This is Prog-07.gz in view mode; [Up]
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer Subject: Re: Getting OpenStep/Mach program to run on Windows 95 Date: 1 Jul 1997 03:41:54 GMT Organization: Digital Fix Development Message-ID: <5p9ua2$lku$1@news.digifix.com> References: <erich-3006970646150001@ppp-207-104-16-89.snrf01.pacbell.net> <5p8fo9$scq@shelob.afs.com> <erich-3006971932520001@ppp-207-104-16-142.snrf01.pacbell.net> In-Reply-To: <erich-3006971932520001@ppp-207-104-16-142.snrf01.pacbell.net> On 06/30/97, Eric Harley wrote: >In article <5p8fo9$scq@shelob.afs.com>, Greg_Anderson@afs.com wrote: > >> Eric Harley writes >> > Okay, I have an OpenStep application that I just wrote and would like to >> > get it on Windows 95 without buying OpenStep for NT. Where and how do I >> > get the shared libraries for doing this? >> >> (1) The runtime is still not officially "free," so what you propose >> to do is illegal on its face. > >What kind of time frame are the runtimes on? > The release of the premier of Rhapsody from what I've been able to get out of various people at Apple. >> (2) If you don't have OS/NT, then you don't have a binary executable >> capable of running under Windows, so you're screwed either way. >> Code compiled with the OS/Mach release will only run under Mach. > >But if I did have OS/NT then I could run the program under NT and 95? > Only if you compile it under OS/NT -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: gvandyk@icon.co.za Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 1 Jul 1997 05:02:00 GMT Organization: E.S. Systems cc (Financial Systems Development) Message-ID: <5pa308$2on$1@hermes.is.co.za> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5p567o$7m4$1@vader.wolfware.ipc.net> In-Reply-To: <5p567o$7m4$1@vader.wolfware.ipc.net> On 06/29/97, Christopher Wolf wrote: >> I personally find the new PB _far_ superior to the old PB, and it's >> excruciating to go back nowadays when I have to do some 3.3 >> development. I know a number of others who feel the same way. The >> new PB makes keeping the open files organized _much_ easier. I couldn't agree with you more, however the editing capabilities has a lot to be desired. I agree it is much better that 3.3's PB but if you're used to Emacs or something similar then it is a real pain. Then again PB is not an editor but a project management tool. I just wish they can extend PB to have the same type of integration to external editors that 3.3's PB had. In particular when you highlight an error it should highlight the line in your editor of choice and not only in PB. It does however open the file in your editor of choice if you double click on the file. -- Regards, Gerrit van Dyk email: gvandyk@icon.co.za (NeXTMail welcome) E.S. Systems cc The OBJECT is the ADVANTAGE
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 1 Jul 1997 06:28:05 GMT Organization: WolfWare Message-ID: <5pa81l$n6q$1@vader.wolfware.ipc.net> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5p567o$7m4$1@vader.wolfware.ipc.net> <5pa308$2on$1@hermes.is.co.za> In-Reply-To: <5pa308$2on$1@hermes.is.co.za> On 06/30/97, gvandyk@icon.co.za wrote: >On 06/29/97, Christopher Wolf wrote: >>> I personally find the new PB _far_ superior to the old PB, and it's >>> excruciating to go back nowadays when I have to do some 3.3 >>> development. I know a number of others who feel the same way. The >>> new PB makes keeping the open files organized _much_ easier. I didn't write the above, just agreed with it. >I couldn't agree with you more, however the editing capabilities has a >lot to be desired. I agree it is much better that 3.3's PB but if >you're used to Emacs or something similar then it is a real pain. I'm actually pretty happy with the editor now although I agree it's nowhere near as powerful/flexible as emacs. The things that I always missed horribly in Edit.app under 3.3 were the ability to select ranges of text using the keyboard and the ability to use Page-Up/Page-Down keys to page through the text. Having to resort to the mouse to do these things always bugged me but the new 4.2 PB editor lets you do both from the kb so I don't have to take my hands away from it. >Then again PB is not an editor but a project management tool. Then again Emacs isn't an editor but rather an operating system :-) >I just >wish they can extend PB to have the same type of integration to >external editors that 3.3's PB had. Yup, more flexibility like this would certainly be nice. -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
Newsgroups: comp.sys.next.programmer Subject: sockets in 3.3 From: apl@kcbbs.gen.nz (Andrew Lindesay) Date: 1 Jul 97 04:52:09 GMT Message-ID: <17497181.17529.22036@kcbbs.gen.nz> Organization: Kappa Crucis Unix BBS, Auckland, New Zealand I'm writing a little program that listens on one socket and connects to the same socket. So both the client & server are in the same executable NeXTSTEP app. This means: (A)SERVER <---connect--- (A)CLIENT or (A)SERVER <---CONNECT--- (B)CLIENT or (A)SERVER <---connect--- (A)CLIENT <---connect--- (B)CLIENT Now the second scenario works; and I haven't tested multiple connections yet. Oh yer (A) and (B) are running processes of the executable. This program uses Mach threads so in my mind, a process _should_ be able to connect a socket to itself! Am I wrong in assuming this? I suppose if I can't get it to work, I'll have to make a seperate Server application which would make it a bit confusing for some users in a single-use situation. Andrew (apl@kcbbs.gen.nz)
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 1 Jul 1997 06:58:42 GMT Organization: 21st Century Software, New York City Sender: daver@ts3-3.nj.cnct.com Message-ID: <5pa9r2$mbb$1@darla.visi.com> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5p567o$7m4$1@vader.wolfware.ipc.net> <5pa308$2on$1@hermes.is.co.za> <5pa81l$n6q$1@vader.wolfware.ipc.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 01 Jul 1997 01:58:42 CDT Cc: cwolf@wolfware.com In <5pa81l$n6q$1@vader.wolfware.ipc.net> Christopher Wolf wrote: > Yup, more flexibility like this would certainly be nice. Just to add my two cents, 4.x is definitely better (and it looks like it was going to be even more powerful; poke around in PB's nibs and you'll see all sorts of interesting panels.. hmmm) but I would really appreciate more integrated documentation support. AutoDoc comments are workable, but having a rtf.template and an indexed documentation editor would really be great. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
From: ;aosdfjsda;lj@;lkajf;lkas.com Organization: Internet MCI Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5p9vak$edr$2474@news.internetmci.com> Message-ID: <cancel.5p9vak$edr$2474@news.internetmci.com> Control: cancel <5p9vak$edr$2474@news.internetmci.com> References: <5p9vak$edr$2474@news.internetmci.com> Date: Tue, 01 Jul 1997 05:12:43 GMT EMP/ECP spam cancelled by hweede@berlin.snafu.de. This is an ongoing spam whose Breidbart index already is above 20. See my report "sexy-girls.com" or "summary of auto-cancellations" in news.admin.net-abuse.bulletins. Subject was: -Hot Teen caught Undressing teenund.jpg.
Newsgroups: comp.sys.next.programmer Subject: Re: Shipping Frameworks Message-ID: <33B7E8DE.7072@running-start.com> From: Ralph Zazula <zazula@running-start.com> Date: Mon, 30 Jun 1997 10:11:58 -0700 References: <5ovjqk$75f$1@owl.slip.net> Organization: Running Start, Inc. MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi - You can include the .dll files for the frameworks inside of the app wrapper. If you have resources that need to be found, you would place them in .app/Resources since the framework now thinks its wrapper is the .app folder. The potential gotcha here is frameworks that have the same name for a resource... Ralph -- Ralph Zazula Running Start, Inc. zazula@running-start.com 520/760-4890 (4891 FAX) http://www.running-start.com Steve Dekorte wrote: > > I have an app that I'd like to distribute but it uses a number > of frameworks which I don't want the end user to have to install. > > Is there a way to statically link selected frameworks or to > stick them in the app wrapper like bundles? >
From: sanjeev@ee.umr.edu (Sanjeev Agarwal) Newsgroups: comp.sys.next.programmer Subject: Help: NXImage.... Date: 1 Jul 1997 10:32:03 GMT Organization: UMR Missouri's Technological University Message-ID: <5pamb3$3b6$1@news.cc.umr.edu> Hi, I have a problem. I am working on a image processing application on NextStep 3.2. I am using NXImage class with NXBitmapImageRep to specify the bitmap for the image. For this I provide the image data as a unsigned char array to the "initData: pixelsWide: pixelsHigh:.... procedure of NXBitmapImageRep. Now I make some changes to the image data (by changing pixel value in the array)!! How do I make it visible on the window. I mean when I say display should the changes not show up on the window??? Please help... If I need to take some other path to composite the NXImage please let me know... Thanx in advance ... Sanjeev
From: embuck@palmer.cca.rockwell.com (Erik M. Buck) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 1 Jul 1997 14:10:53 GMT Organization: Rockwell Collins Message-ID: <5pb35d$6ui2@castor.cca.rockwell.com> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> Cc: andrew_abernathy@omnigroup.com The 4.2 Project Builder is still not all that reliable. It crashes every few hours. At least once a day, the debugger and Project Builder stop communicating and Project Builder must be restarted. The colored syntax stuff occasionally gets it wrong. I spent several minutes looking for a missing close quote because the syntax coloring told me my entire document was quoted. Of course it was not. The coloring was just wrong. I really really miss the 3.3 debugger. The 4.2 debugger is much less powerful. Watch points no longer work. future-break works unreliable. There is no browse mode. When a program dies, there is no post-mortem (sp?). The stack is lost and all symbol table info is lost. (Are the frameworks unloaded ?) Nevertheless, 4.2 Project Builder is still very productive and it crashes less that VC++. I still get good work done. I just miss some features. By the way, has anyone else noticed that NeXT lost their polish somewhere between 3.3 and 4.0 ? Everything used to be so well though out. I feel like I am using an experimental system now.
From: embuck@palmer.cca.rockwell.com (Erik M. Buck) Newsgroups: comp.sys.next.programmer Subject: Re: Round Windows Date: 1 Jul 1997 13:57:40 GMT Organization: Rockwell Collins Message-ID: <5pb2ck$6ui1@castor.cca.rockwell.com> References: <beauvois-3006970655030001@amour.la.utk.edu> Cc: beauvois@usa.net In <beauvois-3006970655030001@amour.la.utk.edu> beauvois@usa.net wrote: > > Following Thomas Poff's idea, I think round windows would be a very > interesting addition. > > But how would one implement such a thing ? > Implement it as a transparent rectangular window and a clipping path. You can have arbitrary shaped windows. My personal favorite window is an eight pointed star.
From: mtrombin@ix.netcom.com (Mark Trombino) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 1 Jul 1997 18:19:36 GMT Organization: Egghead Billy, Inc. Message-ID: <5pbhno$cf4@dfw-ixnews7.ix.netcom.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5p90bq$r1u$1@darla.visi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In <5p90bq$r1u$1@darla.visi.com> David Young wrote: > I generally follow a policy where in all instance setFooBar:(NSObject > *)sender methods I retain (or copy) the incoming object (and release the > previous value). In methods that return a value, I usually do return > [[someValue copy] autorelease], unless there's a specific reason not to. This was another question of mine and I'm glad that you brought it up! I was wondering what guidelines there were (if any) for when to return an object, or when to return a copy of an object. In general, is it always preferred to return a copy? I know it is when working with distributed objects... -- |-------------------------------------------------------------------------| | Mark Trombino | J A M S o f t | | mtrombin@ix.netcom.com | Audio DSP Tools for Openstep & Rhapsody | |-------------------------------------------------------------------------|
From: Dru Nelson <dnelson@slip.net> Newsgroups: comp.sys.next.programmer,comp.unix.programmer,comp.unix.questions Subject: Re: Are these entry-level-job handicaps show-stoppers?" Date: 1 Jul 1997 18:52:01 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5pbjkh$ig1$2@owl.slip.net> References: <332DA1F1.41A4@hom.net> <5h7eed$ntq$1@news.platinum.com> <01bc85a2$01ec70c0$bc65a8c6@metnews.hip.cam.org> I understand your situation (all too well I'm afraid.) Anyhow, I think Mr. Longsine's advice is good. If you follow it you will have much greater chances of getting a job. Also, you might want to mention your location. This can be important sometimes when looking for a future job. Just keep learning, there is enough need out there for people who can do things. The openstep market for programmers is starting to pick up, so in 6 months it will be very healthy. dnelson@slip.net
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 1 Jul 1997 19:13:57 GMT Organization: 21st Century Software, New York City Sender: daver@ts3-3.nj.cnct.com Message-ID: <5pbktl$k5t$1@darla.visi.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5p90bq$r1u$1@darla.visi.com> <5pbhno$cf4@dfw-ixnews7.ix.netcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 01 Jul 1997 14:13:57 CDT Cc: mtrombin@ix.netcom.com In <5pbhno$cf4@dfw-ixnews7.ix.netcom.com> Mark Trombino wrote: > This was another question of mine and I'm glad that you brought it up! I was > wondering what guidelines there were (if any) for when to return an object, > or when to return a copy of an object. In general, is it always preferred to > return a copy? I know it is when working with distributed objects... If your instance variable is mutable then I would return a copy simply to avoid the having the return value changed (even though it might be cast as an immutable object on return, anything could happen...). In the case of immutable objects, they often implement -copy as -retain anyway, so it's not as though you'd incur a huge performance loss from "return [foo copy]" versus "return foo". If you return a copy, and then the calling method forgets to release, you're still a lot better off than if you had returned the original and the calling method released it. The autorelease on return is just a policy I follow and is totally optional. Just be consistent with your policy and you should be fine. Note that this doesn't make sense in all situations. There will be exceptions, like -[NSView subviews] and such. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
From: thomas@zippy.sonoma.edu (Thomas Poff) Newsgroups: comp.sys.next.programmer Subject: Re: Round Windows Date: 1 Jul 1997 20:50:35 GMT Organization: Information Resources and Technology Message-ID: <5pbqir$cfp$1@nuke.csu.net> References: <beauvois-3006970655030001@amour.la.utk.edu> <5pb2ck$6ui1@castor.cca.rockwell.com> Hi, So how do you start making a clear window? Can you still use appkit and use window type "plain_type" and take over it's drawing functions? Or do you have to go _totally_ down to the postscript level? Mainly I'm curious tho' if I get a good hint or too I'll build something some weekend and post it to the peanuts site. Thomas Erik M. Buck (embuck@palmer.cca.rockwell.com) wrote: : In <beauvois-3006970655030001@amour.la.utk.edu> beauvois@usa.net wrote: : > : > Following Thomas Poff's idea, I think round windows would be a very : > interesting addition. : > : > But how would one implement such a thing ? : > : Implement it as a transparent rectangular window and a clipping path. You : can have arbitrary shaped windows. My personal favorite window is an eight : pointed star. -- <>+<> ////// __v__ __\/__ `\|||/ /---\ """"""" | _ - | (_____) . / ^ _ \ . (q p) | o o | <^-@-@-^> (| o O |) .( O O ), |\| (o)(o) |/| _ooO_<_>_Ooo_ooO_U_Ooo_ooO__v__Ooo_ooO_u_Ooo_ooO__(_)__Ooa__oOO_()_OOo___ [_____}_____!____.}_____{_____|_____}_____i____.}_____!_____{_____}_____] __.}____.|_____{_____!____.}_____|_____{.____}_____|_____}_____|_____!___ [_____{_____}_____|_____}_____i_____}_____|_____}_____i_____{_____}_____]
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: Source to ObjectAlloc.app .?.. and how to extend it Date: 1 Jul 1997 23:10:54 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5pc2pu$ni7$1@concorde.ctp.com> Hi, What I was just wondering right now if there wouldn't be a way to make ObjectAlloc smarter and even better at tracking down these alloc problems that are coming up every now and then. ObjectAlloc already traces all "alloc" method...so I am ondering what if the could be coupled with GDB and then if would somehow record which obejct did send the alloc/release/autorelease messages...and then in an then allow you to trace down a badly release object if you have zombies enabled. In really hard cases it could even record some kind of stackframe or calling hierarchy so that could actually see which methods have been envolved in release and retaining that specific instance. I am not able to find the source for ObejctAlloc...and I'm wondering I "Apple" would be willingto release if via NeXTanswers so that we could try to put that in on our own...our even better...if they find the time they could add something similar themself. Or is there a public source for an app which does the same thing..actually it seems to be fairly trivial at first glance. Aloha Tomi
From: noyau*NOSPAM*@*NOSPAM*apple.com (Eric Noyau) Newsgroups: comp.sys.next.programmer Subject: Re: Source to ObjectAlloc.app .?.. and how to extend it Date: 2 Jul 1997 00:47:30 GMT Organization: Apple Computer, Inc. Message-ID: <5pc8f2$10gm$1@news.apple.com> References: <5pc2pu$ni7$1@concorde.ctp.com> In article <5pc2pu$ni7$1@concorde.ctp.com> tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) writes: > Hi, > > [...] > ObjectAlloc already traces all "alloc" method...so I am ondering what if > the could be coupled with GDB and then if would somehow record which > obejct did send the alloc/release/autorelease messages...and then in an > then allow you to trace down a badly release object if you have zombies > enabled. > > In really hard cases it could even record some kind of stackframe or > calling hierarchy so that could actually see which methods have been > envolved in release and retaining that specific instance. > [...] The command line tool 'oh' is doing exactly that. Try 'man oh'. -- Eric
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: <sales@golightspeed.com> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5pcicr$1v9$181@news.pacificrim.net> Control: cancel <5pcicr$1v9$181@news.pacificrim.net> Date: 02 Jul 1997 03:41:32 GMT Message-ID: <cancel.5pcicr$1v9$181@news.pacificrim.net> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: Žiboehme@abm08.abm.de (Ivo Boehme) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 2 Jul 1997 07:50:41 GMT Organization: Nacamar Data Communications Message-ID: <5pd18h$el0$2@news.nacamar.de> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5p567o$7m4$1@vader.wolfware.ipc.net> <5pa308$2on$1@hermes.is.co.za> <5pa81l$n6q$1@vader.wolfware.ipc.net> cwolf@wolfware.com (Christopher Wolf) wrote: > On 06/30/97, gvandyk@icon.co.za wrote: > >On 06/29/97, Christopher Wolf wrote: > > [...] > > I'm actually pretty happy with the editor now although I agree it's nowhere near > as powerful/flexible as emacs. The things that I always missed horribly in > Edit.app under 3.3 were the ability to select ranges of text using the keyboard > and the ability to use Page-Up/Page-Down keys to page through the text. Having ^^^^^^^^^ Page-Down was present in 3.3 (Control-v). Page-Up was possible in 3.2 or maybe 3.1, I think (Alternate-v). Line-Selection is Command-1 in 3.3, which I use frequently (both 3.3 and line-selection ;-)) Just to mention it. > to resort to the mouse to do these things always bugged me but the new 4.2 PB > editor lets you do both from the kb so I don't have to take my hands away from > it. > --- Ivo
From: ians@cam-ani.co.uk (Ian Stephenson) Newsgroups: comp.sys.next.programmer Subject: Re: Round Windows Date: Wed, 2 Jul 1997 10:09:10 GMT Organization: Cambridge Animation Systems Ltd Sender: news@cam-ani.co.uk Message-ID: <ECos7B.DDw@cam-ani.co.uk> References: <5pbqir$cfp$1@nuke.csu.net> In article <5pbqir$cfp$1@nuke.csu.net> thomas@zippy.sonoma.edu (Thomas Poff) writes: > Hi, > > So how do you start making a clear window? Can you still use appkit and > use window type "plain_type" and take over it's drawing functions? NeXTAnswers says you can't do it. (it goes on to say it could be added, but they've just got far better things to do). The recommended workaround is to grab the background off the screen, and draw it inside a square window, then draw your contents over the top. $an
Subject: Re: CVS and .nib files Newsgroups: comp.sys.next.programmer,comp.sys.next.software References: <5p0g4q$dlh@fido.transit.ru> <5p0sdu$nea@iserver.stem.com> In-Reply-To: <5p0sdu$nea@iserver.stem.com> From: vincent@sente.ch (Vincent Kohler) Message-ID: <33ba574c.0@epflnews.epfl.ch> Date: 2 Jul 97 13:27:40 GMT On 06/27/97, dbriggs@stem.com wrote: > >Hello, Vladimir -- & others who use CVS on NEXTSTEP and OPENSTEP > >rtr@user.transit.ru (transit shell) wrote: >>Hi, >> >>Could anybody suggest me how to wrap/unwrap .nib files to store them into cvs >>repository? >> >>Thank you in advance. >> >>Vladimir > [..] >A SOLUTION: >Bob Vadnais wrote a Palette for InterfaceBuilder, called CVSPalette. [..] Another solution is to configure correctly the cvswrappers file within $CVSROOT/ in order to have a real wrapped nib. This option works well, too. Unfortunately it won't work in client/server because cvswrappers are not supported under this mode. I had some bad experience with CVSPalette which didn't copied the CVS subdir in a reliable way, at least with my configuration. Cvswrappers are very useful for diagram, quantrix and others wrapped files. Here is an extract of my cvswrappers: *.nib -f '$CVSROOT/CVSROOT/unwrap %s' -t '$CVSROOT/CVSROOT/wrap %s \ %s' -m 'COPY' *.rtfd -f '$CVSROOT/CVSROOT/unwrap %s' -t '$CVSROOT/CVSROOT/wrap %s \ %s' *.draw -f '$CVSROOT/CVSROOT/unwrap %s' -t '$CVSROOT/CVSROOT/wrap %s \ %s' *.quantrix -f '$CVSROOT/CVSROOT/unwrap %s' -t '$CVSROOT/CVSROOT/wrap \ %s %s' -m 'COPY' Regards Vincent -- ---------------------------------------- Vincent Kohler vincent@sente.ch (NextMail welcome) Sen:te Parc Scientifique - EPFL - Switzerland
Subject: Re: (PB) EOF NLS_LANG Newsgroups: comp.sys.next.programmer References: <EC89pM.8BA@x-lan.alienor.fr> In-Reply-To: <EC89pM.8BA@x-lan.alienor.fr> From: vincent@hnw.sente.ch (Vincent Kohler) Message-ID: <33ba57f0.0@epflnews.epfl.ch> Date: 2 Jul 97 13:30:24 GMT On 06/23/97, Moukdarath Valerie wrote: >I work on NeXTstep 3.2 and EOF 1.1 and Oracle 7.1 > >My oracle NLS_LANG is american_america.we8dec > >I don't know what NLS_LANG to use on EOF to read correctely my date > >ex : "Boite de crème à été" in my database is not the same when I read it with EOF. > > >é may be replace by "Y" or "î" it depends on what caracter set i use. > >valérie > We used the following option in a project to keep a consistent character set NLS_LANG = AMERICAN_AMERICA.WE8ISO8859P1 BTW, your e-mail was wrong is your post Regards Vincent -- ---------------------------------------- Vincent Kohler vincent@sente.ch (NextMail welcome) Sen:te Parc Scientifique - EPFL - Switzerland
From: sg13@rcs.urz.tu-dresden.de (Steffen Greiffenberg) Newsgroups: comp.sys.next.programmer Subject: Accessing a Selected EObjectValue with Openstep Date: 2 Jul 1997 14:16:18 GMT Organization: TU Dresden (URZ) Message-ID: <5pdnri$2n2$1@rks1.urz.tu-dresden.de> hi, i'm trying to display a stringvalue from a table with openstep. With NeXT i used the function objectForKey. After trying: NSRunAlertPanel(@"FULLNAME", [[theTable selectedObject] valueForKey:@"FULLNAME"], 0, 0, 0); i recieve the following message: *** Assertion failure in -[NSTextFieldCell _objectValue:forString:], NSCell.m:864 Invalid parameter not satisfying: aString != nil the key "FULLNAME" is an attribute. The id "theTable" is connected with the EOModel where FULLNAME is defined as an VARCHAR2(100) an connected with NSString. What goes wrong here? Can anybody help? TIA Steffen Please replay directly to sg13@rcs.urz.tu-dresden.de
Subject: Looking for an expert system shell Newsgroups: comp.sys.next.programmer From: marco@sente.ch (Marco Scheurer) Organization: Sen:te Message-ID: <33ba6a8e.0@epflnews.epfl.ch> Date: 2 Jul 97 14:49:50 GMT Hi, I'm looking for an expert system shell I could easily use with NEXTSTEP 3.3. More specifically, I am looking for a forward chaining inference engine. Any reference to available systems, source code, commercial or not would be greatly appreciated. Thanks, -- Marco Scheurer (marco@sente.ch) Sen:te
From: ftouhi@IRO.UMontreal.CA (Majid Ftouhi) Newsgroups: comp.sys.next.programmer Subject: file bool.h Date: 2 Jul 1997 16:56:58 GMT Organization: Universite de Montreal Distribution: world Message-ID: <5pe18q$dih@epervier.CC.UMontreal.CA> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii hi there: if you have the file: /usr/local/lib/g++-include/bool.h send it to me please Thanks in advance
From: robdoss@nvc.cc.ca.us (Robert C. Doss) Newsgroups: comp.sys.next.programmer,comp.sys.next.misc Subject: Converitng Mach-O to S-Records Date: 2 Jul 1997 18:18:25 GMT Organization: Napa Valley College Message-ID: <5pe61h$1hq@wiley.napanet.net> I'm looking for a utility which would convert the Mach-O object files produced by /bin/as into Motorola S-Record format. I intend to program my embedded MC680xx processor on my NeXT using Motorola /bin/as but I need to have the object file be an S-Record so I can burn the ROMs on my embedded system. Let me make this clear: I do not intend to run the program under NEXTSTEP, instead I want to run it on an embedded system with software developed ubder NEXTSTEP. -- *---------------------------------------------------------------------* |Robert C. Doss Jr. |Internet:RobDoss@nvc.cc.ca.us|NeXT mail & MIME ok| |Napa Valley College|Fax: (707) 253-3063 |MIME mail preferred| *---------------------------------------------------------------------* k that .spa-am and nospam to reply.) Well, if there *were* anything to say, it would be with the understanding that the PR/Marketing people want to make the announcements on products, so anything I have to say wouldn't actually exist until after then, so what I might have to say now doesn't exist, and what I may say in future can't be said, so theoretically what exists, doesn't, for the immediate future. (With apologies to Joe Straczynski)
From: HDDL9@VSEZD.V7G0X Newsgroups: comp.sys.next.programmer Subject: Get rich on the net Date: 21 JUN 1997 22:51:03 +0200 Organization: Unisource Espana NEWS SERVER Message-ID: <5ohevu$1h2$39@talia.mad.ibernet.es> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear Friend, Greetings! I am a retired Attorney, and about two years ago a man came to me with a letter. The letter he brought me is basically the letter in your hands. He asked me to verify the fact that this letter was legal. I told him that I would review it and get back to him. When I first read the letter my client brought me I thought it was some off-the-wall idea to make money. However, a week and-a-half later we met in my office to discuss the issue. I told him the letter he originally brouht to me was not 100% legal. I suggested a few changes with which the letter would be 100% legal. I was curious about this letter, and my client told me how this plan works and suggested that I give it a try. I thought it was a long shot and would be a wast of time, so I decided against participating. However, before my client left, I asked him to keep me updated of his results. About two months later he called me to tell me he had received over $800,000 in cash!! I didn't believe him, so he suggested that I try this idea and find out for myself. I thought about it for a couple of days and decided that I didn't really have anything to lose, so I asked my client for a copy of the letter. I followed the instructions exactly, sent out 200 copies, and sure enough, the money started arriving! It came slowly at first, but after 3 weeks, I was getting more mail than I could open in a day! The money stopped coming after about three months. I kept a precise record of my earnings, and they totaled $968,498.00!! I was earning a good living as a lawyer, but anyone in the legal profession, or even anyone who owns their own business will tell you, there is a lot of stress that comes with the job. I told myself that if things worked out, I would retire from my job and travel. I decided to try the letter again, but this time I sent out 500 letters. Well , three months later, my earnings totaled $2,344,178.00!!! I just could not believe it. I met with my former client (and now my friend) for lunch to find out how it worked. He told me there are quite a few similar letters going around. What made THIS ONE DIFFERENT WAS THE FACT THAT 6 ADDRESSES APPEAR ON THE LIST, AND NOT 5 LIKE MOST OF THE OTHERS. This fact alone results in your name being on far more returns. Of course I was thankful that he had shown this plan to me, but he was just as thankful that I had suggested the changes to ensure that this plan was legal since most people don't want to risk doing something illegal. I'll bet that by now you are curious to know the little change I suggested in my friend's original letter. Well, if you send a letter like this out, to be legal you must actually sell a product or service if you expect to receive a dollar. I told my friend that anyone sending a dollar back to him must receive something in return. You have received this letter due to the participation of each of the 6 persons given below. This letter and the plan given within this letter are a "product" that you have received. Thus, you will be sending $1.00 to each of the 6 persons given below in exchange for receiving this letter and for asking them to provide the "service" of adding you on their mailing list. THIS IS A SERVICE AND IS 100% LEGAL BUSINESS OPPORTUNITY WHICH IS COVERED IN TITLE 18, SECTION 1302 AND 1342 OF THE POSTAL AND LOTTERY LAWS. NOW, PLEASE FOLLOW THESE TRIED AND TRUE STEPS TO OBTAIN THE AMOUNT YOU DESIRE: 1) Immediately send a $1 dollar bill to each of the 6 people on the list below. Wrap this dollar in a note with the words: "Please put me on your mailing list." (Remember, in this way, you're not just sending a dollar to someone; you're paying for a legitimate service.) Include your address. You do not need to include your name. This is the key to the program! Also, let each person know what number they are in the letter so he/she knows which cycle he is receiving the $1.00 from. Make sure that you retain EVERY name and or address sent to you, either on computer or hard copy, but do not discard the names and notes that people send to you. This is PROOF that you are truly providing a service, and should the I.R.S. or some other government agency question you, you can provide them with this proof! Therefore you will start a list of all the people who send you a dollar requesting to be added on a mailing list. Actually, you will want to safeguard this list because it can generate even better responses and much more money later! See step #4 below. 2) Remove the name next to the #1 on the list, and move the rest of the names up one position. Then place your name in the #6 spot. Thus, #6 becomes #5, #4 becomes #3, etc., and YOU will be #6 on the list. DO NOT EDIT THIS LETTER ANYMORE BEYOND THIS POINT. When you are mailing this letter, you may edit the references to the internet. 3) Post this letter in at least 250 newsgroups, I do 5 groups per post. You can simply copy the contents of this letter and past it to the newsgroups of your choice. IMPORTANT NOTE: Please pay attention to the newsgroups in which you post. If there is an abundance of similar letters, you may want to seek another newsgroup. Thus, pat respect to the harmony of the newsgroup and use common sense. There are plenty of newsgroups on the Internet. (I think there are close to 18,000 of them.) ALSO, PLEASE POST IN APPROPRIATE GROUPS AND SPREAD YOUR POSTINGS OVER MANY DIFFERENT LOCAL & INTERNATIONAL AREAS (SPREAD THE WORD)! For added success, you can also mail this letter as well, in addition to posting in newsgroups. Order 200 (or more) names from a mailing list company. Two that have been most effective for these names are: S.E. Mailing List Advon Distributors P.O. Box 15061 P.O. Box B-11 Ft. Lauderdale, FL 33318S Helly, ID 83274 (954) 742-9519 (800) 992-3866 4. Keep a copy of this letter so you can use it a second time. Also keep EACH AND EVERY NAME AND ADDRESS SENT TO YOU as instructed in #1. Then, send out more copies in about 6 months, but mail it to the addresses you receive (your new mailing list) with each dollar! It will work again, only much better! Here are the addresses that you must send a dollar to with your slip of paper requesting placement on their mailing list, as instructed above: 1. Ron van Hoof 14 Moshier St. Greenwich, CT 06831 USA 2. Richard Blanchette Pob 455 Palisade, Colorado 81520 USA 3. Caroline Mowinckel Molnar Utca 21 2 Em Nr-10 H-1056 Budapest Hungary 4. Odd Hilsen Avda Espana 93 Sitio de Calahonda Mijas Costa - Malaga Spain 5. Peter Kjoge Camino de la Condesa 9, 5-A E-29640 Fuengirola Spain ***THIS PLAN WILL NOT WORK UNLESS YOU SEND $1.00 TO EACH OF THE ABOVE PERSONS AS DIRECTED. How do the numbers work for potential income earnings? Assume for the sake of example that you get a 7.5% return rate. This is a very conservative, my first attempt was about 9.5%, and my second was over 11%. 1. You send out 250 letters, and 18 people (250 x 7.5% = 18) will send $1 each. $18.00 2. Those 18 send out 250 letters each, and 337 people (18 x 250 = 4,500x 7.5% = 337), send you $1 each. $337.00 3. Those 337 send out 250 letters each (337 x 250 = 84,250 x 7.5% = 6,318), and 6,318 people send you $1 each. $6,318 4. Those 6,318 send out 250 letters each (6,318 x 250 = 1,579,500 x 7.5% = 118,462), and 118,462 people send you $1 each. $118,462.00 5. Those 118,462 send out 250 letters each (118,462 x 250 = 29,615,500 x 7.5% = 2,221,162), and 2,221,162 people send you $1 each. $2,221,162.00 ADD 1-5 AND TOTAL AT $2,346,297.50 ALL THIS IS AT A 7.5% RETURN ON YOUR LETTERS. THAT'S 18 REPLIES ON YOUR ORIGINAL 250 POSTINGS. Some valuable, general suggestions: 1. DO NOT TRY AND CHEAT THIS PROCESS. IF YOU DO, YOU WILL NOT ACHIEVE THE DISIRED RESULTS. 2. The Internet is a very effective medium for exposing this vehicle. However, pay attention to newsgroups, and avoid posting this letter in the newsgroups where there are a lot of upset epeople complaining about "spamming." In other words, we are asking that you use common sense, and have courtesy to others on the Internet. This may be opposite of what many other similar postings may be saying. Keep in mind that there are plenty of people out there who believe and want this as much as you do. 3. Keep a positive frame of mind while engaging in this process, and you will receive the amount of money you desire. 4. In order to achieve maximum results DO NOT CHANGE THE CONTENTS OF THIS LETTER. However, you may edit applicable portions relating to Internet when sending this letter via regular mail. 5. When the money begins to arrive, it is a good gesture to give 10% or more to your church or favorite charity with a joyful spirit. The Internet has proved to be a very quick and effective means to reach millions of people with your message. This message exemplifies a UNIVERSAL LAW OF ABUNDANCE that there is plenty to go around for everyone. You will not be hurting anyone and will be creating the opportunity for those who read your message to benefit--just as you will--from the abundance that exists in this universe.
From: Žoloft@hegel1.cs.chalmers.se.cs.chalmers.se (Olof Torgersson) Newsgroups: comp.sys.next.programmer Subject: libfl on NT? Date: 2 Jul 1997 18:36:38 GMT Organization: Chalmers University of Technology Message-ID: <5pe73m$eh8$1@nyheter.chalmers.se> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi, I'm trying to convert a 3.3 project using flex and bison generated files to OPENSTEP. I managed to do it for mach, but I would also like to port it to NT. While flex and bison seems to come with NextDeveloper I can find no libfl needed to link the resulting programs. Does such a library exist for NT and if so, where can I find it? Olof Torgersson --- Olof Torgersson oloft@cs.chalmers.se Department of Computing Science +46 31 772 54 06 Göteborg University & Chalmers University of Technology S-412 96 GÖTEBORG, SWEDEN http://www.cs.chalmers.se/~oloft/
Newsgroups: comp.sys.next.programmer From: allan@ali.bc.ca (Allan Noordvyk) Subject: Re: Round Windows Message-ID: <ECpGBz.uC@gateway.ali.bc.ca> Sender: nobody@gateway.ali.bc.ca Mime-Version: 1.0 Cc: mpaque@wco.com Content-Type: text/plain; charset=us-ascii Organization: ALI Technologies Date: Wed, 2 Jul 1997 18:50:23 GMT References: <5pbqir$cfp$1@nuke.csu.net> <5pe56t$1tp@mpaque.mpaque> In comp.sys.next.programmer Mike Paquette wrote: > In article <5pbqir$cfp$1@nuke.csu.net> thomas@zippy.sonoma.edu (Thomas > Poff) writes: > > So how do you start making a clear window? Can you still use appkit and > > use window type "plain_type" and take over it's drawing functions? > > /* Checked for 4.2 OPENSTEP for Mach */ > NSWindow * funkyWindow; > funkyWindow = [[NSWindow alloc] initWithContentRect:frame > styleMask:NSBorderlessWindowMask > backing:NSBackingStoreNonretained > defer:NO]; > PSsetautofill(NO, [funkyWindow windowNumber]); > [funkyWindow makeKeyAndOrderFront:self]; > > /* At this point, funkyWindow is on-screen and filled with the > * contents of the windows beneath it. You can play with drawing > * into it's contentView. > */ > Unfortunately, this only works until the user the window is moved. Snarfing the background as the window is on the hoof is pretty hard to accomplish. Of course, if the window can't be moved, then this isn't a problem. -- Allan Noordvyk, Software Artisan e-mail: allan@ali.bc.ca ALI Technologies Voice: 604.821.6317 Richmond, Canada Fax: 604.279.5468
From: SLKUJF@DSOIUFD.COM Newsgroups: comp.sys.next.programmer Subject: 800 SERVICE AT 7.9 CENTS PER MINUTE! Date: 2 Jul 1997 21:30:06 GMT Organization: All USENET -- http://www.Supernews.com Message-ID: <5peh8u$8rs$1@usenet85.supernews.com> Inbound dedicated toll-free service Now only 7.9 cents/minute * Use existing 800/888#s only rate will change! * Billing done in six second increments! * No set per month charge! no minimums * Does not effect your home or business outbound long distance service. A one time association fee is required at time of activation of $49.95 is due upon sign up Free first 30 minutes of usage if ordered by july 5,1997 Billing Information: last______________________first_______________________mi,_____ address____________________________city_______________________ state___________zip code____________________ phone:__________________fax:_______________________ Where do you wish to have this 800/888# ring?________________________ Do you have a 800/888# you wish to receive this rate on?_________________ We take: Mastercard,Visa,Amex,Discover,Money Orders or Checks Card Number_______________________________exp____/_____ signature___________________________date_______/____/____ fax completed application to: 213-650-9031 Or call for more information: 213-650-1027 Virtual Office Communications Company. 7985 santa monica bl. 109-416 West Hollywood CA. 90046-5112
Newsgroups: comp.sys.next.programmer From: yuwaraj@ecf.toronto.edu (Murugathas Yuwaraj) Subject: Q:Java development in NS3.0? Sender: news@ecf.toronto.edu (News Administrator) Message-ID: <ECp4wH.EE@ecf.toronto.edu> Date: Wed, 2 Jul 1997 14:43:29 GMT Organization: University of Toronto, Engineering Computing Facility Hello everyone, Could someone tell me if JDK is available for NS3.0? Thank you kindly. Cheers, Thas
Newsgroups: comp.sys.next.programmer,comp.unix.programmer,comp.unix.questions Subject: Re: Are these entry-level-job handicaps show-stoppers?" Message-ID: <33B98E02.5859@running-start.com> From: Ralph Zazula <zazula@running-start.com> Date: Tue, 01 Jul 1997 16:08:50 -0700 References: <332DA1F1.41A4@hom.net> <5h7eed$ntq$1@news.platinum.com> <01bc85a2$01ec70c0$bc65a8c6@metnews.hip.cam.org> Organization: Running Start, Inc. MIME-Version: 1.0 To: Mark <mp@digiplace.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi - Mark wrote: > > A few months back Gary Longsine posted a thoughtful, detailed response > (quoted at the bottom of this post) in reply to somebody wondering about > getting an entry-level programming job. > > I'm wondering if he or anyone else would like to comment on my situation - > it's similar but with a difference: I don't have the CS degree. > > I lack no confidence that with some diligent work I can become competent in > Unix programming, Objective-C, the OpenStep API, and fundamental OOP > principles. > [...] Here are some thoughts - there are many different roles to be filled in a large software development organization. You may find that you can find a path to becoming a software engineer by starting with a position in: - support - testing - release control - training - system installation/setup (e.g., grunt) - pre-sales support If there is a particular company you want to work for, you may want to search out these types of positions. The real key is to leverage your insider position to learn as much as you can about the platform and make sure you take time to learn new skills. Chances are, companies will encourage you to move to higher-skilled positions. It is up to you to make them notice you. Once you have established yourself as a compentent contributor, things like the lacking degree should become less of a barrier. Ralph -- Ralph Zazula Running Start, Inc. zazula@running-start.com 520/760-4890 (4891 FAX) http://www.running-start.com
From: szallies@gehirnbrei.energotec.de (Constantin Szallies) Newsgroups: comp.sys.next.programmer,comp.sys.next.software Subject: Re: CVS and .nib files Date: 1 Jul 1997 15:04:08 GMT Organization: Technet GmbH Message-ID: <5pb698$osk$1@oxygen.technet.net> References: <5p0g4q$dlh@fido.transit.ru> <5p0sdu$nea@iserver.stem.com> dbriggs@stem.com wrote: > >Hello, Vladimir -- & others who use CVS on NEXTSTEP and OPENSTEP > >rtr@user.transit.ru (transit shell) wrote: >>Hi, >> >>Could anybody suggest me how to wrap/unwrap .nib files to store them into cvs >>repository? >> >>Thank you in advance. >> >>Vladimir > >THE PROBLEM: >GNU's CVS (Concurrent Version System) is useful for archiving source >code and documentation. It has trouble with InterfaceBuilder, though. >InterfaceBuilder destroys the CVS information inside the *.nib files >(directories) when it modifies a *.nib. (Actually, it fails to >copy the CVS stuff from the old version of the *.nib into the new.) >Consequently, one can archive a *.lproj directory with CVS, but if >one checks out a version, then modifies its *.nib files, and then >tries to archive the new version, -- frustration! > SOLUTION 2: To solve the nib file problem, checkout the file cwswrappers in the repository CVSROOT and add the following lines: *.nib -f '$CVSROOT/CVSROOT/unwrap %s' -t '$CVSROOT/CVSROOT/wrap %s %s' *.rtfd -f '$CVSROOT/CVSROOT/unwrap %s' -t '$CVSROOT/CVSROOT/wrap %s %s' *.draw -f '$CVSROOT/CVSROOT/unwrap %s' -t '$CVSROOT/CVSROOT/wrap %s %s' *.tiff -m 'COPY' *.eps -m 'COPY' *.snd -m 'COPY' Then commit. The binaries wrap and unwrap come with the cvs distribution and go into the CVSROOT directory. Works well. We are using local cvs + NFS imported repository. Greetings -- # Constantin Szallies, Energotec GmbH # szallies@energotec.de # 49211-9144018
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: SLKUJF@DSOIUFD.COM Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5peh8u$8rs$1@usenet85.supernews.com> Control: cancel <5peh8u$8rs$1@usenet85.supernews.com> Date: 03 Jul 1997 00:04:03 GMT Message-ID: <cancel.5peh8u$8rs$1@usenet85.supernews.com> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: leigh@cs.uwa.edu.au (Leigh Smith) Newsgroups: comp.sys.next.programmer Subject: PB V4.2 continue-after-error preference missing? Date: 3 Jul 1997 03:08:52 GMT Organization: The University of Western Australia Message-ID: <5pf544$l5h$1@enyo.uwa.edu.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Apologies if this is a FAQ: I'm on the OS4.2 PR2 (Prelude...) and can't find the ProjectBuilder Continue After Error preference that is described in the OpenStep conversion release notes as in the Preferences panel under Build. Scoping my question up a level: How can I get PB to give me all the errors at once, rather one at a time? Many thanks -- Leigh Computer Science, University of Western Australia Smith +61-9-9380-3778 leigh@cs.uwa.edu.au (NeXTMail/MIME) "Home pages are the pet rock of the 90s. We all have them, we all think they're very cute. But in a few years we're going to look back and be pretty embarrassed." -- Tony Shepps <toad@pond.com> "Why wait?" -- Peter Langston
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 3 Jul 1997 03:33:38 GMT Organization: WolfWare Message-ID: <5pf6ii$t1j$1@vader.wolfware.ipc.net> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5p567o$7m4$1@vader.wolfware.ipc.net> <5pa308$2on$1@hermes.is.co.za> <5pa81l$n6q$1@vader.wolfware.ipc.net> <5pd18h$el0$2@news.nacamar.de> In-Reply-To: <5pd18h$el0$2@news.nacamar.de> On 07/01/97, Ivo Boehme wrote: >Page-Down was present in 3.3 (Control-v). Page-Up was possible in 3.2 or >maybe 3.1, I think (Alternate-v). Line-Selection is Command-1 in 3.3, which I >use frequently (both 3.3 and line-selection ;-)) Just to mention it. Well I guess you learn something new everyday. Thanks. Will have to try it since I still use 3.3 every day at my day job. - Chris -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: arti@address.in.signature (Art Isbell) Newsgroups: comp.sys.next.programmer Subject: Re: libfl on NT? Date: 3 Jul 1997 04:56:28 GMT Organization: LavaNet, Inc. Distribution: world Message-ID: <5pfbds$lv5@mochi.lava.net> References: <5pe73m$eh8$1@nyheter.chalmers.se> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Žoloft@hegel1.cs.chalmers.se.cs.chalmers.se (Olof Torgersson) wrote: > > I'm trying to convert a 3.3 project using flex and bison generated files to > OPENSTEP. I managed to do it for mach, but I would also like to port it > to NT. flex and bison are included with OS/NT, so no additional libraries are needed. However, we had been using flex 2.5.2 under NS 3.3 because it's so much nicer than the old BSD lex included with NS. To our dismay, when we ported to NT, our flex source would no longer compile under the flex included with OS/NT. The reason is that the flex version is VERY OLD and we were taking advantage of newer flex features. It's so old that it doesn't respond to the --version flag. It still uses flex.skel from 1990! So you may need to rewrite your flex source somewhat. Not sure about bison because we had been using NS 3.3's yacc. -- Art Isbell NeXT/MIME Mail: arti at lava dot net Trego Systems (for whom I don't speak) Voice/Fax: +1 808 394 0511 OPENSTEP/NT Voice Mail: +1 808 394 0495 managed care solutions US Mail: Honolulu, HI 96825-2638
From: gvandyk@icon.co.za Newsgroups: comp.sys.next.programmer Subject: Re: Round Windows Date: 3 Jul 1997 05:33:46 GMT Organization: E.S. Systems cc (Financial Systems Development) Message-ID: <5pfdjq$3m9$1@hermes.is.co.za> References: <5pbqir$cfp$1@nuke.csu.net> <5pe56t$1tp@mpaque.mpaque> <ECpGBz.uC@gateway.ali.bc.ca> In-Reply-To: <ECpGBz.uC@gateway.ali.bc.ca> In comp.sys.next.programmer Mike Paquette wrote: > In article <5pbqir$cfp$1@nuke.csu.net> thomas@zippy.sonoma.edu (Thomas > Poff) writes: > > So how do you start making a clear window? Can you still use appkit and > > use window type "plain_type" and take over it's drawing functions? > > /* Checked for 4.2 OPENSTEP for Mach */ > NSWindow * funkyWindow; > funkyWindow = [[NSWindow alloc] initWithContentRect:frame > styleMask:NSBorderlessWindowMask > backing:NSBackingStoreNonretained > defer:NO]; > PSsetautofill(NO, [funkyWindow windowNumber]); > [funkyWindow makeKeyAndOrderFront:self]; > > /* At this point, funkyWindow is on-screen and filled with the > * contents of the windows beneath it. You can play with drawing > * into it's contentView. > */ > Is PSsetautofill() supported under NT. The reason I am asking is that I have a window that also needs to be "transparent". I am using PSsetautofill() and everything works very well, except that I get a warning everytime I compile stating that PSsetautofill is not defined. I then looked in the documentation and couldn't find anything either. I am running OS4.1 for mach -- Regards, Gerrit van Dyk email: gvandyk@icon.co.za (NeXTMail welcome) E.S. Systems cc The OBJECT is the ADVANTAGE
From: dcoyle@ctp.com (David A. Coyle) Newsgroups: comp.sys.next.programmer Subject: Re: CVS and .nib files Date: 3 Jul 1997 08:35:10 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5pfo7u$1s3$1@concorde.ctp.com> References: <5pfda8$3hg$1@hermes.is.co.za> In article <5pfda8$3hg$1@hermes.is.co.za> gvandyk@icon.co.za writes: > On 07/02/97, Vincent Kohler wrote: > > I am doing the same and I agree it works very well. I also had a lot > of problems with CVSPalette, sometimes it wrote the CVS files and > sometimes it didn't, totally iconsistent, so I started using wrappers. Well, the palette must actually be loaded for it to work. It seems IB won't actually load a palette until requested (ie: selected). If you want to do this (and wrappers *really *are *better (most of the time(*))) in your CVSIBFixes palette, you could put in something like this, which makes your palette the one selected at startup, and therefor guaranteed to be loaded: // NSApp notifications - (void)applicationWillTerminate:(NSNotification*)aNotification { /* Tasks: 1) read the defaults and see where I sit in the array of loaded palettes 2) set the active palette to be that index. the two keys are: ActivePalette and Palettes my name is CVSIBFixes (I could also use [self class]...) */ NSUserDefaults* theDefaults = [NSUserDefaults standardUserDefaults]; NSArray* palettesArray = [theDefaults arrayForKey:@"Palettes"]; NSEnumerator* myEnum = [palettesArray reverseObjectEnumerator]; // I'm guessing we're likely to be near the end! id anObject; while (anObject = [myEnum nextObject]) { // see if we are contained in that string //NSLog(@"Enumerating...current object is: %@", [anObject description]); if ([anObject rangeOfString:@"CVSIBFixes"].length) { // if not there, range will be 0 break; } } [theDefaults setInteger:[palettesArray indexOfObject:anObject] forKey:@"ActivePalette"]; [theDefaults synchronize]; return; } Indeed, the whole palette can be done in three notifications (IB's - (void)docWillSave:(NSNotification *)aNotification - (void)docDidSave:(NSNotification *)aNotification and the aforementioned NSApp's: - (void)applicationWillTerminate:(NSNotification*)aNotification and an NSFileManager. The only issue involved is making it run on NT as well...and even that's not too hard (*) footnote if you're paying attention: if you're using the SCM module in PB, and you try to check in several nibs at once, some can get screwed up. We don't know why, but probably some notification/tomeout deal. Anyway, if you're using SCM, I find it behhoves one to check in each modified nib separately. ------- /\/\ Dave Coyle <dcoyle@ctp.com> / /_ \ Cambridge Technology Partners \ / / 118/119 Lower Baggot Street \/\/ Dublin 2, Republic of Ireland Tel: +353 1 607 9008 WWW: http://www.ctp.com>
From: dcoyle@ctp.com (David A. Coyle) Newsgroups: comp.sys.next.programmer Subject: Re: Round Windows Date: 3 Jul 1997 08:39:37 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5pfog9$1vt$1@concorde.ctp.com> References: <ECpGBz.uC@gateway.ali.bc.ca> In article <ECpGBz.uC@gateway.ali.bc.ca> allan@ali.bc.ca (Allan Noordvyk) writes: > In comp.sys.next.programmer Mike Paquette wrote: <code removed> > > > > Unfortunately, this only works until the user the window is moved. > Exactly: look at the click-for-help in (e.g.) PB. When you get the big yellow help text, move the window out from under it: watch the shadow on the help box carefully... ;-) (thanks to Georg who pointed this out) May the Force be with you! Dave ------- /\/\ Dave Coyle <dcoyle@ctp.com> / /_ \ Cambridge Technology Partners \ / / 118/119 Lower Baggot Street \/\/ Dublin 2, Republic of Ireland Tel: +353 1 607 9008 WWW: http://www.ctp.com>
From: jwdb@fygir.nl (Jan-Willem de Bruijn) Newsgroups: comp.sys.next.programmer Subject: [Q] Can a window without a title bar be dragged Date: 3 Jul 1997 10:47:15 GMT Organization: XS4ALL, networking for the masses Sender: fygir@194.229.196.70 Message-ID: <5pfvvj$3tg$1@news2.xs4all.nl> In NEXTSTEP 3.3 the Window class had a method dragFrom::eventNum: allowing windows to be dragged by clicking the mouse somewhere other than the title bar. After running the OpenStep conversion scripts the code has somehow changed to dragFromPoint:eventNumber:, but NSWindow does not recognize this. Does anybody know of a way around this? Thanks in advance, -- Jan-Willem de Bruijn - F Y G I R logistic information systems
From: embuck@palmer.cca.rockwell.com (Erik M. Buck) Newsgroups: comp.sys.next.programmer Subject: Re: Round Windows Date: 3 Jul 1997 14:37:16 GMT Organization: Rockwell Collins Message-ID: <5pgdes$kht1@castor.cca.rockwell.com> References: <5pbqir$cfp$1@nuke.csu.net> <5pe56t$1tp@mpaque.mpaque> <ECpGBz.uC@gateway.ali.bc.ca> Cc: allan@ali.bc.ca To make a transparent window do the following - { NSImage *image; NSBitmapImageRep *bitmap; NSRect grabRect; [myWindow setFrame:aRect display:NO]; /* Grab the screen bits into the window. */ [myWindow orderFrontRegardless]; PSsetautofill(NO, [myWindow windowNumber]); [myWindow orderOut:nil]; [myWindow orderFrontRegardless]; /* Read the bits from the window */ [[myWindow contentView] lockFocus]; grabRect = [[myWindow contentView] bounds]; bitmap = [[NSBitmapImageRep alloc] initWithFocusedViewRect:grabRect]; [[myWindow contentView] unlockFocus]; image = [[NSImage alloc] initWithSize:grabRect.size]; [image addRepresentation:bitmap]; [[myWindow contentView] setImage:image]; PSsetautofill(YES, [myWindow windowNumber]); [myWindow display]; } myWindow can contain any sub-Views. If a sub View is not opaque, the background will show through. The Views can have any clipping path including spline curves, text etc. Use - windowWillResize:toSize: -windowWillMove: etc to recapture the background when the window changes.
From: szallies@gehirnbrei.energotec.de (Constantin Szallies) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 1 Jul 1997 13:52:15 GMT Organization: Technet GmbH Message-ID: <5pb22f$na6$1@oxygen.technet.net> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> mtrombin@ix.netcom.com (Mark Trombino) wrote: >I believe that I finally have a handle on how/when to properly retain objects >in OpenStep: Some people don't consider reference counting as garbage collection (GC) because retain cycles are not collected. And because you have to manually send "retain" and "release", OpenStep's GC is not so comfortable and secure as in Java. But well .. you get 80 per cent of the advantages for only 20 % of the costs! >Using the combination of +(id)alloc and -(id)init increments the object ref >count to 1. So does creating an object by copying another. Using special >creater methods such as +(NSArray *)array or +(NSDictionary *)dictionary do >not, so you need to retain them explicitly if you need them to hang around. Methods like +(NSArray *)array are really alloc/init plus autorelease. This means that the returned object has a reference count of one and is included in some NSAutoReleasePool once. Then the pool is destroyed, the object receives "release" once. This makes "autorelease" a deferred "release" -- at least if the pool is eventually destroyed. >Is this correct? If so, then WHY is it this way? It seems to me that it >would make more sense if you retained every object you wanted to keep. I'm >sure, however, that there is some reason why it isn't that way. Can someone >fill me in? If you want to keep an array 1) use alloc/init or 2) +(NSArray *)array plus retain. This increases the reference count to two and after the pool is destroyed, the count drops to one. The autorelease pool concept is really a big enhancement for reference counting! Greetings -- # Constantin Szallies, Energotec GmbH # szallies@energotec.de # 49211-9144018
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Shipping Frameworks Date: 3 Jul 1997 17:59:05 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5pgp99$299$1@owl.slip.net> Stil having a problems putting Frameworks into the app wrapper. Ralph's suggestion didn't work. (Was there bundle loading code that goes along with this?) Any help or suggestions would be much appreciated, -- Steve Dekorte - OpenStep consultant - San Francisco
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: Q:Java development in NS3.0? Date: Thu, 3 Jul 1997 13:54:17 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <MniyR9_00YUs0KYkU0@andrew.cmu.edu> References: <ECp4wH.EE@ecf.toronto.edu> Excerpts from netnews.comp.sys.next.programmer: 2-Jul-97 Q:Java development in NS3.0? by Murugathas Yuwaraj@ecf.t > Could someone tell me if JDK is available for NS3.0? Yes; the JDK is not available for NS3.0. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 3 Jul 1997 18:40:29 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5pgrmt$2ur$1@owl.slip.net> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5pb35d$6ui2@castor.cca.rockwell.com> Erik M. Buck <embuck@palmer.cca.rockwell.com> wrote: > By the way, has anyone else noticed that NeXT lost their polish somewhere > between 3.3 and 4.0 ? Everything used to be so well though out. I feel like > I am using an experimental system now. Yes, I've noticed that. Has NeXT been able to retain it's talent? (aka show it's engineers the money) -- Steve Dekorte - OpenStep consultant - San Francisco
From: Bill Perkins <perk@iag.net> Newsgroups: comp.sys.next.programmer,comp.unix.programmer,comp.unix.questions Subject: Re: Are these entry-level-job handicaps show-stoppers?" Date: Thu, 03 Jul 1997 16:07:57 EDT Organization: Internet Access Group, Orlando, Florida Message-ID: <5ph10f$5ol$1@news.iag.net> References: <332DA1F1.41A4@hom.net> <5h7eed$ntq$1@news.platinum.com> <01bc85a2$01ec70c0$bc65a8c6@metnews.hip.cam.org> OK, I'll take a crack at this: I don't have a degree of any sort; some college, not enough for that piece of paper saying that I know what I'm doing, as yet. However, about 13 years ago, I got lucky when somebody realized that I knew something about computers, so they put me in a position where I could really show them. Since that time, I have not had the very _best_ of luck finding jobs, but it hasn't been real terrible, either. Entry level (at least, as far as I can tell) means that you know what a computer is, you know how to run at least one compiler, you have a favorite editor, and you know how to run a debugger effectivly. It also means that when the boss says "I need a small program to generate a report of statistics from these data files" that you will know at least enough 'C' to be able to open files, parse the data in that file (if you know the format of the data), and print out some simple information as the file is read. If you can pass a simple programming test (most companies have them) and not get too self-concious about your lack of experience/schooling/whatever, you _might_ be able to get a foot in the door. I also recommend the books and procedures outlined in the prior message; UNIX (Linux, FreeBSD, etc) is a wonderful programming environment. However, to be safe, also be familiar with the Wintel programming "environment". As crummy as it is, _most_ companies out there _really_ want you to be able to deal with Visual C and all the other Micro$haft stuff. Given a limited budget, I would try and find someone with an unused copy of the Micro$lush compilers and have them "transfer" it to you (don't pirate it, please. It's just not a good habit to get into, IMHO.) Learn enough about it that you know it's limitations. Then get into a _real_ programming environment like UNIX, and have a ball. If you can say you've been writing and debugging software for 6 months to a year, and maybe have some shareware outthere for your prospective boss to look at, it'll help. Sorry to ramble, but I hope this helps. "Mark" <mp@digiplace.com> wrote: >A few months back Gary Longsine posted a thoughtful, detailed response >(quoted at the bottom of this post) in reply to somebody wondering about >getting an entry-level programming job. > >I'm wondering if he or anyone else would like to comment on my situation - >it's similar but with a difference: I don't have the CS degree. > >I lack no confidence that with some diligent work I can become competent in >Unix programming, Objective-C, the OpenStep API, and fundamental OOP >principles. > >However I'm not confident that I can land an entry-level programming job of >some kind because > > * I'm in my mid-thirties > > * never finished my CS university degree > > * and I have no workplace programming experience. Etc. Sorry for the snip, I've got a flaky new processor here. ---------- huge snip ------------------------------------------------------- -- ------------------------------------------------------------------------------- ________ | Bill Perkins / "Ship Arriving Too Late to | perk@iag.net / Save a Drowning Witch" | programmer-at-large / /\ F. Zappa | ALL assembly languages done here. -------------------------------------------------------------------------------
From: Dru Nelson <dnelson@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 3 Jul 1997 22:35:57 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5ph9gd$5dp$1@owl.slip.net> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> Ok, is a non mutable class copy REALLY just a retain/autorelease ? Dru
From: penrose@sfc.keio.ac.jp (Penrose Christopher) Newsgroups: comp.sys.next.programmer Subject: HELP!: NSThread severely degrades NSEvent queue access speed Date: 04 Jul 1997 00:03:36 GMT Organization: Keio University Shonan Fujisawa Campus, Kanagawa Japan Distribution: world Message-ID: <PENROSE.97Jul4090336@ccs01.sfc.keio.ac.jp> Howdy Folks: I am developing an application which is multithreaded. There is a single method that is spawned with: [NSThread detachNewThreadSelector:@selector(mySelector:) toTarget:self withObject:sender]; In this method (as referenced by mySelector) there are no AppKit objects manipulated, nor are any messages passed to any AppKit objects. Elsewhere, in a custom view, mouseDown: is overridden to implement the dragging of another view. The dragging is smooth before the thread is detached and it is smooth after mySelector: is executed if it is not executed in a seperate thread. However, as soon as the application becomes multithreaded, e.g. mySelector: is executed in a seperate thread, the dragging behavior degrades drastically: the view stutters when dragged. A huge (sometimes as humungous as 500ms) latency is introduced. This latency remains in the application ever after, even though the detached thread has exitted. Why is this happening? Are there any ways to alleviate this unacceptable latency? I am tempted to use cthread_fork() to see if I can fool the AppKit into better performance, but I'd prefer this application to be Openstep compliant. Thanks for any help! Christopher Penrose penrose@sfc.keio.ac.jp penrose@silvertone.princeton.edu
Newsgroups: comp.sys.next.programmer From: Information Subject: Metrics Message-ID: <737cd$1131d.d6@news.psrinc.com> Date: Fri, 04 Jul 1997 01:49:13 GMT Great Site URL:http://www.psrinc.com/metsys.htm
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <737cd$1131d.d6@news.psrinc.com> Date: 4 Jul 1997 01:36:05 GMT Control: cancel <737cd$1131d.d6@news.psrinc.com> Message-ID: <cancel.737cd$1131d.d6@news.psrinc.com> Sender: Information Spam cancelled by sef@kithrup.com
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 4 Jul 1997 03:26:50 GMT Organization: WolfWare Message-ID: <5phqhq$ied$1@vader.wolfware.ipc.net> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> In-Reply-To: <5ph9gd$5dp$1@owl.slip.net> On 07/03/97, Dru Nelson wrote: > >Ok, > > is a non mutable class copy REALLY just a retain/autorelease ? > >Dru No. In many cases it's just a retain though. (No autorelease.) Copy operations are like init operations - they return an object with a retain count of 1 - you have to handle releasing it yourself. - Chris -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: Joseph Panico <jpanico@online.disney.com> Newsgroups: comp.sys.next.programmer Subject: Re: WOApps & Sessions Date: Thu, 26 Jun 1997 08:30:45 -0400 Organization: Disney Online Message-ID: <33B260F5.A9B7CE13@online.disney.com> References: <33B0F48E.38AADB32@ctp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Al Sheehan wrote: > Hi all, > > I'm doing a project in which there are a number of WOApp's. Now, form > the login screen you choose which WOApp you want to start. > I am wondering is there a way of forcing the session number to be what > > ever one you generate, and is there a way that the WOApp's > can capture this session number before fully launching, in order to When you say session number, are you referring to the SessionIDs that WOF uses to track user sessions, or do you also have your own notion of what a session number is and how it is tied to a particular persistent user? I don't think you've given us enough information. > authenticate. > Any ideas, thoughts or Six foot blondes looking for a good time ?? > > Al ... -- Joe Panico Disney Online jpanico@online.disney.com
From: kevin parks <kpp@dartmouth.edu> Newsgroups: comp.sys.next.software,comp.sys.next.programmer Subject: *.wn -> .rtf ? Date: Thu, 03 Jul 1997 23:59:01 +0000 Organization: dartmouth college - music dept. Message-ID: <33BC3CC3.2630@dartmouth.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit CC: comp.sys.next.misc Hi! I may be having to shut down my old beloved NeXTs for good soon and I need to make some backups of all my WriteNow texts and have been converting them to rtf on the NeXT end of things and porting them to them MAC and that has worked without a hitch so far. However I have like 5 years worth of files and this could take forever. Is there any way to convert all my WriteNow files to rtf in batch. instead of opening each individual file and doing a save as. like a: % mkrtf -R *.wn Claris's open menu has an option for WriteNow NeXT, but it simply doesn't work. Probably because any NeXT wn file is really a folder in wrapper. any help would be appreciated. I look in my file viewer at all my articles and correspondence, etc with all those .wn extensions and just sigh. cheers, kevin parks
From: "miguel angel gutierrez" <maguti@mx2.redestb.es> Newsgroups: comp.sys.mac.printing,comp.sys.mac.programmer.help,comp.sys.mac.system,comp.sys.misc,comp.sys.newton.marketplace,comp.sys.newton.programmer,comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.northstar,comp.sys.palmtops,comp.sys.powerpc.tech,c Subject: DINERO FACIL LO ASEGURO FUNCIONA Date: 1 Jul 1997 01:21:35 GMT Organization: Unisource Espana NEWS SERVER Message-ID: <01bc85bc$d8e0faa0$LocalHost@miguel> ********************** DINERO FACIL *********************** ¿Quieres ganar dinero fácil y rápido?. Si es así por favor sigue leyendo. Esto realmente funciona. Es la manera mas justa y honesta de poder ganar algun dinero extra sin invertir apenas nada. Invierte 5 minutos de tu tiempo en leer esto y CAMBIARA tu vida. El Internet ha crecido enormemente, su tamaño se duplica cada 4 meses. Piensa que si ves cada vez mas este tipo de noticia es por que funciona y dado que son mas los puntos positivos que los negativos. Piensa que puede pasar, quizá funcione además, solo son 3000 pesetas. Muchos escépticos pensarán que es un engaño, PIÉNSALO POR UN MOMENTO!! Hay tantos nuevos suscriptores a Internet de todo el Mundo sin contar a los grandes proveedores en EEUU, ellos se interesaran y lo harán funcionar por lo simple del sistema. Así es como funciona en 3 sencillos pasos: PASO NUMERO 1: Invierte 3000 pesetas, escribiendo tu nombre y dirección en tres hojas con la leyenda "Por favor agregeme a su lista de correo", (de esta manera no estas regalando un dólar a alguien estas pagando por un servicio completamente legal). Envuelve cada billete de 1000 en cada hoja de papel y envíalas por correo a estas tres direcciones: 1. M.Angel Gutierrez. C/ Sant Lluis nº 49, 1º2ª 37300 Peñaranda de bracamonte (salamanca) 2. Susana López Rodriguez C/ Baja de S Idelfonso,6 18010 -Granada 3. Ernesto Caballero C/ San Luis nº 49, 2º1ª 08024 Barcelona PASO NUMERO 2: Ahora elimina el #1 de la lista y recorre los otros nombre para arriba. De la manera siguiente el #2 es en #1, el #3 es el #2. Y pon tu nombre y dirección en el #3 de la lista. PASO NUMERO 3: Publica este artículo en por lo menos 250 grupos de noticias. Hay por lo menos 20000 grupos de noticias en cualquier hora del día. Mientras en mas grupos te anuncies, mas gente te vera y te mandara dinero. PASO NUMERO 4: Ahora es cuando empieza el negocio para ti, y empezaras a ver resultados entre los próximos 7 a 14 días. Recuerda que el Internet esta en crecimiento constante. No hay manera posible de perder. **************************************************************************** Para cuando ya no veas tu nombre en la lista, tomas el ultimo anuncio y empiezas de nuevo. El resultado final depende de TU PERSONA. Debes seguir este articulo y anunciarlo cuantas veces puedas y donde puedas. Mientras mas te anuncies mas dinero te llegara por correo. Es demasiado simple y sencillo como para dejar pasar esta oportunidad!!! Si no estas completamente seguro o piensas que no puede realizarse, no lo hagas. Pero por favor imprime este articulo y pásalo a alguien que realmente necesite el dinero y ve que pasa. PASOS GENERALES PARA AUTOMATIZAR EL PROCESO: Haz todos los paso necesarios para cambiar este articulo como se explica en el PASO NUMERO (2) y cuando estés satisfecho grábalo en formato .txt para importarlo en el cuerpo del programa de lector de noticias (ejemplo Internet News), De esta manera solo tendrás que cambiar el nombre del grupo o dirección de correo para distribuirlo. Consejo Importante: Es importante para aquellos no muy familiarizados con el email. Por todos los medios utilicen un procesador de textos para asegurarse de que el articulo lo vean clara y correctamente. Reformatear este articulo con un editor simple como "notepad," o "WordPad" si utilizas "Windows 95". Procura que el texto sea claramente visible y no se necesite mover a la derecha para leerlo. Hacerlo de esta forma asegura que el articulo se vera bien en todos los lectores de noticias. Puedes checarlo con tu lector antes de conectarse en línea, leyéndolo en tu propio lector. Enseguida empieza a buscar los grupos de noticias en donde te piensas anunciar, Netscape 3.0 es un lector muy bueno, por que puedes seleccionar varias docenas de grupos a la vez, permitiéndote distribuir tu articulo a miles de lugares en menos de una hora o dos, Selecciona todos los grupos a donde quieras mandar el articulo (como sabes se hace presionando deteniendo CTRL mientras oprimes el botón izquierdo del mouse). De esta manera se pueden seleccionar varios grupos en una pasada, quizá una docena cada vez. Después de eso veras los grupos seleccionados en el campo de grupos. Después selecciona tu grupo de noticias y oprime "a Noticias" procura poner un titulo adecuado en el Tema, luego oprime en agregados donde aparecerá otra ventana. Localiza el archivo que deseas enviar, oprime el archivo y luego ábrelo, después de eso envíalo y ya esta. Repite este proceso una y otra vez seleccionando de 10 en 10 los grupos, selecciona los de mayor afluencia y no anuncies este articulo con títulos muy rimbombantes solo ahuyentara a la gente. Este es una manera honesta y legitima de hacer dinero regularmente. Pero solo será así si el mensaje es sincero y claramente entendido por los demás lograras el éxito deseado. NOTA DE SOLICITUD: El sistema se basa en que todos seamos honestos, pero es demasiado tentador no molestarse en enviar por correo los sobres con los billetes de a dólar adentro. El éxito de este programa depende si se lleva a cabo y distribuido en un 500%. El sistema tampoco funcionara si la gente toma ventaja de ti y no sigue las instrucciones. Puedes sin embargo si deseas permanecer en el anonimato, usar un seudónimo, pero asegúrate de que tu dirección sea correcta. NOTA FINAL: Muchas de las ideas para hacer dinero, no importa que tan bien planeadas e implementadas simplemente no se levantan. Y en muchos de los casos debido a los costos de publicidad, pero la publicidad conseguida por medio de Internet es honestamente muy impresionante. Así que por favor estudia este articulo minuciosamente y tomate el tiempo que quieras, ya que cuando te decidas a participar estarás en camino a recibir bastante dinero. Por la misma naturaleza de este sistema no veras los resultados la primera semana. PERO AL COMENZAR LA SEGUNDA SEMANA TUS INGRESOS DEL CORREO PROVENIENTES DE TODO EL MUNDO VERDADERAMENTE TE SORPRENDERA!!!!! Por favor piensa esto seriamente, por que es una de las pocas oportunidades de hacer dinero rápido que realmente funciona. BUENA SUERTE!!! -- MIGUEL
Newsgroups: c,comp.sys.mac.printing,comp.sys.mac.programmer.help,comp.sys.mac.system,comp.sys.misc,comp.sys.newton.marketplace,comp.sys.newton.programmer,comp.sys.next.bugs,comp.sys.next.programmer,comp.sys.northstar,comp.sys.palmtops,comp.sys.powerpc.tech,control From: news@news.msfc.nasa.gov Message-ID: <cancel.01bc85bc$d8e0faa0$LocalHost@miguel> Control: cancel <01bc85bc$d8e0faa0$LocalHost@miguel> Subject: cmsg cancel <01bc85bc$d8e0faa0$LocalHost@miguel> no reply ignore Organization: Semi-Automatic Lupine Remover Date: Fri, 4 Jul 1997 04:47:31 GMT Sender: "miguel angel gutierrez" <maguti@mx2.redestb.es> ignore Make Money Fast post canceled by J. Porter Clark.
From: Joseph Panico <jpanico@online.disney.com> Newsgroups: comp.sys.next.programmer Subject: Re: Cross-Platform Applications under NextStep Question Date: Thu, 26 Jun 1997 08:33:18 -0400 Organization: Disney Online Message-ID: <33B2618E.B0997BD1@online.disney.com> References: <erich-2206971904580001@ppp-207-105-88-93.snrf01.pacbell.net> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Eric Harley wrote: > Hello, > I have heard alot of buzz about NextStep being a good development > platform > for other machines. > > I would like to build a program on NextStep and get it working under > the > Macintosh The MacOS is currently not supported by the OpenStep layer. You'll have to wait for apple to port the Yellow box to MacOS. > and Windows platforms without alot of modifications. Is this > possible? > > I currently own a 486DX4-100Mhz with 17MBs of RAM. Is this adequete > for > NextStep on Intel? Yes, that will work. Performance will probably be acceptable, but not great. > Thanks alot! > > -Eric Harley > erich@powerwareintl.com -- Joe Panico Disney Online jpanico@online.disney.com
From: kevin parks <kpp@dartmouth.edu> Newsgroups: comp.sys.next.software,comp.sys.next.programmer Subject: *.wn -> .rtf ? Date: Fri, 04 Jul 1997 00:05:16 +0000 Organization: Dartmouth College Message-ID: <33BC3E3C.49F5@dartmouth.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit CC: comp.sys.next.misc Hi! I may be having to shut down my old beloved NeXTs for good soon and I need to make some backups of all my WriteNow texts and have been converting them to rtf on the NeXT end of things and porting them to them MAC and that has worked without a hitch so far. However I have like 5 years worth of files and this could take forever. Is there any way to convert all my WriteNow files to rtf in batch. instead of opening each individual file and doing a save as. like a: % mkrtf -R *.wn Claris's open menu has an option for WriteNow NeXT, but it simply doesn't work. Probably because any NeXT wn file is really a folder in wrapper. any help would be appreciated. I look in my file viewer at all my articles and correspondence, etc with all those .wn extensions and just sigh. cheers, kevin parks
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 4 Jul 1997 06:42:37 GMT Organization: monoChrome, Inc., NJ, USA Message-ID: <5pi60t$o16$1@news2.digex.net> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> cwolf@wolfware.com (Christopher Wolf) wrote: > No. In many cases it's just a retain though. (No autorelease.) > Copy operations are like init operations - they return an object > with a retain count of 1 - you have to handle releasing it > yourself. Pardon me if this has been answered before (if so a pointer to the location would be appreciated...). But why the heck can't apple/next just put in a kick but garbage collection scheme that will work decently the bulk of the time, and offer a more manual option for those that wish to employ that instead... Those that are in the know and want to do it themselves will likely be able to and figure out any issues, and those not in the know or too lazy or inexperienced will get the benefits of garbage collection gratis... -- Thanks, be well, take care, later, John Kheit; Self expressed... __________________________________________________________________ monoChrome, Inc. ASCII, MIME, PGP, SUN, & NeXTmail OK NeXT/OPENSTEP Developer mailto:jkheit@cnj.digex.net Telepathy, It's coming... http://www.cnj.digex.net/~jkheit New York Law School You're dangerous because you're honest
From: fpottier@pauillac.inria.fr (Francois Pottier) Newsgroups: comp.sys.next.programmer,comp.sys.next.software Subject: Re: CVS and .nib files Date: 4 Jul 1997 07:30:52 GMT Organization: INRIA Rocquencourt, BP 105, 78153 Le Chesnay Cedex, France Message-ID: <5pi8rc$a6g@news-rocq.inria.fr> References: <5p0g4q$dlh@fido.transit.ru> <5p0sdu$nea@iserver.stem.com> <33ba574c.0@epflnews.epfl.ch> <5pfda8$3hg$1@hermes.is.co.za> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Cc: In article <5pfda8$3hg$1@hermes.is.co.za>, <gvandyk@icon.co.za> wrote: >On 07/02/97, Vincent Kohler wrote: >The only problem I had with wrappers is that I couldn't find the >"wrap" and "unwrap" Utilities and I wrote my own. Was this the right >thing to do or did I miss something somewhere My distribution of cvs didn't have wrap and unwrap either (it's a cvs 1.9 which I found on the RedHat Linux CD). I simply use tar instead and everything seems to work fine. -- François Pottier Francois.Pottier@inria.fr http://pauillac.inria.fr/~fpottier/
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 4 Jul 1997 07:52:52 GMT Organization: 21st Century Software, New York City Sender: daver@ts3-4.nj.cnct.com Message-ID: <5pia4k$gnt$2@darla.visi.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 04 Jul 1997 02:52:52 CDT Cc: dnelson@slip.net In <5ph9gd$5dp$1@owl.slip.net> Dru Nelson wrote: > is a non mutable class copy REALLY just a retain/autorelease ? No. Sometimes it's just a retain. The point is, though, that you shouldn't even _know_ it's a retain or not. It has no bearing on the object other than performance. Since the object isn't mutable, there's no need to make a new object, just make another note of its use. Sort of like shared libraries, or copy on write memory. Autorelease is a separate mechanism. Autorelease just means the object will get a release message at some point later, specifically, when the current pool is freed. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: Java->ObjC Anyone? Date: Fri, 04 Jul 1997 04:43:50 -0600 Organization: ErgoTech Message-ID: <33BCD3E6.7999@ergotech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Since Apple has embraced java I think it's politically correct to ask this question. Java and objective-C use the same object model and are very similar in many way. I suspect that it would be possible to pre-compile java into objective-C source code. This would provide two things; JIT java and the ability to link java and objective-C together. JIT java would allow java code to be compiled to a real executable, forget bytecode and the java VM. The ability to link the two together would allow development of bundles and application extensions in java and a possible migration path to java from objective-C. Is anyone working on this? Is anyone interested in working on it? In my naiveté did I miss something that makes this impossible? It seems to be such an obvious step that I'm concerned that I totally overlooked some major obstacle. Jim
From: gtupar@ctp.com (Georg Tuparev) Newsgroups: comp.sys.next.programmer Subject: Re: CVS and .nib files Date: 4 Jul 1997 10:49:20 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5pikfg$10b$1@concorde.ctp.com> References: <5pi8rc$a6g@news-rocq.inria.fr> Hey folks! Below are my modified versions. Note that this hack does not work in client/server mode! After a long conversation with the NeXT technical support I decided to write a complete CVS solution that works in client-server mode, notifies the PB, IB and EOM and does not fuck my project regularly. When we receive our official OS 4.2/MACH and finish the current project (probably beg. of august) I'll put everything together and make it public. BTW, I'll also submit a patch to the diff. With the version below we can work on NT and mach, but it is not fun :-( Another warning: Better do not use SCM if you are working on UI application or your project has an eomodeled directory! Use plane CVS instead.... Have fun -- georg -- cvswrappers: ... *.nib -k 'b' -f 'unwrap %s' -t 'wrap %s %s' -m 'COPY' *.eomodeld -k 'b' -f 'unwrap %s' -t 'wrap %s %s' -m 'COPY' *.wo -k 'b' -f 'unwrap %s' -t 'wrap %s %s' -m 'COPY' *.rtfd -k 'b' -f 'unwrap %s' -t 'wrap %s %s' -m 'COPY' *.template -m 'COPY' .. unwrap: #! /bin/sh # # unwrap - extract the combined package (created with wrap) # #ident "@(#)cvs/examples:$Name: $:$Id: unwrap,v 1.1 1996/04/18 17:24:07 tuparev Exp $" # move the file to a new name with an extension rm -rf $1.cvswrap mv $1 $1.cvswrap # untar the file if `gzip -t $1.cvswrap > /dev/null 2>&1` then gzcat -d $1.cvswrap | gnutar --preserve --sparse -x -f - else gnutar --preserve --sparse -x -f $1.cvswrap fi # remove the original rm -rf $1.cvswrap wrap: #! /bin/sh # # wrap - Combine a directory into a single tar package. # #ident "@(#)cvs/examples:$Name: $:$Id: wrap,v 1.1 1996/04/18 17:24:07 tuparev Exp $" # This script is always called with the current directory set to # where the file to be combined exists. but i may get called with a # path to where cvs first started executing. (this probably should be # fixed in cvs) so strip out all of the directory information. The # first sed expression will only work if the path has a leading / # if it doesn't the one in the if statement will work. DIRNAME=`echo $1 | sed -e "s|/.*/||g"` if [ ! -d $DIRNAME ] ; then DIRNAME=`echo $1 | sed -e "s|.*/||g"` fi # # Now tar up the directory but we now will only get a relative path # even if the user did a cvs commit . at the top. # gnutar --preserve --sparse -cf - $DIRNAME | gzip --no-name --best -c > $2 In article <5pi8rc$a6g@news-rocq.inria.fr> fpottier@pauillac.inria.fr (Francois Pottier) writes: > In article <5pfda8$3hg$1@hermes.is.co.za>, <gvandyk@icon.co.za> wrote: > >On 07/02/97, Vincent Kohler wrote: > > >The only problem I had with wrappers is that I couldn't find the > >"wrap" and "unwrap" Utilities and I wrote my own. Was this the right > >thing to do or did I miss something somewhere > > My distribution of cvs didn't have wrap and unwrap either (it's a cvs > 1.9 which I found on the RedHat Linux CD). I simply use tar instead > and everything seems to work fine. -- ------- /\/\ Georg Tuparev <georg_tuparev@ctp.com> | Currently in Dublin / /_ \ Cambridge Technology Partners 118/119 Lower Baggot Street \ / / Apollo House, Apollolaan 15 Dublin 2, Republic of Ireland \/\/ 1077 AB Amsterdam, The Netherlands Tel: +353(1)607-9083 Tel: +31(20)575-0492 Fax: +31(20)575-0500 WWW: http://www.ctp.com
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: NSString encoding inside property lists Date: 4 Jul 1997 11:05:03 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5pilcv$qq$1@concorde.ctp.com> Hi, What kind of encoding does an NSString use when it writes inself into a ASCII property list (which should be identical to what you get when calling -description on an NSDictionary) I poked around in the documentation and was not really able to find information on that...and NSString supports all sorts of wierd encodings. My best geuss up to now is that NeXT uses escaped octals ("\205") for chars which ar in the NeXT ASCII encoding table as defined in Appendix C. For chars which are not included it uses the 7-Bit Unicode encoding (NSUTFStringEncoding..as NeXT wrote: <<should we document this format?>>...and I think : "YES please" ) The Unicode stuff is escaped too ("\U4711") But then...this is just a guess and since we need to provide our clinet with a spec about how to create NSString compatible "ASCII-dicts" I would like to have more then just propable matching guesses. Any hint ? Reference ? Aloha Tomi P.S. We don't have the Unicode books yet..they are ordered but not here. So if all this is described in there a simple reference is all I need.
From: atze@aspohr.dart.de (Alexander Spohr) Newsgroups: comp.sys.next.programmer Subject: Re: Problem EOF procedure Date: 4 Jul 1997 11:26:24 GMT Organization: INTERSHOP Communications Message-ID: <5piml0$kob@linux1.netconx.de> References: <5oe5ju$asd@hpuniv.univ-lr.fr> bruno@gruick.univ-lr.fr (Bruno Garnier) wrote: > Hi All, > I work on OpenStep4.1 and EOF 2.0 and Oracle 7.3 > I try to use a stored procedure that have 100 parameters > BUt I can't retrieve it on EOModeler > SO I insert manually this stored procedure into my model > but when i try to execute this procedure in my project > I've got always the same error message : > *** -[OracleNumberColumn setColumnLengthAndClientTypeForAttribute:] unknown > number type: "*nil*" we have the same problem using EODatabaseContext batchFetchRelationship:forSourceObjects:editingContext: the generated sql is wrong. it contains nil entries. (t0.*nil* = *nil* AND t1.PRODUCT_ID = :productId) in fact these entries need not to be there to make the query work as supposed: (t1.PRODUCT_ID = :productId) if someone knows how to enable the right usage please help. Atze --- Alexander Spohr, NetMatic GmbH, Hamburg, Germany Mail: Alexander_Spohr@NetMatic.com Faces and faces Fax: +49 (0) 40 / 492 23 68 See them and complain not WWW: http://www.NetMatic.com/ And am content with all
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 4 Jul 1997 14:37:07 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5pj1qj$4pk$1@concorde.ctp.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> John Kheit <jkheit@cnj.digex.net> wrote: >cwolf@wolfware.com (Christopher Wolf) wrote: >> No. In many cases it's just a retain though. (No autorelease.) >> Copy operations are like init operations - they return an object >> with a retain count of 1 - you have to handle releasing it >> yourself. > >Pardon me if this has been answered before (if so a pointer to the >location would be appreciated...). But why the heck can't apple/next >just put in a kick but garbage collection scheme that will work >decently the bulk of the time, and offer a more manual option for >those that wish to employ that instead... Those that are in the >know and want to do it themselves will likely be able to and figure >out any issues, and those not in the know or too lazy or inexperienced >will get the benefits of garbage collection gratis... While this has been discussed during the past 2or more years the answer to me seems quite simple. Like someone already mentioned...its the 80%/20% typ of thing. Doing distributed garbage collection is not that trivial and NeXT needed a working solution to especially solve the DO problems which did really hurt badly during 3.x times. Now what is NeXTpple's solution ? IMHO it is Java. ObjC shares all the problems that C based languges have (pointers etc.). Java doesn't have these problems since it was designed "not to have them". I personally would be very happy I they would just provide bridges from the system frameworks to "smarter" languages so that you can choose if you wnat the low level control or smart GC. That's what Apple is at least heading for with the Java integration in Rhapsody. Aloha Tomi P.S. The sad thing is, that they are providing a Java-C++ like crappy syntax for ObjC instead of offering a Smalltalk-ObjC style syntax for Java. Sigh...
Newsgroups: comp.sys.next.programmer From: pshum@rock.ali.bc.ca (Paul Shum) Subject: Strange behavior of Text object Message-ID: <ECstuq.En5@gateway.ali.bc.ca> Sender: nobody@gateway.ali.bc.ca Organization: A.L.I. Technologies, Inc. Date: Fri, 4 Jul 1997 14:35:13 GMT Hi! I am using NeXTStep 3.3. I run into a problem which is really strange. I would like people to suggest ways to find out what is wrong or tell me what I did wrong or even locating the problem area. As my schedule is beginning to fall behind, I appreciate for any constructive suggestion. I am writing a feature bundle (say F) which will be loaded by an application (say A). My feature bundle is used to save a view with a text into an EPS file. When I was creating the F, I used an old version of A to test my source. Everything were fine. Now, when I tested F with a new version of A, the EPS file created has the text inverted (or more accurate is mirror image about the x-axis). The view on the screen is perfect. The text is upright. But, the created EPS file has a inverted image. When I used the old version of A to load my F, everything went back to normal which is a correct EPS file is generated. The above is only a description of the behavior. The actual functionality of F is much more complicated. Additional information: 1. A is compiled in a different machine. I just used the executable. 2. I suspect that the machines to compile A and F are different in terms of loading appropriate patch or software environment setup. Questions: 1. Who is the killer? A causes the inverted image or F causes the inverted image. It seemed to me that A causes the problem but it does not make sense because F has its own view and it generates its own EPS file. But, if F causes the problem, why does it work with the old version of A? (F has its own window.) 2. I compared the two EPS source. There are three differences: a. the correct image has one additional set of gsave and grestore. b. the ordering is a bit different. c. there is one additional EPS command "concat" for the correct version. The EPS code is generated with F without any change. Why does it generate different code? 3. Why does it look O.K. on the screen and the generated EPS file image is inverted? Your response is very much appreciated. If you need any further information, please do let me know. Thanks! Paul S. L. Shum _____________________________________________________________ Are you a Christian who happens to be a computer professional or a computer professional who happens to be a Christian?
From: suckow@bln.sel.alcatel.de (Ralf Suckow) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 4 Jul 1997 16:30:57 GMT Organization: Alcatel/Bell Distribution: world Message-ID: <5pj8g1$bh0@btmpjg.god.bel.alcatel.be> References: <5pi60t$o16$1@news2.digex.net> John Kheit <jkheit@cnj.digex.net> writes > cwolf@wolfware.com (Christopher Wolf) wrote: > > No. In many cases it's just a retain though. (No autorelease.) > > Copy operations are like init operations - they return an object > > with a retain count of 1 - you have to handle releasing it > > yourself. > Pardon me if this has been answered before (if so a pointer to the > location would be appreciated...). But why the heck can't apple/next > just put in a kick but garbage collection scheme that will work > decently the bulk of the time, and offer a more manual option for > those that wish to employ that instead... Those that are in the > know and want to do it themselves will likely be able to and figure > out any issues, and those not in the know or too lazy or inexperienced > will get the benefits of garbage collection gratis... In general I understand your complaints (esp. after using Java for almost two years), but in the above case its just not good style to not autorelease the copied object. Any object that is given out of a method to the caller should to be autoreleased at least in the case when the callee may have no reference to it anymore. This is the case with copy operations. So, the statement "in many cases it's just a retain though ... you have to handle releasing it yourself"of Christopher is simply missing the common practice. Either it is returned autoreleased, or you don't need to release it anyway. On the other hand, I like Java where due to interpretation automatic reference counting is easy and GC is not an 20/80 issue. Yours, Ralf ------------------------ Ralf.Suckow@bln.sel.alcatel.de | All opinions are mine.
From: Chuck_Esterbrook@orcacomputer.com Newsgroups: comp.sys.next.programmer Subject: OPENSTEP on 95 is several times slower than on NT! Date: 4 Jul 1997 17:44:51 GMT Organization: Virginia Tech, Blacksburg, Virginia Message-ID: <5pjcqj$mti$1@solaris.cc.vt.edu> We recently received OPENSTEP Enterprise 4.2 and are able to deploy our software product on Windows 95. However, we have discovered a serious problem with OPENSTEP 4.2 on Windows 95: our application runs 4 to 5 times slower than on Windows NT. This is the same executable binary on the same machine. Our software product, the Visual Simulation Environment, contains a discrete-event simulation engine that performs a lot of internal processing. It is not multi-threaded, does not use EOF, PDO, or any network features. We use only the Foundation Kit, the AppKit, and our own frameworks built on top of those two. I have been trying to identify the bottle-neck. Unfortunately, the Sampler application does not work on Windows 95. When I click on the "Start Sampling" button, it just beeps at me. Therefore I cannot find which method or function takes so long. Sampler works on NT, but it is in 95 that the slow down occurs. 1) Have others experienced this 95 slow down problem? If so is there a known workaround? I know that NeXT/Apple is busy with Rhapsody, but if this speed problem is a general OPENSTEP/95 problem it makes OPENSTEP almost unusable on Windows 95 (since there is no chance of competing with native Windows products when the speed difference is so enormous). 2) Can others run Sampler? Thanks, -- Chuck Esterbrook, Software Eng. http://www.orcacomputer.com/~chuck --------------------------------------------------------------------- chuck_esterbrook@orcacomputer.com / vo 540 231-3475 / fx 540 231-3480 Orca Computer, Inc. / 1800 Kraft Dr. Suite 111 / Blacksburg, VA 24060
Newsgroups: comp.sys.next.programmer,control From: news@news.msfc.nasa.gov Message-ID: <cancel.33bcb18f.0@news.genesisnetwork.net> Control: cancel <33bcb18f.0@news.genesisnetwork.net> Subject: cmsg cancel <33bcb18f.0@news.genesisnetwork.net> no reply ignore Organization: Semi-Automatic Lupine Remover Date: Fri, 4 Jul 1997 16:56:42 GMT Sender: a1@a.a ignore Make Money Fast post canceled by J. Porter Clark.
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 4 Jul 1997 20:05:21 GMT Organization: WolfWare Message-ID: <5pjl21$29t$1@vader.wolfware.ipc.net> References: <5pi60t$o16$1@news2.digex.net> <5pj8g1$bh0@btmpjg.god.bel.alcatel.be> In-Reply-To: <5pj8g1$bh0@btmpjg.god.bel.alcatel.be> On 07/04/97, Ralf Suckow wrote: >John Kheit <jkheit@cnj.digex.net> writes >> cwolf@wolfware.com (Christopher Wolf) wrote: >> > No. In many cases it's just a retain though. (No autorelease.) >> > Copy operations are like init operations - they return an object >> > with a retain count of 1 - you have to handle releasing it >> > yourself. > >In general I understand your complaints (esp. after using Java >for almost two years), but in the above case its just not good >style to not autorelease the copied object. > >Any object that is given out of a method to the caller should to >be autoreleased at least in the case when the callee may have no reference >to it anymore. This is the case with copy operations. > >So, the statement "in many cases it's just a retain though ... >you have to handle releasing it yourself"of Christopher >is simply missing the common practice. Either it is returned >autoreleased, or you don't need to release it anyway. Uh no. Copy is an EXPLICIT exception to this "common practice" that is clearly documented by NeXT. Copy is supposed to behave like alloc/init. In ProgrammingTopics/ObjectOwnership.rtf: ----- START QUOTE ----- Now that the concepts behind the Foundation Framework's object ownership policy have been introduced, they can be expressed as a short list of rules: If you allocate, copy, or retain an object, you are responsible for releasing the newly created object with release or autorelease. Any other time you receive an object, you're not responsible for releasing it. ----- END QUOTE ------ And in the NSCopying protocol documentation: ----- START QUOTE ----- The NSCopying protocol declares a method for providing functional copies of an object. The exact meaning of ›copy— can vary from class to class, but a copy must be a functionally independent object with values identical to the original at the time the copy was made. A copy produced with NSCopying is implicitly retained by the sender, who is responsible for releasing it. ----- END QUOTE ------ Note that both of these sources say the SENDER of the copy message is responsible for releasing the returned object. And also from the NSCopying protocol docs: ----- START QUOTE ----- Immutable classes can implement NSCopying very efficiently. Since immutable objects don't change, there is no need to duplicate them. Instead, NSCopying can be implemented to retain the original. For example, copyWithZone: for an immutable string class can be implemented in the following way. - (id)copyWithZone:(NSZone *)zone { return [self retain]; } ----- END QUOTE ------ - Chris -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: stark@easynet.fr (Frederic Stark) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 4 Jul 1997 23:34:25 GMT Organization: Self Message-ID: <5pk1a1$ngr@buggy.news.easynet.net> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5pb35d$6ui2@castor.cca.rockwell.com> <5pgrmt$2ur$1@owl.slip.net> <33BD742E.FB3@ergotech.com> > Steve Dekorte wrote: > > > > Erik M. Buck <embuck@palmer.cca.rockwell.com> wrote: > > > By the way, has anyone else noticed that NeXT lost their polish somewhere > > > between 3.3 and 4.0 ? Everything used to be so well though out. I feel like > > > I am using an experimental system now. There have been huge changes between 3.3 and 4.0, and I think they wanted to do too much things at once (ie: openstep and new developper tools) My 0.002$ on PB: I'd like to be able to alway use torn-off windows because I just hate the way the browser and the text get out-of-sync. [I mean, select a file in a subproject, tear it of, select a file in another subproject, click on the torn-off window, the browser get back to the first subproject, try to click somewhere in the browser (to open a related file), and the browser changes its selection after it become main, but before it accepts the event. I want the guy who designed that each time] [off-topic-on] Or, in 4.2pre, type the following method: - (float) and save here. Bang, project builder goes away. I would even have reported it if I could have launch BugNeXT.app [off-topic-off] I don't find new ProjectBuilder really easier to manage big projects. I would really love to have a shelf in it to access frequently used files/subprojects and to move them around. And, a way to open the header of a implementation file other than: * Alt-drag miniaturize button to content of browser text * Click-on "Headers" * Alt-T While we are at it, I hope they loose the source of the integration of gdb in project builder (hint, hint, guys at apple, "rm -rf"). Terminal does a much better job at handling text/selections/Alt-k. gdb window in PB is soooo slooooow. And is often inacurate if a lot of text is dumped on it (some output is just lost. Nice isn't it?). And when you click on the middle of the text and type, nothing happends. Ghosh. *Who* did that ? Or try to use the keyboard in gdb window: type 'n', and this half-dumb project builder make your source key window, so you can happilly override your source with gdb commands, Happends to me 3 or 4 times per debugging session. [Ok at least, now we can undo...] I understand that NeXT needed to have a OPENSTEP compliant PB, so couldn't base things on Terminal, but this should have never been released with such trivial bugs ! I could continue ranting for hours on the subject, and it is not really fair because I used VC++ a lot (and metrowerks a bit) and those are true crap compared to PB. As we says in French: "Qui aime bien chatie bien" Cheers, -- fred PS: Btw, new PB have nice features (undo region, auto-save, multi-file replace) and is a lot better than before.
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: Fri, 04 Jul 1997 16:07:42 -0600 Organization: ErgoTech Message-ID: <33BD742E.FB3@ergotech.com> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5pb35d$6ui2@castor.cca.rockwell.com> <5pgrmt$2ur$1@owl.slip.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Steve Dekorte wrote: > > Erik M. Buck <embuck@palmer.cca.rockwell.com> wrote: > > By the way, has anyone else noticed that NeXT lost their polish somewhere > > between 3.3 and 4.0 ? Everything used to be so well though out. I feel like > > I am using an experimental system now. > > Yes, I've noticed that. Has NeXT been able to retain it's talent? > (aka show it's engineers the money) > > -- > Steve Dekorte - OpenStep consultant - San Francisco How's WO? I've always suspected that to have been so much of a focus at NeXT that it really hurt the OpenStep development. Jim
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 5 Jul 1997 08:39:21 GMT Organization: 21st Century Software, New York City Sender: daver@ts1-4.nj.cnct.com Message-ID: <5pl17p$1h2$1@darla.visi.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5pia4k$gnt$2@darla.visi.com> <5pk6lp$s0v$1@owl.slip.net> <5pkv83$bh3$1@vader.wolfware.ipc.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 05 Jul 1997 03:39:21 CDT Cc: cwolf@wolfware.com In <5pkv83$bh3$1@vader.wolfware.ipc.net> Christopher Wolf wrote: > Is this a really wide-spread misconception that you are supposed to autorelease > objects returned by the copy method???? See my other post on this thread where > I explicitly show 3 places in NeXT's documentation where it says that it is the > SENDER's responsibility to release objects returned by copy. (Just as it is the > SENDER's responsibility to release objects returned by alloc/init.) He's getting confused by a couple posts back where I said I usually do return [[something copy] autorelease] in accessor methods. I think. Language is too imprecise, I'm going to sleep. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
From: lars.farm@ite.mh.se (Lars Farm) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Sat, 5 Jul 1997 12:45:05 +0200 Organization: pv Message-ID: <19970705124505517300@dialup76-3-36.swipnet.se> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> NNTP-Posting-User: s-49817 Charles William Swiger <cs4w+@andrew.cmu.edu> wrote: > Excerpts from netnews.comp.sys.next.programmer: 4-Jul-97 Re: [Q] > OpenStep Garbage Co.. by John Kheit@cnj.digex.net > > Pardon me if this has been answered before (if so a pointer to the > > location would be appreciated...). But why the heck can't apple/next > > just put in a kick but garbage collection scheme that will work > > decently the bulk of the time, > > They've already come about as close as they can without encountering > some fundamental issues-- namely, the fact that under C-derived > languages, it's possible to create arbitrary pointer references and it's > possible to do pointer arithmetic. These two properties make it > impossible for a garbage collector to be certain whether memory regions > are in use or not. This is not correct. A conservative garbage collector will work just fine with C or C++ and I assume therefor also with Obj-C. You'd have to work hard to fool the collector. Much harder than with non collecting allocators (where the collector is you). Arbitrary pointer references (I suppose you mean block referenced only by a pointer to the interior of an allocated block(?)) and pointer arithmetic is certainly not a problem. The things that might make the collector fail - collect and reuse live data - are things that invokes "undefined behaviour" and therefor is illegal and will probably fail anyway, with or without GC. Things like deliberately hiding address bitpatterns by for instance storing pointers to allocated blocks on disk (only) for later retrieval or splitting the address word in bytes stored away from each other and later combining to a proper address or other similarly contrived and illegal tricks. There are commercial conservative collectors available and Hans-Juergen Boehms free and excellent conservative collector works just fine in C or C++ on many systems, including Macintosh (CW and Symantec), NT, OS2 and various Unixes. Does it work on NeXT? I suppose so because of its Unix foundation. I'm a Mac programmer and know little about the NeXT stuff, but will hopefully learn. Speed is also often used as an argument against GC, but this is not a valid argument. Boehms free GC is a substantially faster allocator than Metrowerks or Symantecs non collecting allocators. Collections can be configured to be threaded and incremental where the OS permits or if one prefers as full stop-the-world collections. The latter is what most people are afraid of. On a Macintosh a full stop collection wont take longer than perhaps 10 ms and they are far between. With incremental collection there is no such issue. Many claim that reference counting is as good as GC and is enough, but this is not the case. Reference counting wont solve cyclic references and is not automatic. If you must manually increment/decrement a counter then where is the difference from saying new/delete? It's a very minor gain compared to proper garbage collection. -- Lars Farm; lars.farm@ite.mh.se - Limt/Channelmatic: lars.farm@limt.se
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: Re: libfl on NT? Date: Sat, 05 Jul 1997 05:19:36 -0600 Organization: ErgoTech Message-ID: <33BE2DC8.7851@ergotech.com> References: <5pe73m$eh8$1@nyheter.chalmers.se> <5pfbds$lv5@mochi.lava.net> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit CC: arti@lava.net Art Isbell wrote: > > Žoloft@hegel1.cs.chalmers.se.cs.chalmers.se (Olof Torgersson) wrote: > > > > I'm trying to convert a 3.3 project using flex and bison generated files to > > OPENSTEP. I managed to do it for mach, but I would also like to port it > > to NT. > > flex and bison are included with OS/NT, so no additional libraries are > needed. However, we had been using flex 2.5.2 under NS 3.3 because it's so > much nicer than the old BSD lex included with NS. To our dismay, when we > ported to NT, our flex source would no longer compile under the flex included > with OS/NT. The reason is that the flex version is VERY OLD and we were > taking advantage of newer flex features. It's so old that it doesn't respond > to the --version flag. It still uses flex.skel from 1990! > > So you may need to rewrite your flex source somewhat. Not sure about > bison because we had been using NS 3.3's yacc. You can convince the latest flex (2.5.4?) to compile on OS/NT and it seems to work fine although I haven't given it too much of a workout. Make links (shortcuts) from gcc.exe to cc.exe and create /bin with sh and sh.exe. You also need to create a couple of links during compilation (parse.XXX to any file that the parser says it can't find). You need to use -DYY_NEVER_INTERACTIVE when you compile the flex code to get rid of linker errors caused by references to isatty. We do need a replacement for bison on OS/NT. I believe that the licensing for bison still requires that you release the grammar under the gnu license. This is usually not acceptable for commercial applications. Jim
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: <ASE1000@1stfamily.com> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <61d7cd$123a30.19@NEWS> Control: cancel <61d7cd$123a30.19@NEWS> Date: 29 Jun 1997 23:50:13 GMT Message-ID: <cancel.61d7cd$123a30.19@NEWS> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: dyoung@vvi.com Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 5 Jul 1997 14:13:48 GMT Organization: VVI Data Control Specialists Message-ID: <5plkqs$3o9$1@news2.digex.net> Originator: gsupport@ Erik M. Buck wrote: > By the way, has anyone else noticed that NeXT lost their polish somewhere > between 3.3 and 4.0 ? Everything used to be so well though out. I feel > like I am using an experimental system now. Art Isbell wrote: > The NS 3.3 -> OS 4.0 changes were, by far, the most extensive changes > NeXT has made to its development environment. Essentially everything > changed. A small company like NeXT just didn't have the personnel to make > such sweeping changes without a degradation in the development environment. > NeXT was in a position that its only potentially successful commercial > direction was a Windows development environment, and that meant cleaving the > BSD UNIX underpinnings from the bottom of the development environment, a > major undertaking. > Apple simply must return the polish to Rhapsody ASAP which includes > comprehensive documentation. From my perspective the NS3.* -> OS4.* change is very positive. All the development tools are better and the underlying fundamentals of the NeXT supplied frameworks took a very positive step forward in several areas like object life times, transporting objects via DO, naming conventions, and increased functionality of each object class as well as new object classes. These changes have nothing to do with needing to support NT, although it definitely helps. The original NeXT framework was designed before 1987. The guys at NeXT learned a lot in the following decade and were building up to a major change in their frameworks. It wasn't a drastic overnight process, but rather a process that evolved over 10 years of design thought. The most concrete acknowledgment that OS4.* is better is in the redesign of third party code, and our code (about 1/2 million source code lines) definitely benefited in several substantial ways. -- David Young VVI-DCS ; http://www.vvi.com
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Fri, 4 Jul 1997 20:08:06 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <AnjN1ai00iWl05Awc0@andrew.cmu.edu> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> In-Reply-To: <5pi60t$o16$1@news2.digex.net> Excerpts from netnews.comp.sys.next.programmer: 4-Jul-97 Re: [Q] OpenStep Garbage Co.. by John Kheit@cnj.digex.net > Pardon me if this has been answered before (if so a pointer to the > location would be appreciated...). But why the heck can't apple/next > just put in a kick but garbage collection scheme that will work > decently the bulk of the time, They've already come about as close as they can without encountering some fundamental issues-- namely, the fact that under C-derived languages, it's possible to create arbitrary pointer references and it's possible to do pointer arithmetic. These two properties make it impossible for a garbage collector to be certain whether memory regions are in use or not. That is why Java has far more restrictive pointer semantics, since it permits Javas' GC to know for certain the status of all memory regions, which means that no explicit memory management (ala -retain/-release and so forth) is needed. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Dru Nelson <dnelson@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 5 Jul 1997 01:06:01 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5pk6lp$s0v$1@owl.slip.net> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5pia4k$gnt$2@darla.visi.com> David Young <dwy@ace.net> wrote: > In <5ph9gd$5dp$1@owl.slip.net> Dru Nelson wrote: > > is a non mutable class copy REALLY just a retain/autorelease ? > No. How about sometimes/maybe yes. :-) Even though everybody is saying that it is just a retain, that is just for the copy. Before you give it to the caller, you also put an autorelease on it so the caller doesn't have to worry about releasing it (I like free better BTW :) So although copy is just a retain on non immutable objects, there is usually an autorelease done on it after it has been copied. (aren't you mind readers? ;) New question: copy creates an immutable copy for every object immutable or not (if they are into that immutable/mutable distinction?), if I copy a Mutable, is it going to create a new object on the copy every time. My 99% belief is yes. So when I want to work with objects that change (it could happen :), I am creating a lot of objects. The Bottom line, I, the programmer, still have to consider memory allocation for variables when I structure a program, whether there is garbage collection or not. BTW, thanks to all the talk on this. It's interesting. Hopefully, there will be some better general documentation on openstep as time goes on.
From: arti@address.in.signature (Art Isbell) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 5 Jul 1997 03:44:12 GMT Organization: LavaNet, Inc. Distribution: world Message-ID: <5pkfuc$81q@mochi.lava.net> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5pb35d$6ui2@castor.cca.rockwell.com> <5pgrmt$2ur$1@owl.slip.net> Erik M. Buck <embuck@palmer.cca.rockwell.com> wrote: > By the way, has anyone else noticed that NeXT lost their polish somewhere > between 3.3 and 4.0 ? Everything used to be so well though out. I feel like > I am using an experimental system now. The NS 3.3 -> OS 4.0 changes were, by far, the most extensive changes NeXT has made to its development environment. Essentially everything changed. A small company like NeXT just didn't have the personnel to make such sweeping changes without a degradation in the development environment. NeXT was in a position that its only potentially successful commercial direction was a Windows development environment, and that meant cleaving the BSD UNIX underpinnings from the bottom of the development environment, a major undertaking. Apple simply must return the polish to Rhapsody ASAP which includes comprehensive documentation. -- Art Isbell NeXT/MIME Mail: arti at lava dot net Trego Systems (for whom I don't speak) Voice/Fax: +1 808 394 0511 OPENSTEP/NT Voice Mail: +1 808 394 0495 managed care solutions US Mail: Honolulu, HI 96825-2638
From: psyboyvych@inter-nexus.com Newsgroups: comp.sys.next.programmer Subject: Attention CGI/PERL/C++ Elite. Dream Job Offer. Date: Sat, 05 Jul 1997 05:00:20 GMT Organization: Uniserve Message-ID: <5pkk1n$3tj$16@neptune.uniserve.com> To the CGI/PERL/C++ Elite: A private organization needs your help to finnish a highly specialized HTML project called the "Internexus". The first 75 programmers who submit their name will be interviewed for the job. The people chosen will recieve shares in our Web Design company and an optional position as systems administrator and/or HTML author with the company. Approx. $18k to start, part time. You will not need to relocate as the virtual office works great! Submit you name & e-mail to the Program Director. He will contact you to set up an interview time. Please do not send any of your questions as we will not be able to reply. Save them for your interview:) Reply to texas@uniserve.com
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 5 Jul 1997 06:59:28 GMT Organization: 21st Century Software, New York City Sender: daver@ts1-4.nj.cnct.com Message-ID: <5pkrcg$qd7$1@darla.visi.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5pia4k$gnt$2@darla.visi.com> <5pk6lp$s0v$1@owl.slip.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 05 Jul 1997 01:59:29 CDT Cc: dnelson@slip.net In <5pk6lp$s0v$1@owl.slip.net> Dru Nelson wrote: > How about sometimes/maybe yes. :-) Well, from the class implementor's perspective, sometimes yes. But, for the developer using the class, the answer is: maybe, but don't count on it, so effectively no. > Even though everybody is saying that it is just a retain, that is just > for the copy. Before you give it to the caller, you also put an autorelease > on it so the caller doesn't have to worry about releasing it > (I like free better BTW :) So although copy is just a retain on non > immutable objects, there is usually an autorelease done on it after > it has been copied. (aren't you mind readers? ;) Yeah, I see what you mean now. If I return a NSString from a class' accessor method, it's [[string copy] autorelease], but what's really happening might be [[string retain] autorelease]. Either one is exactly what I want, though; I want to give the caller of an accessor method an object guaranteed to exist for at least the lifetime of the current autorelease pool, and no longer. Implementing [object copy] is a separate function from the above. > New question: copy creates an immutable copy for every object immutable > or not (if they are into that immutable/mutable distinction?), if I > copy a Mutable, is it going to create a new object on the copy every > time. My 99% belief is yes. So when I want to work with objects that > change (it could happen :), I am creating a lot of objects. If you copy a mutable instance, you get an instance of that class' immutable counterpart. For example, [[NSMutableString string] copy] returns a NSString, NOT a NSMutableString. Chances are, yes, you will get a lot of objects. All this is detailed in Protocols/NS[Mutable]Copying.rtfd in the Foundation framework documentation. > The Bottom line, I, the programmer, still have to consider memory > allocation for variables when I structure a program, whether there is garbage > collection or not. Yep. Except with automatic garbage collection, you often can't change its behavior (Java) and it tends to make you forget about what memory usage and the overheads of object instantiation and deallocation have on your program. > BTW, thanks to all the talk on this. It's interesting. Yea. > Hopefully, there will be some better general documentation on openstep as > time goes on. Heh, I don't know, the Foundation Reference is pretty good. It's the last word, in almost every case. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 5 Jul 1997 08:05:23 GMT Organization: WolfWare Message-ID: <5pkv83$bh3$1@vader.wolfware.ipc.net> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5pia4k$gnt$2@darla.visi.com> <5pk6lp$s0v$1@owl.slip.net> In-Reply-To: <5pk6lp$s0v$1@owl.slip.net> On 07/04/97, Dru Nelson wrote: >David Young <dwy@ace.net> wrote: >> In <5ph9gd$5dp$1@owl.slip.net> Dru Nelson wrote: >> > is a non mutable class copy REALLY just a retain/autorelease ? > >> No. > >How about sometimes/maybe yes. :-) > >Even though everybody is saying that it is just a retain, that is just >for the copy. Before you give it to the caller, you also put an autorelease >on it so the caller doesn't have to worry about releasing it >(I like free better BTW :) So although copy is just a retain on non >immutable objects, there is usually an autorelease done on it after >it has been copied. (aren't you mind readers? ;) Is this a really wide-spread misconception that you are supposed to autorelease objects returned by the copy method???? See my other post on this thread where I explicitly show 3 places in NeXT's documentation where it says that it is the SENDER's responsibility to release objects returned by copy. (Just as it is the SENDER's responsibility to release objects returned by alloc/init.) - Chris -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: Sat, 05 Jul 1997 05:08:16 -0600 Organization: ErgoTech Message-ID: <33BE2B20.64CA@ergotech.com> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> <5pb35d$6ui2@castor.cca.rockwell.com> <5pgrmt$2ur$1@owl.slip.net> <5pkfuc$81q@mochi.lava.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Art Isbell wrote: > > Erik M. Buck <embuck@palmer.cca.rockwell.com> wrote: > > By the way, has anyone else noticed that NeXT lost their polish somewhere > > between 3.3 and 4.0 ? Everything used to be so well though out. I feel > like > > I am using an experimental system now. > Apple simply must return the polish to Rhapsody ASAP which includes > comprehensive documentation. Some bug fixes to 4.2 wouldn't hurt. Are there any plans for this or is it just not worth reporting them? Once you get in a groove you can get PB to crash 3-4 times or more per day. Jim
From: "kdevries" <kdevries@theonramp.net> Subject: PROGRAMMERS- GEAC, DBS, MSA Newsgroups: comp.os.os2.programmer.tools,comp.sys.acorn.programmer,comp.sys.amiga.programmer,comp.sys.apple2.programmer,comp.sys.atari.programmer,comp.sys.ibm.pc.programmer,comp.sys.next.programmer,comp.sys.psion.programmer,comp.unix.programmer,comp.unix.sco Message-ID: <01bc8680$d22afde0$0dbb020a@default> Date: 2 Jul 97 00:42:07 GMT Fabulous opportunities exist for programmers with DBS e:series (Geac) experience. Opportunities nationwide. Excellent salary and benefits. Can work as a permanent employee or as a contractor. Send resume or brief work history to : KDEVRIES@THEONRAMP.NET OR contact: Karen DeVries 512-342-0302, phone 512-342-0305, fax
From: ftouhi@IRO.UMontreal.CA (Majid Ftouhi) Newsgroups: comp.lang.c++.leda,comp.sys.next.programmer Subject: compilation problem? Date: 5 Jul 1997 16:59:07 GMT Organization: Universite de Montreal Distribution: world Message-ID: <5plugr$16i@epervier.CC.UMontreal.CA> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Hi there: i have installed LEDA3.5 on NextStep for Intel: And i tried to compile this example: #include<LEDA/d_array.h> int main() { d_array<string,int> N(0); string s; while ( cin >> s ) N[s]++; forall_defined(s,N) cout << s << " " << N[s] << endl; return ( 0 ); } i get this : >g++ Essai.C -lP -lG -lL -lm ld: Undefined symbols: _cuserid can anyone help please??
From: Robert Anderson<megacash4u@hotmail.com> Newsgroups: comp.sys.next.programmer Subject: This news group is invited to a Secret online book ! All Ages Welcome! Date: 5 Jul 1997 17:12:15 GMT Organization: Anderson Services Distribution: inet Message-ID: <5plv9f$im1@news1.mnsinc.com> You are about to learn how ordinary people can simply and easily achieve the once thought un-achievable. These Secrets will work for anyone. Simply apply them. This online book is one to study. Write any ideas or notes down. Read this one Secret at a time. Taking notes as ideas cross through your mind. Do not skip any Secrets. I have arranged the Secrets to increase your readiness for the coming Information. Each Secret builds on the other for maximum results. I wrote this online book for you in a way that simplifies understanding. This should make the secrets easier to apply. Enjoy! CLICK HERE TO GET STARTED!!! http://members.aol.com/Lesson4u
From: maddog@blkbox.com (Gary Fox) Newsgroups: comp.sys.next.programmer Subject: RIP for Ultre 3000 Date: Sat, 05 Jul 1997 15:29:47 +0100 Organization: Mad Dog Productions Message-ID: <maddog-ya02408000R0507971529470001@news.blkbox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit I have heard that NeXT Step can Print to any PostScript Printer at any resolution, Is this still true. If so I need to find out how I can get it to RIP to an Imager from a Windows NT Machine.
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Sat, 5 Jul 1997 16:21:01 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <gnjemhO00iWp03KEs0@andrew.cmu.edu> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> In-Reply-To: <19970705124505517300@dialup76-3-36.swipnet.se> Excerpts from netnews.comp.sys.next.programmer: 5-Jul-97 Re: [Q] OpenStep Garbage Co.. by Lars Farm@ite.mh.se >> They've already come about as close as they can without encountering >> some fundamental issues-- namely, the fact that under C-derived >> languages, it's possible to create arbitrary pointer references and it's >> possible to do pointer arithmetic. These two properties make it >> impossible for a garbage collector to be certain whether memory regions >> are in use or not. > > This is not correct. It isn't? How so? > A conservative garbage collector will work just fine with C or C++ and I > assume therefor also with Obj-C. Why would one have to use a conservative GC in the first place? It's because the GC cannot be sure which regions are used or not because of C pointer semantics, so it _must_ use a conservative strategy. A Java GC does not have to be conservative because Java's pointer semantics guarantee that the garbage collector can determine with certainty whether a region is in use or not. > You'd have to work hard to fool the collector. Not very hard at all-- a circularly linked list or other self-referential data structure is generally sufficient to prevent a conservative GC from ever freeing the memory. > Much harder than with non collecting allocators (where the collector is you). > Arbitrary pointer references (I suppose you mean block referenced only by a > pointer to the interior of an allocated block(?)) I meant the fact that this type of behavior is permitted: #define SOME_INT (12345) // or pick your own magic number void *ptr = (void *) SOME_INT; > and pointer arithmetic is certainly not a problem. You are mistaken. It's perfectly valid for me to take a pointer to a memory region, and subtract 10000 from it if in the future, I always add 10000 to the pointer before dereferencing it. Of course, this is a simple and contrived example to demonstrate the problem. A non-contrived example would be something like simulating a base-and-bounds register combination for managing memory regions, where any pointer references within a particular region explicitly are added to the base register of that region in order to determine the actual memory address being referenced. Doing this type of thing will confuse even a conservative garbage collector into thinking the memory region is not in use when it is. > The things that might make the collector fail - collect and > reuse live data - are things that invokes "undefined behaviour" and > therefor is illegal and will probably fail anyway, with or without GC. You are mistaken. Go read a second edition of the K&R and you will discover that the ANSI-C language has precisely defined semantics for pointer arithmetic. > Things like deliberately hiding address bitpatterns by for instance > storing pointers to allocated blocks on disk (only) for later retrieval > or splitting the address word in bytes stored away from each other and > later combining to a proper address or other similarly contrived and > illegal tricks. There is nothing contrived about self-referential data structures or C pointer arithmetic semantics. They are a well-defined and legal part of the language, and it is possible to use such techniques without any problems under the standard ANSI-C environment. The fact that garbage collection algorithms break when confronted with conformant, working ANSI-C code indicates that adding GC was the problem. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: sams@best.com (Samuel G. Streeper) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 5 Jul 1997 15:42:17 -0700 Organization: BEST Internet Communications Message-ID: <sams.868141660@shellx> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> Charles William Swiger <cs4w+@andrew.cmu.edu> writes: >Why would one have to use a conservative GC in the first place? It's >because the GC cannot be sure which regions are used or not because of C >pointer semantics, so it _must_ use a conservative strategy. Also, all the conservative GC's I've seen have to scan the process's entire variable space, and if you don't tie into the VM system pretty tightly you will thrash the disk scanning the address space, hopefully not goofing up the VM's least recently used (or whatever) algorithm. This is not a very good solution. Also, someone just noted that it's easy for programmers to misunderstand the refcounting scheme and screw up a program because they don't understand their responsibilities (ie autoreleased status of copied objects.) If there is much turnover in your programming staff, the learning curve associated with this fragile system may well decelerate your deployment. The king has no clothes and this problem is not being properly addressed. It will become a big issue, if obj-c or at least openstep is ever to become a significant force. -sam
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5plv9f$im1@news1.mnsinc.com> Date: 5 Jul 1997 21:57:37 GMT Control: cancel <5plv9f$im1@news1.mnsinc.com> Message-ID: <cancel.5plv9f$im1@news1.mnsinc.com> Sender: Robert Anderson<megacash4u@hotmail.com> Spam cancelled by sef@kithrup.com
From: behappy@iveshown.com Subject: Enlow Gives Away Businesses... Newsgroups: comp.sys.next.programmer Organization: EEI Message-ID: <33bf0e22.0@news1.ibm.net> Date: 6 Jul 97 03:16:50 GMT Pissed of Private Investigator, tells "the rest of the story." What Mark Fraunfelder with Wired forgot to tell... http:/michaelenlow.by.net/spamwar.htm
From: luomat@peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.software,comp.sys.next.programmer Subject: Re: *.wn -> .rtf ? Date: 6 Jul 1997 02:50:17 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Message-ID: <5pn159$pst$2@bashir.peak.org> References: <33BC3E3C.49F5@dartmouth.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: kpp@dartmouth.edu In <33BC3E3C.49F5@dartmouth.edu> kevin parks wrote: > Is there any way to convert all my WriteNow > files to rtf in batch. instead of opening each individual > file and doing a save as. There's a wn2rtf binary at ftp://next-ftp.peak.org/pub/next/apps/utils/converters/wn2rtf.1.0.N.b.gz You can use it like this: for i in `/bin/ls *.wn` do short=`basename $i .rtf` wn2rtf $i > $short.rtf done exit 0 I am not sure how well it deals with anything which is not text. YMMV/etc/etc TjL ps -- Hebrew started as of 2 July, and ends on 22 August, so please understand if responses are slow. I am taking a 2 semester class in 8 weeks for 6 credits -- TjL <luomat@peak.org>
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 6 Jul 1997 11:32:36 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5pnvok$mca$1@concorde.ctp.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <sams.868141660@shellx> sams@best.com (Samuel G. Streeper) wrote: >The king has no clothes and this problem is not being >properly addressed. It will become a big issue, if obj-c >or at least openstep is ever to become a significant force. > Pardon ? C++ _is_ a significant force and it deals with this issue in an even crappier way (by ignoring it) So OpenStep at least goes 50% of the way. Java's GC does 98% (since its not very efficient..mark and sweep) And it sure will take some time until the crowds realize that their king needs some really good clothes (Smalltalk, Dylan, etc.) Aloha Tomi
From: lars.farm@ite.mh.se (Lars Farm) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Sun, 6 Jul 1997 16:39:24 +0200 Organization: pv Message-ID: <199707061639241455966@dialup156-4-13.swipnet.se> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> NNTP-Posting-User: s-49817 For better explanations than I can give ( and a working free GC ) see: http://reality.sgi.com/employees/boehm_mti/gc.html Charles William Swiger <cs4w+@andrew.cmu.edu> wrote: > Why would one have to use a conservative GC in the first place? One wants GC (well, at least I do;-), one wants it to be entirely transparent and one wants it to perform well. How this is implemented is a don't care. It so happens that conservative collectors fills those requirements and works very well in C/C++ where other garbage collection techniques fail and that is why one wants to use a conservative collector. Your question might be interpreted to imply that conservative collectors are bad. I have no reason to beleive they are bad and in my experience they are not bad. Speed is comparable to or faster than other techniques and they are safe. So my question to you is: Why would one not want to use conservative GC? Perhaps you are asking why one wants GC at all? or why one would prefer GC over reference counting or fully manual memory management? There are lots of reasons: Development effort. Reusability. Funcionality. Safety. Debugging. Space performance. CPU-performance... See for example: http://reality.sgi.com/employees/boehm_mti/issues.html > It's > because the GC cannot be sure which regions are used or not because of C > pointer semantics, so it _must_ use a conservative strategy. Yes. This can also be interpreted as if you consider this a problem, but it is not a problem. Conservative GC works fine. > A Java GC does not have to be conservative because Java's pointer > semantics guarantee that the garbage collector can determine with > certainty whether a region is in use or not. The only thing this might theoretically buy you is speed and Boehm claims that his conservative collector is faster than the Java GCs he has encountered. I don't know Java and am not particularly interested, but for C++, experience shows that his allocator with conservative collections beats most manual malloc/free new/delete without collectors so where is the problem? There is no problem! > > You'd have to work hard to fool the collector. > > Not very hard at all-- a circularly linked list or other > self-referential data structure is generally sufficient to prevent a > conservative GC from ever freeing the memory. No. If there is no path from a root pointer to a block then it can be collected even if there are pointers pointing at it from other unreachable blocks including itself. The key is that there has to be a path from the root set to the block. Not that there might be pointers pointing at it. If those other pointers are unreachable then the block can still be collected. root : free store set : +---------------------------------------+ : | +-------------+ +-------------+ | +--+ : +->| |----->| | | | |-------->| | | +-+ +--+ : | |<-----| | : +-------------+ +-------------+ P A B If you break the link or links between root set (your program variables) and the free store then there is no way to reach any of the blocks in the group of blocks on the free store and they are all garbage that can all be collected. A conservative collector can handle this. The root set is global + local vars including registers (+ anything that you might want to manually add to the root set). The root set is expanded by the dynamically allocated blocks reachable from the roots. If it is impossible to construct a path from the root set to a block then it is unreachable by the user and therefor garbage that can be collected. Self-referential data or a circular linked list will certainly not fool a conservative collector. When no one is holding on to it, it can be collected. Reference counting is unable to cope with this kind of data because, looking at figure above, there are three references to A and dropping P wont turn that into zero. It is still referenced by two pointers and A wont go away unless you manually make it go away. If you must do that then why bother with reference counts at all? This is manual deallocation. > > Much harder than with non collecting allocators (where the collector is > > you). Arbitrary pointer references (I suppose you mean block referenced > > only by a pointer to the interior of an allocated block(?)) > > I meant the fact that this type of behavior is permitted: > > #define SOME_INT (12345) // or pick your own magic number > void *ptr = (void *) SOME_INT; GC operates on dynamically allocated data. Pointers that you've received from the allocator. There is no way that you can predict the actual address of a block received from the allocator so that it can be hardcoded like your example. The above will not fool the collector because it has nothing to do with the free store. There is a small possibility that 12345 will be interpreted as a pointer to allocated data and therefor will prevent collection of that block for a while. This probability is low and usually quite harmless. The worst that can happen is that a block of memory is held back from reuse until the false pointer is gone. Certainly there are ways to fool the collector, but they all include strange and wonderful treatment of pointers. Techniques best avoided anyway and almost always illegal. Things that might happen to work on this particular version of this particular compiler, but that will break on another compiler or the next version of the compiler. In this area of "undefined behaviour" there are things that may fool the collector. There are always better ways than utilizing "undefined behaviour". By the way "undefined behaviour" is a term defined in the C++ language spec. It means illegal, but no diagnostic required. The program may or may not appear to compile, and anything may happen when the program is run (including what one expects). It's simply not defined and therefor not something to rely on. Avoiding such techniques should be especially important on OpenStep/NeXT/Rhapsody or whatever it will be called because of its portability claims. These programs will run on many different processors, be compiled with many different compilers and run on many different core OS. Avoiding "undefined behaviour" ought to be very important in that context. One way of reducing that problem is to localize the use of implementation peculiarities to one single module - the collector. Makes porting easier. > > and pointer arithmetic is certainly not a problem. > > You are mistaken. It's perfectly valid for me to take a pointer to a > memory region, and subtract 10000 from it if in the future, I always add > 10000 to the pointer before dereferencing it. Of course, this is a simple > and contrived example to demonstrate the problem. No, I'm not misstaken and no, your example is definitely illegal. A pointer must point at an object otherwise it's not valid and may not be used. For arrays a pointer must point at one of the elements of the array or at the imaginary nth+1 element directly after the array. No other pointers are valid. Using such pointers is illegal and invokes "undefined behaviour". When reading the sections in the language spec (5.7/5 and /6) describing pointer arithmetic please note the recurring phrase "...otherwise, the behavior is undefined." and undefined behaviour is bad news. > A non-contrived example would be something like simulating a > base-and-bounds register combination for managing memory regions, where > any pointer references within a particular region explicitly are added > to the base register of that region in order to determine the actual > memory address being referenced. > > Doing this type of thing will confuse even a conservative garbage > collector into thinking the memory region is not in use when it is. For a conservative collector to work there has to be a pointer, reachable from the roots, to or into, the allocated block. Pointers derived from that by pointer arithmetic (legal or not) pointing outside the block has no effect. So all it takes is: Thing* holder = new Thing; Thing* p = holder; p -= 50000; /* no problem */ (instead of holder -= 50000;) This is illegal anyway and wont work on some segmented architectures. > > The things that might make the collector fail - collect and > > reuse live data - are things that invokes "undefined behaviour" and > > therefor is illegal and will probably fail anyway, with or without GC. > > You are mistaken. Go read a second edition of the K&R and you will > discover that the ANSI-C language has precisely defined semantics for > pointer arithmetic. I am still not misstaken and legal pointer arithmetic will not fool a conservative collector. The pointer arithmetic C++ (I know the C++ language spec but in this area I have no reason to beleive ANSI-C is different) allows you to do is rather more restricted than most people expect. For one thing, all arithmetic is constrained by the definition of a valid pointer. Valid pointers can't point anywhere. They can only point at, into or directly after an array. If they don't then they by definition are unusable and using them anyway results in "undefined behaviour" - a program error. > The fact that garbage collection algorithms break when confronted with > conformant, working ANSI-C code indicates that adding GC was the problem. Conservative garbage collection does not fall apart when confronted with conformant, working ANSI-C code;-) It may fail with non conformant code where the programmer plays unportable and illegal tricks with pointers, like I explained in the first post. A short quote from Hans Boehm's web page mentioned above: > Conservative garbage collectors require mild restrictions on both the > source code and the compiler. The source program must not ``hide'' > pointers in such a way that they are invisible to the collector. There > are very few ways in which a strictly conforming ANSI C program could > hide pointers. Writing and retrieving them from a file is one way. For > some collector configurations, using memcpy to copy pointers to > unaligned locations is another. Casts tfrom pointers to integers and > back to pointers can cause problems. The most common problem is probably > to refer to an object by pointing to a known offset before the beginning > of the object. The last one is clearly not legal C code. All are > uncommon even in existing code. Another way to hide pointers that would succeed to fool the collector taken from Macintosh programming would be to store the *only* reference to an object in the RefCon field of a WindowRecord. The WindowRecord is allocated by a system call with memory outside of the collectors reach. Easily avoided. - Lars -- Lars Farm; lars.farm@ite.mh.se - Limt/Channelmatic: lars.farm@limt.se
From: dre32d@msn.com Newsgroups: comp.sys.next.programmer Subject: sell/your/photos$$$$$ Date: Sun, 6 Jul 1997 11:16:36 Organization: Miracle Net Communications, Inc. Message-ID: <5pod4e$47k$827@roadrunner.miracle.net> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit We are a New York based international photo agency representing professional and amateur photographers in search of having their work promoted in order to conduct their profession or hobby minus the time consuming inconvenience of presenting and selling their own work. You do not have to be a professional photographer to sell your photographic work successfully, and as any pro will tell you, quite often throughout history an amateur stumbles on a scene that he or she photographs and ends up with a photo that becomes a classic. Those who publish magazines, books, newspapers, greeting cards, record albums, or just about anyone who works with images in their profession, are in desperate need of fresh material, and are more than willing to review photos submitted by professional photo representatives. Sometimes the photos of amateurs are sought after more so than professionals because of accidental originality and innovative themes. At Photo Phoenix International, we delight in and value the work of people who view photography as a fine art and understand a photo representative's significance in terms of a photographer's career, reputation, and monetary gain. The photos we are seeking are some of the following: ANIMALS, ARCHITECTURE, ART, AUTOMOBILES, BUSINESS, CELEBRITIES, DOCUMENTARY, ENTERTAINMENT, EROTICA/NUDITY, FASHION, FOOD, HEALTH, HISTORY, HOBBIES, OUTDOORS, PEOPLE, POLITICAL, PORTRAITS, STILL LIFES, RELIGIOUS, SCIENCE, SPORTS, TRAVEL . . . Whether your work portrays conservative, experimental, stylish, or innovative themes, your project proposal should be well thought out before submitting to us. For first contact, submit a query letter and "SAMPLES ONLY" of your photos. You must include "2" self addressed, stamped envelopes--one for correspondence, the other for the return of your material should we not be interested. If we think your work is sellable, we will respond as soon as possible. Do not send entire portfolio unless we ask for it. SEND TO: Photo Phoenix International <33-29 58 Street> <Woodside, New York> <11377> <~Phone: 718-651-8145>
From: wtetwutqrttirtyjr@kjpttypppstwtthy.com Newsgroups: comp.sys.next.programmer Subject: YOUR VIRTUAL OFFICE CAN DO EVERYTHING!!! Date: 6 Jul 1997 16:04:18 GMT Organization: Internet MCI Message-ID: <5pofm2$4e$2@news.internetmci.com> One-Number service puts you in control of your calls while ensuring that your calls, messages and faxes reach you regardless of your location. One-Number service does this by linking your existing telephones, cell phones, pagers, and fax machines through a single telephone number. YOUR One-Number ACCOUNT COMES STANDARD WITH THESE FEATURES - Consolidates your phone, fax, voice-mail and paging services - Works with your existing products and services, no need to change - Easily lets you designate yourself as available or unavailable - Rings you at up to three numbers simultaneously - Never-busy number accepts multiple incoming calls and faxes - Call Screening (by name) lets you know who is calling - Call Waiting lets you know somebody else is calling - Call Move lets you easily transfer calls among different phones - Call Conferencing allows conferences with other parties - Fax Notification on your pager for incoming fax messages - Redirect calls to Personal Operator (live person) if desired - AND MUCH MORE Virtual Office Communications Company is not a phone company, a long distance carrier, nor do we sell telephone equipment, cellular phones, cellular service, pager or pager service. Instead, we offer service to integrate your existing products, adding value to what you already have by letting your phones, pagers, fax machines and voice-mail systems work together seamlessly. To learn more please visit our web site at: http://www.mynumber.com Please do not reply by e-mail as we are not set up to respond to e-mail requests.
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 6 Jul 1997 18:48:34 GMT Organization: 21st Century Software, New York City Sender: daver@ts1-4.nj.cnct.com Message-ID: <5popa2$sfp$1@darla.visi.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <199707061639241455966@dialup156-4-13.swipnet.se> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 06 Jul 1997 13:48:34 CDT Cc: lars.farm@ite.mh.se In <199707061639241455966@dialup156-4-13.swipnet.se> Lars Farm wrote: > For better explanations than I can give ( and a working free GC ) see: This entire point is moot as conservative GC's which scan the data area of a process break when using distributed objects. You'd have to combine it with a reference counting scheme in order to effectively vend objects to the network. I've written Objective-C programs which deal with live data areas of 256M and greater; while they're not a majority of the code out there, a VM scanner would clearly have some ill effects on software of that type. The OpenStep system works if you follow the policy. The policy is clear and well defined. I have no problem with a vendor shipping, say, a third party garbage collector for people to use, but making GC part of OpenStep or the language proper is uncalled for. With the addition of Java to Rhapsody's OpenStep implementation, making GC part of ObjC seems silly. Dave -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
From: ftr45r@aol.com Newsgroups: comp.sys.next.programmer Subject: writers=seeking=publication Date: Sun, 6 Jul 1997 15:36:42 Organization: Miracle Net Communications, Inc. Message-ID: <5posc6$67e$5667@roadrunner.miracle.net> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit We are a New York based international literary agency with two branch offices, one of which is in* Florida. We are seeking new and> previously published authors, so please adhere to the >following-- guidelines.>>>>>>>>>>>>>>> All fiction: send brief <synopsis, first chapter, and include a self addressed, stamped envelope (SASE). All nonfiction: brief synopsis, first chapter, SASE.>>>>>>>>>>>>> Short-Stories: brief synopsis, 3 pages, SASE.>>>>>>>>>>>>>> Poetry: send 3 poems, SASE.>>>>>>>>>>>>>> Please do not send complete manuscript unless we ask for it. <<Send to: Woodside International Literary Agency>> =33-29 58 Street>>>>>>>>// =Woodside, New York>>>>>>>>// =11377>>>>>>> =Phone (main office): =718--651-8145>>>>>>>//
From: Sarah Marsden<routera@bigfoot.com> Newsgroups: comp.sys.next.programmer Subject: free report Date: 6 Jul 97 06:12:34 GMT Organization: Zen Marketing Group Message-ID: <33bf3766.0@hades.ndirect.co.uk> Hello, I have been asked to write to offer you a free report. 'The Ultimate Guide To Killer Marketing' The Report Features: The Five Golden Rules For Web Sites How To Get Your Customers To See Your Name Everyday. How To Beat Your Competition How To Take The Risk Out Of Marketing Free Software How To Make $30000 A Year On Two Hours Work A Day And Much More. Why is it free - We are a newly launched company looking to increase awareness of our name. For your free copy by email (.txt) just hit reply now and add the words 'please send' to the subject box. Send Email To routera@bigfoot.com Sarah Marsden Zen Group
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Sun, 6 Jul 1997 19:12:05 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <Ink2N5e00iVC01p0U0@andrew.cmu.edu> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <199707061639241455966@dialup156-4-13.swipnet.se> In-Reply-To: <199707061639241455966@dialup156-4-13.swipnet.se> Excerpts from netnews.comp.sys.next.programmer: 6-Jul-97 Re: [Q] OpenStep Garbage Co.. by Lars Farm@ite.mh.se >> Why would one have to use a conservative GC in the first place? [ ... ] > Your question might be interpreted to imply that conservative collectors > are bad. I have no reason to beleive they are bad and in my experience > they are not bad. Speed is comparable to or faster than other techniques > and they are safe. So my question to you is: Why would one not want to > use conservative GC? You are attempting to make a virtue out of necessity. The point of my question was to demonstrate the fact that one is required to use a conservative garbage collector with C code because non-conservative garbage collectors are likely to fail when used with most non-trivial C programs. > Perhaps you are asking why one wants GC at all? or why one would prefer > GC over reference counting or fully manual memory management? No, I wasn't asking those questions. I'm not interested in an advocacy debate over the merits of garbage collection. Like many other concepts, GC is an appropriate and efficient solution for some problem domains and an inappropriate and/or inefficient solution for others. [ ... ] >> It's because the GC cannot be sure which regions are used or not because of >> C pointer semantics, so it _must_ use a conservative strategy. > > Yes. Well, that was the point of my question above. > This can also be interpreted as if you consider this a problem, but > it is not a problem. Conservative GC works fine. Sometimes. Sometimes not. >> A Java GC does not have to be conservative because Java's pointer >> semantics guarantee that the garbage collector can determine with >> certainty whether a region is in use or not. > > The only thing this might theoretically buy you is speed and Boehm > claims that his conservative collector is faster than the Java GCs he > has encountered. Java's semantics buy you more than that-- the Java GC is guaranteed that it can always free all unused memory because (for example) it cannot be fooled by integer aliases that appear to be valid pointers. A conservative GC is therefore likely to use more space than a non-conservative GC. >>> You'd have to work hard to fool the collector. >> >> Not very hard at all-- a circularly linked list or other >> self-referential data structure is generally sufficient to prevent a >> conservative GC from ever freeing the memory. > > No. If there is no path from a root pointer to a block then it can be > collected even if there are pointers pointing at it from other > unreachable blocks including itself. The key is that there has to be a > path from the root set to the block. Not that there might be pointers > pointing at it. If those other pointers are unreachable then the block > can still be collected. > > root : free store > set : +---------------------------------------+ > : | +-------------+ +-------------+ | > +--+ : +->| |----->| | | > | |-------->| | | +-+ > +--+ : | |<-----| | > : +-------------+ +-------------+ > P A B > > If you break the link or links between root set (your program variables) > and the free store then there is no way to reach any of the blocks in > the group of blocks on the free store and they are all garbage that can > all be collected. A conservative collector can handle this. I understand how a mark-and-sweep or copying collector works. The problem comes in that unless you explicitly NULL old pointers to dynamic structures, they will continue to point to them and the GC will not free that memory. This means that such structures will stay around either forever if the pointer variable is global or automatic, or will stay around until after the call stack has recursed past the frame containing the local pointer variables. And if you're going to have to explicitly decide to NULL pointers to avoid leaking uncollectable memory, then there is no gain in convenience over something like NeXT's current reference counting scheme where you'd send a -release message. And if you use local pointer variables and are willing to let old memory stay uncollected until the call stack pops back and frees the local pointers, what's the difference in convenience between that and using -autorelease? And, of course, there is the issue of pointer references not present in the address space of the current process-- such as memory being referenced via Distributed Objects. [ ... ] > Reference counting is unable to cope with this kind of data because, > looking at figure above, there are three references to A and dropping P > wont turn that into zero. It is still referenced by two pointers and A > wont go away unless you manually make it go away. If you must do that > then why bother with reference counts at all? This is manual > deallocation. That depends entirely upon how smart your reference counting scheme is. OPENSTEP/NEXTSTEP uses a system where valid pointers to objects memory have an implicit reference count of 1, and where the system maintains a global structure that keeps track of objects which have a higher reference count number. Simply assigning pointers around does not change the reference count, so it is perfectly possible to create self-referential structures which will be autoreleased later without any manual memory management. [ ...GC advocacy & "undefined behaviour" snipped.... ] >> You are mistaken. It's perfectly valid for me to take a pointer to a >> memory region, and subtract 10000 from it if in the future, I always add >> 10000 to the pointer before dereferencing it. Of course, this is a simple >> and contrived example to demonstrate the problem. > > No, I'm not misstaken and no, your example is definitely illegal. A > pointer must point at an object otherwise it's not valid and may not be > used. For arrays a pointer must point at one of the elements of the > array or at the imaginary nth+1 element directly after the array. No > other pointers are valid. Using such pointers is illegal and invokes > "undefined behaviour". If one was using a strictly conforming implementation of the ANSI spec, I take it back-- you're right. However, it is also permissible by the ANSI spec for conforming implementations to define the results for "undefined" operations, so long as they do not change the meaning of any strictly conforming program. Therefore, "undefined behavior according to ANSI" does not mean the same thing as "illegal" (presumably in the sense that the results of the operation will not be well defined). One expectation people have of reasonable implementations is that pointers can be converted to some available integral type (generally int or long) and back to a pointer without a loss of representation. Another is that addition or subtraction on pointers produces a result which is congruent mod 2**n to the true mathematical result which would be obtained from doing the operation using unsigned arithmetic on variables of size n, where n is the number of bits required to represent the pointer. [Where the non-pointer variable is multiplied by the size of the data type being referred to by the pointer in question, of course.] My personal preference would be for the ANSI spec to adopt these expectations given that they are commonly made. Neither of the above requires the implementation to provide an integral type which is the same size in bits as a pointer, nor should they cause problems under a segmented memory architecture. Anyway, this is getting off-topic. Time to stop here.... :-) -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
Date: 6 Jul 1997 22:16:32 GMT From: clewis@ferret.ocunix.on.ca (Chris Lewis) Sender: dre32d@msn.com Message-ID: <cancel.5pod4e$47k$827@roadrunner.miracle.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5pod4e$47k$827@roadrunner.miracle.net> Control: cancel <5pod4e$47k$827@roadrunner.miracle.net> WOODSIDE spam cancelled by clewis@ferret.ocunix.on.ca Original Subject: sell/your/photos$$$$$ Total spams this type to date: 8882 Total this spam type for this user: 3625 Total this spam type for this user today: 3625 Originating site: miracle.net Complaint addresses: postmaster@miracle.net abuse@psi.com
From: news@news.tufts.edu Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5posc6$67e$5667@roadrunner.miracle.net> Date: 6 Jul 1997 22:45:24 GMT Control: cancel <5posc6$67e$5667@roadrunner.miracle.net> Message-ID: <cancel.5posc6$67e$5667@roadrunner.miracle.net> Sender: ftr45r@aol.com Spam cancelled by news@news.tufts.edu
From: emclean@slip.net (Emmett McLean) Newsgroups: comp.sys.next.programmer Subject: How to scroll from keyboard Date: 6 Jul 1997 20:56:17 -0700 Organization: Slip.Net Message-ID: <5pppd1$hhr@slip.net> Hi, Is there a way to scroll to the bottom of a terminal session without draging the scroll-bar-circle-box downward til it won't move? I sometimes use Command-F to search for text and it would be nice if I could type Command-Something to return to typing. Thanks, Emmett
From: emclean@slip.net (Emmett McLean) Newsgroups: comp.sys.next.programmer Subject: Seek equivalent to Windows Alt-Tab Date: 6 Jul 1997 20:59:16 -0700 Organization: Slip.Net Message-ID: <5pppik$hli@slip.net> Hi, Is there a NeXT equivalent to Window's Alt-Tab ? I'd like to switch applications from the command line sometimes without lifting my hand from the keyboard. Thanks, Emmett
From: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.next.programmer Subject: Re: How to scroll from keyboard Date: 7 Jul 1997 00:20:36 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5ppqqk$6td$1@sps1.phys.vt.edu> References: <5pppd1$hhr@slip.net> In article <5pppd1$hhr@slip.net>, emclean@slip.net (Emmett McLean) wrote: > Is there a way to scroll to the bottom > of a terminal session without draging > the scroll-bar-circle-box downward > til it won't move? Assuming you're using Terminal.app: Go to Preferences. Select Display from the pop-up list. In the Other Options box, check the "Scroll to the bottom of the window when input is received" checkbox. Then click the Set Default button.
From: webmaster@jpl.nasa.gov Newsgroups: comp.sys.ncr,comp.sys.newton.misc,comp.sys.newton.programmer,comp.sys.next.advocacy,comp.sys.next.bugs,comp.sys.next.hardware,comp.sys.next.marketplace,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software Subject: !Most Recent Mars Photos&Info! Date: Mon, 07 Jul 1997 01:45:28 -1200 Organization: NASA - JPL Message-ID: <33C0F2F8.5C97@jpl.nasa.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Due to the intense pressure on our servers, we have configured our software for this site; http://38.217.84.11/~nasa2/marspath.html to transfer traffic to the least busy sight available. Hopefully this will speed up access times. Remember, the pictures at these sites are updated in real-time. As soon as we get them, you get them. Once again; http://38.217.84.11/~nasa2/marspath.html Thanks for your patience, and thank you to the 100 Million visitors we've had already.
From: dix.lorenz@hamburg.netsurf.de (Dix Lorenz) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.next.sysadmin Subject: PTR: Problems and questions Date: Sun, 6 Jul 1997 21:42:36 +0200 Organization: EDV - Beratung Message-ID: <199707062142361722270@dip039-2.hamburg.netsurf.de> Hi, recently I received the Prelude to Rhapsody package and while it cleared up some questions for me, it opened up some new ones. 8 weeks ago I bought OpenStep Developer/Mach 4.1 and have been fiddling around with it, creating small apps, sometimes also using C++. No problem. What I wanted to know from PTR: How does all this work under Windows? Result: Win 3.51: Developing is just the same. Apps I wrote under Mach I could just recompile and they would run under Windows, just as promised. Cool. BUT: Everything that I tried to compile using C++ complained about a missing header file <new>. I am pretty sure this is not my problem, but a compiler problem. Question: I know PTR is a prerelease version. Is this fixed in the final release? Has someone managed to compile Objective-C++ using PTR? Win95: Mixed results: Some of the apps which got installed ran, some didn't. Mine didn't. Again: Is this because it is prerelease? In other words: how is the current release of OpenStep Developer/NT different from the one in PTR? Thanks, Dix
From: mpaque.spa-am@nospam.wco.com (Mike Paquette) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 6 Jul 1997 23:34:10 -0700 Organization: Electronics Service Unit No. 16 Sender: mpaque@mpaque Distribution: world Message-ID: <5pq2l2$11i@mpaque.mpaque> References: <199707061639281456216@dialup156-4-13.swipnet.se> In article <199707061639281456216@dialup156-4-13.swipnet.se> lars.farm@ite.mh.se (Lars Farm) writes: > Reference counting is nothing but a very poor implementation of GC. It has the advantage of working over an allocation pool spread over multiple address spaces, or processes. This discussion reminds me of the tale of the blind men and the elephant.... -- Mike Paquette (mpaque AT wco.com ; Yank that .spa-am and nospam to reply.) Well, if there *were* anything to say, it would be with the understanding that the PR/Marketing people want to make the announcements on products, so anything I have to say wouldn't actually exist until after then, so what I might have to say now doesn't exist, and what I may say in future can't be said, so theoretically what exists, doesn't, for the immediate future. (With apologies to Joe Straczynski)
From: erich@powerwareintl.com (Eric Harley) Newsgroups: comp.sys.next.programmer,comp.sys.next.software Subject: Porting NeXTStep App to OpenStep Date: Sun, 06 Jul 1997 23:23:18 -0800 Organization: EdgeMedia Networks Message-ID: <erich-0607972323190001@ppp-207-104-16-69.snrf01.pacbell.net> How easy is it to port a NeXTStep application under 3.3 to OpenStep for Mach on Intel 4.2? Considering I dont have access to a NeXTStep machine, will I be able to pull this off?
Control: cancel <33bf0e22.0@news1.ibm.net> From: behappy@iveshown.com Subject: cmsg cancel <33bf0e22.0@news1.ibm.net> Newsgroups: comp.sys.next.programmer Organization: EEI Message-ID: <cancel.33bf0e22.0@news1.ibm.net> Date: 07 Jul 97 07:19:49 GMT Article cancelled by news@dfw-ixnews1.ix.netcom.com.
From: cdr43a@aol.com Newsgroups: comp.sys.next.programmer Subject: *writers/seeking/publication Date: Mon, 7 Jul 1997 04:10:28 Organization: Miracle Net Communications, Inc. Message-ID: <5pq8hj$o9u$6271@roadrunner.miracle.net> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit We are a New York based international literary agency with two branch offices, one of which is in Florida. We are seeking new and previously published authors, so please adhere to the following guidelines. All fiction: send brief synopsis, first chapter, and include a self addressed, stamped envelope (SASE). All nonfiction: brief synopsis, first chapter, SASE. Short Stories: brief synopsis, 3 pages, SASE. Poetry: send 3 poems, SASE. Please do not send complete manuscript unless we ask for it. Send to: Woodside International Literary Agency 33-29 58 Street>>>>>>>> Woodside, New York>>>>>>>> 11377>>>>>>> Phone (main office): 718-651-8145>>>>>>>
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer,comp.sys.next.software Subject: Re: Porting NeXTStep App to OpenStep Date: 7 Jul 1997 08:19:35 GMT Organization: 21st Century Software, New York City Sender: daver@ts3-2.nj.cnct.com Message-ID: <5pq8qn$rrg$1@darla.visi.com> References: <erich-0607972323190001@ppp-207-104-16-69.snrf01.pacbell.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 07 Jul 1997 03:19:35 CDT Cc: erich@powerwareintl.com In <erich-0607972323190001@ppp-207-104-16-69.snrf01.pacbell.net> Eric Harley wrote: > How easy is it to port a NeXTStep application under 3.3 to OpenStep for > Mach on Intel 4.2? Depends on the size of the app. It's not very hard, but some things require a little different way of wrapping your brain around things. > Considering I dont have access to a NeXTStep machine, will I be able to > pull this off? Well, in this case, it's probably pretty hard. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
Newsgroups: comp.sys.next.programmer From: bofh@keltia.freenix.fr (Ollivier Robert) Sender: cdr43a@aol.com Message-ID: <cancel.5pq8hj$o9u$6271@roadrunner.miracle.net> Subject: cmsg cancel <5pq8hj$o9u$6271@roadrunner.miracle.net> Control: cancel <5pq8hj$o9u$6271@roadrunner.miracle.net> Organization: Usenet Canal Historique Date: Mon, 7 Jul 1997 04:10:28 ECP/EMP aka SPAM or pyramidal scheme (MMF) cancelled by bofh@keltia.freenix.fr. It may also be an image too small for newsbot to be activated. See report in news.admin.net-abuse.bulletins. Date: Mon Jul 7 11:40:11 1997 Original subject was: *writers/seeking/publication
Newsgroups: comp.sys.next.programmer From: bofh@keltia.freenix.fr (Ollivier Robert) Sender: wtetwutqrttirtyjr@kjpttypppstwtthy.com Message-ID: <cancel.5pofm2$4e$2@news.internetmci.com> Subject: cmsg cancel <5pofm2$4e$2@news.internetmci.com> Control: cancel <5pofm2$4e$2@news.internetmci.com> Organization: Usenet Canal Historique Date: 6 Jul 1997 16:04:18 GMT ECP/EMP aka SPAM or pyramidal scheme (MMF) cancelled by bofh@keltia.freenix.fr. It may also be an image too small for newsbot to be activated. See report in news.admin.net-abuse.bulletins. Date: Mon Jul 7 11:57:43 1997 Original subject was: YOUR VIRTUAL OFFICE CAN DO EVERYTHING!!!
From: lars.farm@ite.mh.se (Lars Farm) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Mon, 7 Jul 1997 12:39:01 +0200 Organization: pv Message-ID: <1997070712390171582@dialup177-2-12.swipnet.se> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <sams.868141660@shellx> <199707061639281456216@dialup156-4-13.swipnet.se> <5pq2l2$11i@mpaque.mpaque> Cache-Post-Path: mn8!s-49817@dialup177-2-12.swipnet.se Mike Paquette <mpaque.spa-am@nospam.wco.com> wrote: > In article <199707061639281456216@dialup156-4-13.swipnet.se> > lars.farm@ite.mh.se (Lars Farm) writes: > > Reference counting is nothing but a very poor implementation of GC. > > It has the advantage of working over an allocation pool spread over > multiple address spaces, or processes. > > This discussion reminds me of the tale of the blind men and the > elephant.... Quite possibly. I started off in the first post by saying I'm a Mac programmer and that I don't know Obj-C. I do know C++ and I do know about GC and I do know about the general technique called reference counting. So tell me what is it about Obj-C or the NeXT frameworks that makes it less suitable for GC than than C or C++? -- Lars Farm; lars.farm@ite.mh.se - Limt/Channelmatic: lars.farm@limt.se
From: Erik Doernenburg <erik@object-factory.REMOVE_ME.com> Newsgroups: comp.sys.next.programmer,comp.sys.next.software Subject: Re: CVS and .nib files Date: 7 Jul 1997 07:37:42 GMT Organization: Object Factory GmbH (Germany) Message-ID: <5pq6c6$t4r$1@leonie.object-factory.com> References: <5p0g4q$dlh@fido.transit.ru> <5p0sdu$nea@iserver.stem.com> <33ba574c.0@epflnews.epfl.ch> <5pfda8$3hg$1@hermes.is.co.za> <5pi8rc$a6g@news-rocq.inria.fr> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit fpottier@pauillac.inria.fr (Francois Pottier) wrote: > In article <5pfda8$3hg$1@hermes.is.co.za>, <gvandyk@icon.co.za> wrote: > >On 07/02/97, Vincent Kohler wrote: > > >The only problem I had with wrappers is that I couldn't find the > >"wrap" and "unwrap" Utilities and I wrote my own. Was this the right > >thing to do or did I miss something somewhere > > My distribution of cvs didn't have wrap and unwrap either (it's a cvs > 1.9 which I found on the RedHat Linux CD). I simply use tar instead > and everything seems to work fine. My colleague who updated our installation to 1.9 said the same. When I installed 1.7, however, I found the two utilities in the sample cvsroot. Since they are not too long I'll include them below. There's another problem with wrappers though: When you use the cvs server, maybe because you cannot export the cvsroot via NFS, they don't work! The directories will be treated like files and all you get is junk. This is a known problem (at least since 1.7) and eventhough some people seem to be working on it I've never heard of a fix. enjoy erik -------8<--------------------------------------------------------------- #! /bin/sh # # wrap - Combine a directory into a single tar package. # #ident "@(#)cvs/examples:$Name: $:$Id: wrap,v 1.1 1996/01/17 14:38:58 erik Exp $" # This script is always called with the current directory set to # where the file to be combined exists. but i may get called with a # path to where cvs first started executing. (this probably should be # fixed in cvs) so strip out all of the directory information. The # first sed expression will only work if the path has a leading / # if it doesn't the one in the if statement will work. DIRNAME=`echo $1 | sed -e "s|/.*/||g"` if [ ! -d $DIRNAME ] ; then DIRNAME=`echo $1 | sed -e "s|.*/||g"` fi # # Now tar up the directory but we now will only get a relative path # even if the user did a cvs commit . at the top. # gnutar --preserve --sparse -cf - $DIRNAME | gzip --no-name --best -c > $2 -------8<--------------------------------------------------------------- -------8<--------------------------------------------------------------- #! /bin/sh # # unwrap - extract the combined package (created with wrap) # #ident "@(#)cvs/examples:$Name: $:$Id: unwrap,v 1.1 1996/01/17 14:38:59 erik Exp $" # move the file to a new name with an extension rm -rf $1.cvswrap mv $1 $1.cvswrap # untar the file if `gzip -t $1.cvswrap > /dev/null 2>&1` then gzcat -d $1.cvswrap | gnutar --preserve --sparse -x -f - else gnutar --preserve --sparse -x -f $1.cvswrap fi # remove the original rm -rf $1.cvswrap -------8<--------------------------------------------------------------- -- Erik Dörnenburg -- OBJECT FACTORY -- Gesellschaft für Informatik und Datenverarbeitung mbH -- http://www.object-factory.com/~erik
From: rog@ohm.york.ac.uk (Roger Peppe) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 7 Jul 1997 11:23:44 GMT Organization: Department of Electronics, University of York, UK. Sender: rp9@york.ac.uk Message-ID: <5pqjk0$ble$1@netty.york.ac.uk> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$ <19970705124505517300@dialup76-3-36.swipnet.se> On Sat, 5 Jul 1997 12:45:05 +0200, Lars Farm <lars.farm@ite.mh.se> wrote: > Many claim that reference counting is as good as GC and is enough, but > this is not the case. Reference counting wont solve cyclic references > and is not automatic. If you must manually increment/decrement a counter > then where is the difference from saying new/delete? It's a very minor > gain compared to proper garbage collection. at least one system does automatic reference counted garbage collection, while retaining the ability to create self referential structures. it's called Inferno (see http://www.lucent.com/inferno/). (sadly, it doesn't run under openstep) so it is possible - and once you've seen such an animal, other creature in the garbage collection zoo seem inadequate. most data structures are *not* self-referential, and it seems silly to pay the price, in unpredictable time and memory overheads, for the few that are. the openstep implementation of reference counting is however a complete mess; allocation and deallocation primitives should be the simplest and best understood primitives in the language. instead, as is evident from the questions and sometimes hesitant answers seen in this thread, the openstep scheme has created a half baked mish mash of unclear obligations and untrusted promises that makes it that much harder to produce reliable, robust code. cheers, rog.
From: wegmann@talisker.linguistics.ruhr-uni-bochum.de (Frank Wegmann) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.programmer Subject: Re: Sun Sparc TOO! (was Re: Prelude2Rhapsody on black hardware!) Followup-To: comp.sys.mac.advocacy,comp.sys.next.programmer Date: 07 Jul 1997 15:39:48 GMT Organization: Ruhr-Universitaet Bochum, Rechenzentrum Message-ID: <WEGMANN.97Jul7143948@talisker.linguistics.ruhr-uni-bochum.de> References: <rbarris-ya023280001106971149270001@news.intelenet.com> <5o3mv4$rh7$1@news.platinum.com> <5p11qf$ar2$2@nuke.csu.net> In-reply-to: thomas@zippy.sonoma.edu's message of 27 Jun 1997 18:46:39 GMT In article <5p11qf$ar2$2@nuke.csu.net> thomas@zippy.sonoma.edu (Thomas Poff) writes: > Prelude installed on a friend's Sparc trivially. No muss no fuss. > It was a trivial installation. Well, almost. You cannot boot from the PTR CD (at least on my SS20), so you need a bootable NS3.3/OS4.x CD-ROM. Then change disks at the boot prompt and continue. From that point, installation is a no-brainer. Frank -- Frank Wegmann voice: +49 234 700 7677 / +49 234 700 2461 Sprachwiss. Institut fax : +49 234 7094 137 Ruhr-Universitaet Bochum email: wegmann@linguistics.ruhr-uni-bochum.de, 44780 Bochum wegmann@acm.org (NeXTmail, MIME welcome) Germany WWW : http://www.linguistics.ruhr-uni-bochum.de
From: wegmann@talisker.linguistics.ruhr-uni-bochum.de (Frank Wegmann) Newsgroups: comp.sys.next.programmer Subject: URGENT: How To Access (SCSI) CD-ROM drives under OS/NT? Date: 07 Jul 1997 15:53:25 GMT Organization: Ruhr-Universitaet Bochum, Rechenzentrum Distribution: world Message-ID: <WEGMANN.97Jul7145325@talisker.linguistics.ruhr-uni-bochum.de> We are currently porting an application from NEXTSTEP to OPENSTEP, leading to the usual bunch of problems. We now got most of it running under OSNT, even deploying under Windows 95. However, one part is (mission) critical: we need to access a CD-ROM drive (for audio CDs) from the Windows application. We used SCSI ioctls under NEXTSTEP, which happen to work under OS/Mach exactly the same way. But how can we use ioctls under OS/NT? We found header files such as winioctl.h, but without a hint to the appropriate library function. Do we need another dev environment such as VC++? How can we link our OPENSTEP application against Windows DLLs? If anybody has already any experience in that field, this would definitely help us a lot, since we *must* demo the app under Windows RSN. TIA, Frank Wegmann -- Frank Wegmann voice: +49 234 700 7677 / +49 234 700 2461 Sprachwiss. Institut fax : +49 234 7094 137 Ruhr-Universitaet Bochum email: wegmann@linguistics.ruhr-uni-bochum.de, 44780 Bochum wegmann@acm.org (NeXTmail, MIME welcome) Germany WWW : http://www.linguistics.ruhr-uni-bochum.de
From: Iracly <iracly@rtr.transit.ru> Newsgroups: comp.sys.next.programmer Subject: How to split classes hierarhy to bundles Date: Mon, 07 Jul 1997 18:10:42 +0400 Organization: Radio-MSU NOC, Moscow State University Message-ID: <33C0F8E2.273C@rtr.transit.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Is anybody have idea how to divide class hierarhy to different loadable bundles. Under Windows I can just declare class as __export/__import and then link any other DLL (even with derived classes) with import library. Is some analogues under OPENSTEP? When I just try to compile "derived bundle" I got UNRESOLVED EXTERNAL ... message from linker, If I make some dummy implementation of base class in derived bundle, than after I load derived bundle, the original base class was replased with dummy implementation from second one! PS I use OPENSTEP developer v4.2 prerelease 2 for NT Regards Iracly (iracly@com2com.ru or iracly@rat.radio-msu.net)
From: William.Clocksin@CL.cam.ac.uk Newsgroups: comp.sys.next.sysadmin,comp.sys.next.programmer Subject: How to set up Renderman printing? Date: 7 Jul 1997 14:31:00 GMT Organization: University of Cambridge, England Message-ID: <5pquj4$p0v@lyra.csx.cam.ac.uk> Every time I use a RenderMan-based plotting app, the data displays nicely on the screen. But whenever I try to print it, I get a panel asking for a renderer host (which the only choice is localhost), and then error panel complaining that it can't connect to localhost. What should I do to get this working? Thanks. William Clocksin wfc@CL.cam.ac.uk
Newsgroups: comp.sys.next.programmer From: bofh@keltia.freenix.fr (Ollivier Robert) Sender: Sarah Marsden<routera@bigfoot.com> Message-ID: <cancel.33bf3766.0@hades.ndirect.co.uk> Subject: cmsg cancel <33bf3766.0@hades.ndirect.co.uk> Control: cancel <33bf3766.0@hades.ndirect.co.uk> Organization: Usenet Canal Historique Date: 6 Jul 97 06:12:34 GMT ECP/EMP aka SPAM or pyramidal scheme (MMF) cancelled by bofh@keltia.freenix.fr. It may also be an image too small for newsbot to be activated. See report in news.admin.net-abuse.bulletins. Date: Mon Jul 7 16:22:31 1997 Original subject was: free report
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: How to scroll from keyboard Date: Mon, 7 Jul 1997 10:27:39 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <onkDnP200iWR02OSg0@andrew.cmu.edu> References: <5pppd1$hhr@slip.net> In-Reply-To: <5pppd1$hhr@slip.net> Excerpts from netnews.comp.sys.next.programmer: 6-Jul-97 How to scroll from keyboard by Emmett McLean@slip.net > Is there a way to scroll to the bottom > of a terminal session without draging > the scroll-bar-circle-box downward > til it won't move? Yes. If you look in "Info->Preferences/Display" and enable the "Scroll to the botton of the window...", and you need to do is start typing. > Is there a NeXT equivalent to Window's Alt-Tab ? Sort of. Command-up/down arrow will raise and lower windows from the keyboard. This does not change the key window, although I believe you can do a simple hack to the WindowServer PostScript code to get that behavior as well. By the way, these questions probably don't belong in .programmer; .misc would have been better. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Patrick Schulz <pschulz@symphony.amd.com> Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: Mon, 07 Jul 1997 10:05:31 -0500 Organization: AMD Fab30 Message-ID: <33C105BB.7345@symphony.amd.com> References: <5plkqs$3o9$1@news2.digex.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi out there, > Erik M. Buck wrote: [...] > > Art Isbell wrote: [...] > > dyoung@vvi.com wrote: [...] you already discussed a bunch of things I just want to add another topic. I don't know the 4.2 released version of PB, but I'm afraid it's still missing the great variable browser for gdb that PB3.3 (actually it was an Edit bundle) had. I really wonder why there are that less complaints about it. This browser was THE reason for me to do almost all my serious debug-work on my NeXT system. If you have ever worked with large structures (trees/woods) you know what a hassle debugging might be, if you can't dereference your pointers very easily. OK, ups can do that for X11 but you waste the whole screen, same for ddd... But we're talking about OPENSTEP and there's no substitute at all. So, please guys in the PB-team, I can live without a stack view, but give me back that neat variable browser ! Patrick. -- Patrick Schulz; 1704 Nelms Dr. #2025; Austin, TX 78744 email: pschulz@symphony.amd.com (MIME welcome) - vmunix: panic - no coffee detected, user halted.
Newsgroups: comp.sys.next.programmer From: "Eric K. Ringger" <ringger@cs.rochester.edu> Subject: Re: Looking for an expert system shell In-Reply-To: Your message of "02 Jul 1997 14:49:50 GMT." <33ba6a8e.0@epflnews.epfl.ch> Message-ID: <199707071603.MAA13005@micro.cs.rochester.edu> Sender: ringger@cs.rochester.edu (Eric K. Ringger) Cc: comp.sys.next.programmer Organization: University of Rochester Computer Science Dept Date: Mon, 07 Jul 1997 12:03:29 -0400 In comp.sys.next.programmer, Marco Scheurer wrote: >I'm looking for an expert system shell I could easily use with >NEXTSTEP 3.3. More specifically, I am looking for a forward chaining >inference engine. > >Any reference to available systems, source code, commercial or not >would be greatly appreciated. [...] CLIPS might meet your requirements. I suppose you'll need to search for the distribution, since I don't have a URL on hand. I seem to recall that one of the NASA labs developed it, but I could be wrong. Good Luck. --Eric --- Eric K. Ringger mailto:ringger@cs.rochester.edu Dept. of Computer Science Office: +1-716-275-0922; Lab: +1-716-275-1083 University of Rochester Fax: +1-716-461-2018 Rochester, NY 14627-0226 http://www.cs.rochester.edu/u/ringger/ ||||| | | | | | | | | | | | | | | | | |||||
From: nomoreshit@fromnetcops.com Newsgroups: comp.sys.next.programmer Subject: ENLOW WAR ON THE NET Date: 7 Jul 1997 07:41:25 GMT Organization: IMT Message-ID: <5pq6j5$d962305@odin.telapex.com> Angered Spy/Research Expert and now, Leading Business Consultant, tells "the rest of the story." What the news & Wired forgot to tell... Just search the Web - we're now the largest chain of WEB SITES on the net for supporting entrepreneurs and businesses wishing to grow using his techniques. He gave them fair warning... But they didn't listen... CH http://michaelenlow.by.net/spamwar.html
From: lars.farm@ite.mh.se (Lars Farm) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Mon, 7 Jul 1997 18:48:06 +0200 Organization: pv Message-ID: <1997070718480667602@dialup103-4-10.swipnet.se> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <199707061639241455966@dialup156-4-13.swipnet.se> <Ink2N5e00iVC01p0U0@andrew.cmu.edu> NNTP-Posting-User: s-49817 Charles William Swiger <cs4w+@andrew.cmu.edu> wrote: > Java's semantics buy you more than that-- the Java GC is guaranteed that > it can always free all unused memory because (for example) it cannot be > fooled by integer aliases that appear to be valid pointers. A > conservative GC is therefore likely to use more space than a > non-conservative GC. A copying collector needs a heapsize of at least twice the allocated size for copying. A conservative collector can make do with less. There are techniques for conservative collectors to deal with false pointers, so false pointers are more a theoretical problem than a practical problem. It's not zero, but not a real problem either. As for speed, a copying collector must touch every allocated byte when copying, but a conservative collector needs only touch pointers to allocated blocks. Well written conservative collectors (like Boehms) holds their own against copying collectors both on memory usage and speed. It is true that they could become even better with language support, but they are still very good without. Saying GC is ok, but not if it is conservative GC simply doesn't make sense to me and I think that is what you're saying. > The problem comes in that unless you explicitly NULL old pointers to > dynamic structures, they will continue to point to them and the GC will > not free that memory. This means that such structures will stay around > either forever if the pointer variable is global or automatic, or will > stay around until after the call stack has recursed past the frame > containing the local pointer variables. So don't use global variables;-) This "problem" is the same with any kind of GC. It would be the same in Lisp, Smalltalk, Java or Eiffel regardless of how the GC was implemented. It is also the same for reference counting. What does that have to do with conservative GC beeing an inferior kind of GC and therefor not worthy of using? Does this mean that you're arguing against GC in general, not against conservative GC? > And if you're going to have to explicitly decide to NULL pointers to > avoid leaking uncollectable memory, then there is no gain in convenience > over something like NeXT's current reference counting scheme where you'd > send a -release message. ... or say free() in C, delete in C++ or dispose in Pascal. What is your point? > And if you use local pointer variables and are willing to let old memory > stay uncollected until the call stack pops back and frees the local > pointers, what's the difference in convenience between that and using > -autorelease? I don't have to say -autorelease and I don't have to know *when* to say aoutorelease and I don't have to know *which* of all the pointers refering to an object to say it to and I don't have to worry about dangling pointers. (Obviously I don't know what -autorelease is I just treated it as some kind of free() / delete / decr_counter() or such - perhaps it is much more capable than I assume - feel free to educate me:-) > And, of course, there is the issue of pointer references not present in > the address space of the current process-- such as memory being > referenced via Distributed Objects. Yes, that is a problem. There are always lots of problems to solve when writing a program. Only one of those problems are addressed by GC. Your argument is a variation of one of the standard arguments against GC, not against conservative GC in particular: - "If it doesn't solve programming problem X as well as reusing unused memory it's no good". Well, it doesn't. Garbage refers to dynamically allocated RAM, not to arbitrary external resource. I think a "Distributed Object" falls in the second category, as does open files, mutexes, semaphors and such. One nice feature of GC is that it is very single minded. It solves one problem and one problem only. A very nasty problem. It solves it almost completely and it has no effect on anything else (unless you want to). GC returns unused memory. Nothing more. GC is no silver bullet. It's merely a very useful tool. This feature, to solve one single problem and do it almost perfectly, is in every other aspect of Software Engineering considered a Very Good Thing. High coherence. GC is often combined with 'finalization' and finalization can be used to release other external resourses. Without pretending to know anything about NeXT distributed objects I assume finalization could be used for them. Even if not, that is a different problem that has little to do with reusing unused RAM. > Simply assigning pointers around does not change the reference count, so > it is perfectly possible to create self-referential structures which > will be autoreleased later without any manual memory management. Then it isn't reference counting. It doesn't count references. Perhaps only "important" references and semi-manual counting or perhaps even manual counting? Doesn't that expose one to the usual pointer related problems with dangling pointers, etc, etc? > However, it is also permissible by the > ANSI spec for conforming implementations to define the results for > "undefined" operations, so long as they do not change the meaning of any > strictly conforming program. Therefore, "undefined behavior according > to ANSI" does not mean the same thing as "illegal" (presumably in the > sense that the results of the operation will not be well defined). More than anything it means "non portable". Yes, vendors may define whatever they want. This doesn't turn their private definitions for release X.Y into C or C++. Compiler vendors have been known to change their mind for the next version. Try to find a pair of vendors that agree on such private definitions;-) Rhapsody is supposed to actively support portablility to several processors and OSs so I still think portability ought to be even more important than otherwise. - Lars -- Lars Farm; lars.farm@ite.mh.se - Limt/Channelmatic: lars.farm@limt.se
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: POSIX, was Re: [Q] OpenStep Garbage Collection Date: Mon, 7 Jul 1997 13:20:29 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Distribution: world Message-ID: <cnkGJRO00iWR02OKM0@andrew.cmu.edu> References: <199707061639281456216@dialup156-4-13.swipnet.se> <5pq2l2$11i@mpaque.mpaque> In-Reply-To: <5pq2l2$11i@mpaque.mpaque> Excerpts from netnews.comp.sys.next.programmer: 6-Jul-97 Re: [Q] OpenStep Garbage Co.. by Mike Paquette@nospam.wco > This discussion reminds me of the tale of the blind men and the > elephant.... Ouch! Since I'd prefer to avoid further comparision along those lines, clearly it's time to change the subject: Will OPENSTEP or Rhapsody ever have POSIX.1 support? -Chuck PS: Lest anyone think that I did not first try to answer this for myself, let me note for the record that a web search for the string "POSIX" via <URL=http://www.next.com/Search/Welcome.html> results in a grand total of 6 hits, after also enabling a search through the "Apple Computer" database. #1-3: NeXTanswers 1125, 2066, & 2068 make the claim that POSIX support existed circa NEXTSTEP 3.1 to 3.3p1. #4: NA-1893 claims that one can compile Satan with "cc -posix" if one were to use the putenv.c file that NeXT provided in lieu of supplying putenv() with the standard system libraries. #5: There was the 8/17/94 press release that the HP_PA version of NEXTSTEP had shipped. A pity running into that, considering that I know people who acquired HP 712 Geckos with the hopes of not using HP/UX, and I also know of a few comparies who paid a goodly amount of money to purchase the high-end HP 735-xxx series machines in the hopes of using them as compile servers for NEXTSTEP. Finally, #6: the search also revealed that in NA-2501 about the 4.2 release of Foundation, someone mentioned POSIX with regard to the topic of "Avoid[ing] two-digit years and all uses of the "%y" date format specifier." But, other than that hopeful indication that there exists at least one person at NeXT/Apple who still remembers what POSIX is, there's _nada_. Well, in light of my recent chastisement elegantly (and perhaps aptly) implied above by Mike Paquette, maybe I shouldn't criticise. Still, this was an unfortunate series of documents to come across. While putenv(), mkfifo(), setsid(), POSIX signal handling, and the like may not seem that important, I must say that I too am reminded of a certain tale. Something along the lines of: "For want of a nail, the kingdom was lost...." Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Greg_Anderson@afs.com (Gregory H. Anderson) Newsgroups: comp.sys.next.programmer Subject: Re: *.wn -> .rtf ? Date: 7 Jul 1997 13:49:18 GMT Organization: Anderson Financial Systems Inc. Message-ID: <5pqs4u$4t7@shelob.afs.com> References: <5pn159$pst$2@bashir.peak.org> Timothy J. Luoma writes > In <33BC3E3C.49F5@dartmouth.edu> kevin parks wrote: > > Is there any way to convert all my WriteNow > > files to rtf in batch. instead of opening each individual > > file and doing a save as. > > There's a wn2rtf binary at > ftp://next-ftp.peak.org/pub/next/apps/utils/converters/wn2rtf.1.0.N.b.gz > I am not sure how well it deals with anything which is not text. It doesn't. It also has several known bugs, especially in line spacing commands. We rewrote a portion of it several years ago, adding support for files that contain graphics (turns them into RTFDs) and fixing the most obvious bugs. This version is contained in the document filters that ship with WriteUp and PasteUp. -- Gregory H. Anderson | "We're in the land of the blind, Visionary Ophthalmologist | selling working eyeballs, and they Anderson Financial Systems | balk at the choice of color." -- Tony greg@afs.com (NeXTmail OK) | Lovell, on Mac user reactions to NeXT
From: Greg_Anderson@afs.com (Gregory H. Anderson) Newsgroups: comp.sys.next.programmer Subject: Re: Porting NeXTStep App to OpenStep Date: 7 Jul 1997 13:53:24 GMT Organization: Anderson Financial Systems Inc. Message-ID: <5pqsck$4tc@shelob.afs.com> References: <erich-0607972323190001@ppp-207-104-16-69.snrf01.pacbell.net> Eric Harley writes > How easy is it to port a NeXTStep application under 3.3 to OpenStep for > Mach on Intel 4.2? Depends on the application. If it's an MCCA "corporate app" built around the AppKit and custom classes, shouldn't be too bad. If it's built with an older version of EOF, there may be substantial rewriting. > Considering I dont have access to a NeXTStep machine, will I be able to > pull this off? Sure. 4.2 comes with a set of scripts that perform much -- but certainly not all -- of the translations. The 3.3 docs are included in 4.x, so you can always look up the old methods and functions to see how they used to work. And the old Intel binary should still run fine, so you can observe its behavior. -- Gregory H. Anderson | "We're in the land of the blind, Visionary Ophthalmologist | selling working eyeballs, and they Anderson Financial Systems | balk at the choice of color." -- Tony greg@afs.com (NeXTmail OK) | Lovell, on Mac user reactions to NeXT
From: planetary <kris@xmission.xmission.com> Newsgroups: comp.sys.next.programmer Subject: Re: Seek equivalent to Windows Alt-Tab Date: 7 Jul 1997 13:55:23 -0600 Organization: XMission Internet (801 539 0900) Message-ID: <5prhjb$kb3@xmission.xmission.com> References: <5pppik$hli@slip.net> Emmett McLean <emclean@slip.net> wrote: : Is there a NeXT equivalent to Window's Alt-Tab ? Cmd-[Up arrow|Down arrow] .......kris -- Kristopher Magnusson kris@xmission.com (no NeXTmail, please) --------------------------------------------------------------------------- Contains freshness saver packet. DO NOT EAT. nguage. instead, as is evident from > the questions and sometimes hesitant answers seen in this thread, the > openstep scheme has created a half baked mish mash of unclear > obligations and untrusted promises that makes it that much harder to > produce reliable, robust code. Well, I don't know if its that bad! Since I was the one who started this thread, I've learned two things that have made me feel both good and bad... Since it does seem a little unclear to people of how GC works on OpenStep, or at least of the proper use of it, I don't feel so bad for not fully understanding myself. However, the fact that a simple question about GC has turned into such a long thread should raise a flag somewhere at Apple. But what bothers me is the inconsistancy I see in the API. From my perspective, when I'm creating objects, sometimes I need to retain them and sometimes I don't. When I alloc/init an object, its ref count is automatically incremented for me. When I use the dictionary class method +dictionary to creat a dictionary, I have to explicitly retain it to keep it. Also, I will have to release objects that I didn't explicitly retain myself. This seems inconsistant to me, and what I was looking for in my original post was an explanation of why it is this way. For instance, I think that whenever you create an object, whether its by copy, alloc, or special class methods, and you want to keep it around outside the scope of the method its created in, then you should have retain it. Then, its a simple matter of always matching a retain with a release. Its seems simple and consistant -- so there's probably a reason why its not this way! What is that reason? -- |-------------------------------------------------------------------------| | Mark Trombino | J A M S o f t | | mtrombin@ix.netcom.com | Audio DSP Tools for Openstep & Rhapsody | |-------------------------------------------------------------------------|
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 7 Jul 1997 20:08:55 GMT Organization: 21st Century Software, New York City Sender: daver@ts3-2.nj.cnct.com Message-ID: <5pricn$i0l$1@darla.visi.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <199707061639241455966@dialup156-4-13.swipnet.se> <5popa2$sfp$1@darla.visi.com> <1997070718480867755@dialup103-4-10.swipnet.se> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 07 Jul 1997 15:08:55 CDT Cc: lars.farm@ite.mh.se In <1997070718480867755@dialup103-4-10.swipnet.se> Lars Farm wrote: > David Young <dwy@ace.net> wrote: > > The OpenStep system works if you follow the policy. The policy is clear and > > well defined. I have no problem with a vendor shipping, say, a third party > > garbage collector for people to use, but making GC part of OpenStep or the > > language proper is uncalled for. With the addition of Java to Rhapsody's > > OpenStep implementation, making GC part of ObjC seems silly. > > How come GC is acceptable for OpenStep in Java but GC is not acceptable > for OpenStep in other languages? Because GC is part of the Java language specification, of course. Programmers programming in Java don't expect to deal with memory management and would undoubtedly goof up royally when confronted with *GASP* having to *THINK* about cleaning up their objects. Java has "new" but no "delete" (or release). Interesting, isn't it? I don't work for Apple, so I obviously have no idea how they're going to (or already have) reconcile(d) Java GC bridged to ObjC for OPENSTEP, and Java GC distributed across processes and networks for (P)DO. Recalling the discussion of Apple's changes to the ObjC language in comp.lang.objective-c, I'm actually starting to get a bit worried. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 7 Jul 1997 20:42:40 GMT Organization: 21st Century Software, New York City Sender: daver@ts3-2.nj.cnct.com Message-ID: <5prkc0$i0l$2@darla.visi.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$ <19970705124505517300@dialup76-3-36.swipnet.se> <5pqjk0$ble$1@netty.york.ac.uk> <5prfv6$ck3@sjx-ixn4.ix.netcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 07 Jul 1997 15:42:40 CDT Cc: mtrombin@ix.netcom.com In <5prfv6$ck3@sjx-ixn4.ix.netcom.com> Mark Trombino wrote: > But what bothers me is the inconsistancy I see in the API. From my > perspective, when I'm creating objects, sometimes I need to retain them and > sometimes I don't. When I alloc/init an object, its ref count is > automatically incremented for me. When I use the dictionary class method > +dictionary to creat a dictionary, I have to explicitly retain it to keep it. alloc init sets the retainCount to 1. This is where the retainCount always starts. + dictionary and +array and +string are convenience methods. Generally, class methods which return an object "in one shot" (like arrayWithObject and such) are convenience methods. I can't think of any exceptions offhand, but Read The Manual(TM). Convenience methods create an object _just like alloc init_ but also adds the object to the current autorelease pool. So, in this case, the retainCount also starts at 1, but when the current autorelease pool is freed, the object will get a -release, which will make it hit 0. All the convenience methods work this way. It's really not inconsistent. > Also, I will have to release objects that I didn't explicitly retain myself. > This seems inconsistant to me, and what I was looking for in my original > post was an explanation of why it is this way. Yes, you do. alloc init makes retainCount 1; copy also makes retainCount 1. Objects created from these methods do need to be either explicitly released or autoreleased. Think of copy as just being alloc initWithThisObject. This was all pretty confusing to me when I first started. Once you get the hang of it, it really does make sense. I suggest you see /NextLibrary/Frameworks/Foundation.framework/Resources/English.lproj/Documentation/Reference/Classes/NSAutoreleasePool.rtf for more information. > For instance, I think that whenever you create an object, whether its by > copy, alloc, or special class methods, and you want to keep it around outside > the scope of the method its created in, then you should have retain it. > Then, its a simple matter of always matching a retain with a release. Its > seems simple and consistant -- so there's probably a reason why its not this > way! That's not how it is. See above. I hope this clears up some of your confusion. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
Control: cancel <5pq6j5$d962305@odin.telapex.com> From: nomoreshit@fromnetcops.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5pq6j5$d962305@odin.telapex.com> Date: 07 Jul 97 20:38:59 GMT Organization: IMT Message-ID: <cancel.5pq6j5$d962305@odin.telapex.com> Article cancelled by news@dfw-ixnews1.ix.netcom.com.
From: holger@object-factory.REMOVETHIS.com (Holger Hoffstaette) Newsgroups: comp.sys.next.programmer Subject: Re: CVSBundle but no source ? Date: 7 Jul 1997 09:12:25 GMT Organization: Object Factory GmbH (Germany) Message-ID: <5pqbtp$tc$1@leonie.object-factory.com> References: <5pjbrm$7n6$1@concorde.ctp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Thomas Engel wrote: > PB 4.2 is really cool and works nicely, it's cool to have SCM support and all > these funky headers in the PB.framework. > > BUT the CVS bundle gives me a hard time. It messes up the Project and slows > development down so much that I almost dislike using it as much as I like the > idea of it. > > Is there a reason why this bundle can not be supplyed in source code ? Some > small bugs could be fixed so easily and other useful features added so easily > (like timed automatic status updates etc.) > Wouldn't it be a good candidate for the Develper/Examples section ? > > Why not put in on NeXTasnwers orinto the MinExamples ? while (1) { [self shout: @"YES please!" loudly: YES]; } Holger -- holger"at"object-factory.com (NeXTMail, MIME) 'Where I come from it's important to develop stuff that doesn't suck.' -- David Young <dwy"at"ace.net>
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 7 Jul 1997 22:26:37 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5prqet$n8l$1@brachio.zrz.TU-Berlin.DE> References: <5prfv6$ck3@sjx-ixn4.ix.netcom.com> In article <5prfv6$ck3@sjx-ixn4.ix.netcom.com> mtrombin@ix.netcom.com (Mark Trombino) writes: > But what bothers me is the inconsistancy I see in the API. A 'live' object has a reference count >= 1. An object is released when its reference count drops to 0. Therefore, when an object is allocated ( +alloc ) its reference count has to be 1. Conceptually, it is also pretty easy: if you explictly _create_ an object, with +alloc, you are the owner of the object, and therefore responsible for releasing it. If you just get a reference to an object by calling some method, you have to retain the object. This includes the +dictionary, +array, +someObject convenience method. It helps to think of all the convenience method as creating temporary objects. The methods that's a bit strange is -copy. I _think_ that this is like +alloc because probably 99% of users would want a retain here anyhow, so autoreleasing would be a waste. But that's just me guessing. (One more guess: copy gets used a lot in EOF, which is what got us this behaviour, as well as the immutable/mutable distinction that allows -copy to be implemented as -retain for immutable objects). Marcel
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 8 Jul 1997 02:25:51 GMT Organization: WolfWare Message-ID: <5ps8ff$4ju$1@vader.wolfware.ipc.net> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <199707061639241455966@dialup156-4-13.swipnet.se> <5popa2$sfp$1@darla.visi.com> <1997070718480867755@dialup103-4-10.swipnet.se> <5pricn$i0l$1@darla.visi.com> In-Reply-To: <5pricn$i0l$1@darla.visi.com> On 07/07/97, David Young wrote:>I don't work for Apple, so I obviously have no idea how they're going to (or >already have) reconcile(d) Java GC bridged to ObjC for OPENSTEP, and Java GC >distributed across processes and networks for (P)DO. Recalling the discussion >of Apple's changes to the ObjC language in comp.lang.objective-c, I'm >actually starting to get a bit worried. What they said at WWDC... (working from memory) Conceptually you have two run-times working in parallel - the Obj-C runtime and the Java runtime - with a bridge between them. When an object gets vended from the Obc-C side to the Java side it gets "retained" by the Obj-C run-time. When the Java GV cleans up the object on the Java side it gets "released" on the Obj-C side. When an object gets vended from the Java side to the OpenStep side an EXTRA reference to that object is created by the Java run-time. This extra reference on the Java side does not go away until the object is dealloced on the Obj-C side. This scheme assures that as long as an object is valid in either side it will not be freed by either side. -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: Alan Hutson <ahutson@stat.ufl.edu> Newsgroups: comp.sys.next.programmer Subject: biostatistician/systems admin opening Date: Mon, 07 Jul 1997 14:12:13 -0400 Organization: University of Florida, Dept. of Statistics Message-ID: <33C1317D.33FD@stat.ufl.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit CC: ahutson@stat.ufl.edu The General Clinical Research Center (GCRC) at the University of Florida is looking for a Statistical Research Coordinator to handle the day-to-day operations of the GCRC's Data Services Laboratory (DSL). The Statistical Research Coordinator fills multiple roles in assisting our researchers and Center staff. He or she will serve as a systems manager, data manager, and biostatistician. The ideal applicant will have a degree in statistics or its equivalent (preferably MA/MS) with a very strong UNIX computing background and have experience with statistical software packages such as SAS or S+. The applicant will participate in clinical trial design, carry out power and sample size calculations, and perform statistical analyses. The applicant will have the following system responsibilities: planning of future systems, integration of new systems with existing systems, which include Solaris and NEXTSTEP UNIX, a Mac, and Windows PCs, upgrading the operating systems on all machines and maintaining networked services to all machines including the email system, faxes, printing, etc. Technical understanding of services such as http, CGI-bin, telnet, ftp, TCP/IP, NFS, CAP, NIS (yellow pages), NTP, automount, rdist, sendmail, POP, IMAP, and NetInfo is a plus. Excellent communication and organizational skills are a must. The salary range for this job is $31,840 - 57,300 depending upon experience. For more information regarding the University of Florida's GCRC, see http://www.gcrc.ufl.edu or contact Professor Alan Hutson at ahutson@stat.ufl.edu. Please send a letter of application and resume to: Diana Stetter Box 115002 Gainesville, FL 32611-5002 (352) 392-4621 AA/EA/EEO. If accommodation is needed please call (352) 392-4621 or TDD (352) 392-7734.
From: penrose@w09.sfc.keio.ac.jp Newsgroups: comp.sys.next.programmer,comp.sys.next.misc Subject: NSThread severely degrades event queue access Date: 08 Jul 1997 11:32:11 +0900 Organization: Keio University Shonan Fujisawa Campus, Fujisawa Japan Message-ID: <wzk9j2e0k4.fsf@w09.sfc.keio.ac.jp> Howdy Folks: I am developing an application, under Openstep 4.1 for intel, which is multithreaded. In my application, I have a single solitary method that is run in a seperate thread. It is spawned with: [NSThread detachNewThreadSelector:@selector(mySelector:) toTarget:self withObject:sender]; In this method (as referenced by mySelector) there are no AppKit objects manipulated, nor are any messages passed to any AppKit objects. Elsewhere in the application, in a custom view, mouseDown: is overridden to implement the dragging of another view. When this view is dragged on the screen before the application becomes mult-threaded (while [NSThread isMultiThreaded] returns NO) the dragging is smooth and continuous. If I execute mySelector: in the same thread as the application, the dragging performance is also smooth. However, as soon as the application becomes multithreaded, e.g. [NSThread isMultiThreaded] returns YES), the dragging behavior degrades drastically: the view stutters when dragged. A huge (sometimes as humungous as 500ms) latency is introduced. This latency remains in the application ever after, even though the detached thread has exitted. Why is this happening? Are there any ways to alleviate this unacceptable latency? I am tempted to use cthread_fork() to see if I can fool the AppKit into better performance, but I'd prefer this application to be Openstep compliant. Thanks for any help! Christopher Penrose penrose@sfc.keio.ac.jp penrose@silvertone.princeton.edu
From: Timothy Luoma <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: Re: Seek equivalent to Windows Alt-Tab Date: 8 Jul 1997 03:40:17 GMT Organization: The PEAK ftp site for OpenStep and NEXTSTEP Message-ID: <5pscr1$2ck$1@ha2.rdc1.nj.home.com> References: <5pppik$hli@slip.net> <5prhjb$kb3@xmission.xmission.com> planetary <kris@xmission.xmission.com> wrote: >Emmett McLean <emclean@slip.net> wrote: >: Is there a NeXT equivalent to Window's Alt-Tab ? > >Cmd-[Up arrow|Down arrow] which isn't exactly the same, because alt-tab actually makes the window 'key' There is a PS hack floating around to do this. Oops.... here it is. Use at your own risk.... I didn't write it but have used it under 3.2/3.3/4.1 begin 644 change-key.patch M(R,@+2TM+2TM+2T@8F5G:6X@=VEN9&]W4&%C:V%G92TS+C,N<&%T8V@@+2TM M+2TM+2T**BHJ(&]R:6=I;F%L+3,N,R]W:6YD;W=086-K86=E+G!S"5=E9"!$ M96,@,3,@,3<Z-34Z-3,@,3DY-0HM+2T@:&%C:V5D+3,N,R]W:6YD;W=086-K M86=E+G!S"5=E9"!$96,@,3,@,3<Z-34Z-#$@,3DY-0HJ*BHJ*BHJ*BHJ*BHJ M*BH**BHJ(#$P-3(L,3`U-R`J*BHJ"BTM+2`Q,#4R+#$P.#4@+2TM+0H@(#4@ M+3$@<F]L;"!P;W`@,"`U(#$@<F]L;`H@(&%C=&EV94%P<"!E>&-H("]P;W-T M0GE#;VYT97AT('=I;F5X96,*("!](%].6$1E9E!R;V,**R`O86-T:79A=&5! M<'!7:71H5&]P5VEN9&]W('L**R!T<G5E(#`@+V=E=%=I;F1O=W,@=VEN97AE M8PHK(&9A;'-E(&5X8V@@>PHK(&1U<"!C=7)R96YT=VEN9&]W;&5V96P**R!D M=7`@+R]M86EN365N=4QE=F5L(&=T(&5X8V@**R`O+V1O8VM,979E;"!L="!O M<B!["BL@9'5P(&-U<G)E;G1W:6YD;W=D:6-T"BL@9'5P(&YU;&P@;F4@>PHK M("]I8V]N(&=E="`Q(&YE('L**R!T<G5E(&5X:70**R!](&EF"BL@?7L**R!P M;W`**R!](&EF96QS90HK('T@:68**R!P;W`**R!](&9O<F%L;`HK('L**R!D M=7`@+V%C=&EV871E0GE7:6YD;W<@=VEN97AE8PHK(&1U<"!C=7)R96YT=VEN M9&]W8F]U;F1S(&5X8V@@<&]P(&%D9`HK(#$@,R`Q(')O;&P@,"`P(#4@:6YD M97@@,"`P"BL@,B!I;F1E>"!C=7)R96YT;W=N97(**R`Y(&-O<'D@+W1R86YS M;6ET179E;G0@=VEN97AE8PHK(#(@.2`Q(')O;&P@+W1R86YS;6ET179E;G0@ M=VEN97AE8PHK('!O<"!P;W`**R!](&EF"BL@+W)E='5R;E=I;F1O=W,@=VEN M97AE8PHK('T@7TY81&5F4')O8PH@("]?:&%N9&QE0V]M;6%N9"!["B`@;6%R M:R`Q,"`Q(')O;&P*("`Q(&EN9&5X("TQ-B!B:71S:&EF=`HJ*BHJ*BHJ*BHJ M*BHJ*BH**BHJ(#$P.#(L,3`X-R`J*BHJ"BTM+2`Q,3$P+#$Q,3@@+2TM+0H@ M('T@+W)E=D9O<F%L;"!W:6YE>&5C"B`@8VQE87)T;VUA<FL*("`O<F5T=7)N M5VEN9&]W<R!W:6YE>&5C"BL@-2!I;F1E>"`O+T%L=&5R;F%T96UA<VL@86YD M(#`@;F4@>PHK("]A8W1I=F%T94%P<%=I=&A4;W!7:6YD;W<@=VEN97AE8PHK M('T@:68*("!]>PH@(&1U<"`Q-S4@97$@>PH@('1R=64@,"`O9V5T5VEN9&]W M<R!W:6YE>&5C"BHJ*BHJ*BHJ*BHJ*BHJ*@HJ*BH@,3`Y.2PQ,3`T("HJ*BH* M+2TM(#$Q,S`L,3$S."`M+2TM"B`@?2!F;W)A;&P*("!C;&5A<G1O;6%R:PH@ M("]R971U<FY7:6YD;W=S('=I;F5X96,**R`U(&EN9&5X("\O06QT97)N871E M;6%S:R!A;F0@,"!N92!["BL@+V%C=&EV871E07!P5VET:%1O<%=I;F1O=R!W M:6YE>&5C"BL@?2!I9@H@('U["B`@<&]P"B`@86-T:79E07!P('1R=64@+W!O M<W1">4-O;G1E>'0@=VEN97AE8PHC(R`M+2TM+2TM+2!E;F0@=VEN9&]W4&%C 9:V%G92TS+C,N<&%T8V@@+2TM+2TM+2T*"BTM ` end
From: bresink@informatik.uni-koblenz.de (Marcel Bresink) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.programmer Subject: Re: How to set up Renderman printing? Date: 8 Jul 1997 07:31:26 GMT Organization: University Koblenz / Germany Message-ID: <5psqce$jsf$2@newshost.uni-koblenz.de> References: <5pquj4$p0v@lyra.csx.cam.ac.uk> William.Clocksin@CL.cam.ac.uk wrote: > [...] I get a panel asking for a > renderer host (which the only choice is localhost), and then error panel > complaining that it can't connect to localhost. This will happen if your computer has never been connected to a network or you kept the NetInfo database in its initial state. The initial local NetInfo settings for RenderMan don't work in most configurations. Start /NextDeveloper/Demos/RenderManager.app and configure the real name of your machine (or again localhost if you aren't in a network) as public renderer. After that, rendering will hopefully work fine. Marcel --- Marcel Bresink, University of Koblenz, Institute for Computer Science Rheinau 1, D-56075 Koblenz, Germany, Fon: +49-261-9119-421 Fax: ...-497 MIME/NeXT Mail accepted --- WWW: http://www.uni-koblenz.de/~bresink
From: Laurent_Daudelin@fanniemae.com (Laurent Daudelin) Newsgroups: comp.sys.next.programmer Subject: Re: (PB) EOF NLS_LANG Date: 1 Jul 1997 12:46:47 GMT Organization: Fannie Mae Message-ID: <5pau7n$5tg2@postman.fanniemae.com> References: <EC89pM.8BA@x-lan.alienor.fr> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit valerie@mizar (Moukdarath Valerie) wrote: >I work on NeXTstep 3.2 and EOF 1.1 and Oracle 7.1 > >My oracle NLS_LANG is american_america.we8dec > >I don't know what NLS_LANG to use on EOF to read correctely my date > >ex : "Boite de crème à été" in my database is not the same when I read it with EOF. > > >é may be replace by "Y" or "î" it depends on what caracter set i use. > >valérie Valérie, It has nothing to do with EOF. This problem is caused by the fact that, for extended characters (e.g. é, è, etc.), there is no convention like the one we have for ASCII (characters 0 to 127). So, we ends up with different value for extended characters, depending on the platform (OS) you're using. So, I guess that your Oracle database server is probably running on an NT server, while you're developing your application on NeXTStep. There is not much you can do, however... -Laurent. ====================================================================== Laurent Daudelin "Risk-Based Pricing" Fannie Mae 13150 Worldgate Drive, Herndon, VA. USA Phone: 703-708-1830 EMail: Laurent_Daudelin@fanniemae.com Fax: 703-708-1857
Subject: Re: [Q] OpenStep Garbage Collection Newsgroups: comp.sys.next.programmer References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <199707061639241455966@dialup156-4-13.swipnet.se> <5popa2$sfp$1@darla.visi.com> <1997070718480867755@dialup103-4-10.swipnet.se> <5pricn$i0l$1@darla.visi.com> <5ps8ff$4ju$1@vader.wolfware.ipc.net> In-Reply-To: <5ps8ff$4ju$1@vader.wolfware.ipc.net> From: marco@sente.nospam.ch (Marco Scheurer (remove nospam to reply)) Organization: Message-ID: <33c1fa79.0@epflnews.epfl.ch> Date: 8 Jul 97 08:29:45 GMT On 07/08/97, Christopher Wolf wrote: >On 07/07/97, David Young wrote:>I don't work for Apple, so I obviously have no >idea how they're going to (or >>already have) reconcile(d) Java GC bridged to ObjC for OPENSTEP, and Java GC >>distributed across processes and networks for (P)DO. Recalling the discussion >>of Apple's changes to the ObjC language in comp.lang.objective-c, I'm >>actually starting to get a bit worried. > >What they said at WWDC... (working from memory) > >Conceptually you have two run-times working in parallel - the Obj-C runtime and >the Java runtime - with a bridge between them. > There was a small article about the integration of Java and Objective-C that was done for WebObjects in last month's Object Magazine. -- Marco Scheurer (marco@sente.ch) Sen:te
Subject: Re: Looking for an expert system shell Newsgroups: comp.sys.next.programmer References: <199707071603.MAA13005@micro.cs.rochester.edu> In-Reply-To: <199707071603.MAA13005@micro.cs.rochester.edu> From: marco@sente.nospam.ch (Marco Scheurer (remove nospam to reply)) Organization: Message-ID: <33c1fa02.0@epflnews.epfl.ch> Date: 8 Jul 97 08:27:46 GMT On 07/07/97, "Eric K. Ringger" wrote: >In comp.sys.next.programmer, Marco Scheurer wrote: >>I'm looking for an expert system shell I could easily use with >>NEXTSTEP 3.3. More specifically, I am looking for a forward chaining >>inference engine. >> >>Any reference to available systems, source code, commercial or not >>would be greatly appreciated. >[...] > >CLIPS might meet your requirements. I suppose you'll need to search >for the distribution, since I don't have a URL on hand. I seem to >recall that one of the NASA labs developed it, but I could be wrong. > Yes, thank you. CLIPS can be found at: http://www.cs.cmu.edu/afs/cs/project/ai-repository/ai/areas/expert/sys tems/clips/0.html It was developed by NASA, but is now supported by its original developers. The new home page for CLIPS is at http://www.ghg.net/clips/CLIPS.html -- Marco Scheurer (marco@sente.ch) Sen:te
From: bresink@informatik.uni-koblenz.de (Marcel Bresink) Newsgroups: comp.sys.next.programmer Subject: Re: OPENSTEP on 95 is several times slower than on NT! Date: 8 Jul 1997 07:53:15 GMT Organization: University Koblenz / Germany Message-ID: <5psrlb$ksr$1@newshost.uni-koblenz.de> References: <5pjcqj$mti$1@solaris.cc.vt.edu> Chuck_Esterbrook@orcacomputer.com wrote: > However, we have discovered a serious problem with OPENSTEP 4.2 on Windows > 95: our application runs 4 to 5 times slower than on Windows NT. Isn't that a general problem of Windows 95? It's a well-known problem that Windows applications which use lots of memory generally run much slower on 95 that on NT. The cause seems to be an ill implementation of the multi-tasking scheduler of 95 which counteracts the CPU's caching strategy, leading to a bad memory performance. See http://www.heise.de/ct/english/9611015/ for further information. So it's not NeXT to blame... Marcel --- Marcel Bresink, University of Koblenz, Institute for Computer Science Rheinau 1, D-56075 Koblenz, Germany, Fon: +49-261-9119-421 Fax: ...-497 MIME/NeXT Mail accepted --- WWW: http://www.uni-koblenz.de/~bresink
From: mmalcolm crawford <malcolm@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Some initial OpenStep notes Date: 8 Jul 1997 09:48:33 GMT Organization: P&L Systems Message-ID: <5pt2dh$pru$1@ironhorse.plsys.co.uk> References: <maury-0307971643370001@199.166.204.230> <paxon-0307972126420001@ppp-206-170-3-31.okld03.pacbell.net> <5pj9cf$nc$1@ironhorse.plsys.co.uk> <maury-0407971712190001@198.133.37.101> <5plifk$o83$1@ironhorse.plsys.co.uk> <maury-0707971204360001@199.166.204.230> <ECyoG5.Ep4@gateway.ali.bc.ca> <maury-0707971700430001@199.166.204.230> In-Reply-To: <maury-0707971700430001@199.166.204.230> On 07/07/97, Maury Markowitz wrote: >In article <ECyoG5.Ep4@gateway.ali.bc.ca>, allan@cetus.ali.bc.ca (Allan >Noordvyk) wrote: > >> While font and ruler wells would be visually obvious, I would still >> want the speed of font/ruler copy & paste. > > By no means do I suggest removing existing functionality, it's just that >I find the wells system to be rather easy to use, and visually useful >because they give you a "preview" of the property inside them. > Just in case you haven't found it yet, note that you can make the preview in the Font Panel persistent by Shift-clicking the Preview button. We've also extended the Font Panel's functionality in Mesa so that you can drag a Font "swatch" out of the preview well to drop on any cell. *** If other developers want, I think we can give away the (relatively straightforward) code to add this to your app -- any takers please get in touch. *** I do wonder, though, how you'd get a Ruler preview in a well? Unless you have a floating Panel like the current Tabs view... Best wishes, mmalc. --
From: Chuck_Esterbrook@orcacomputer.com Newsgroups: comp.sys.next.programmer Subject: Re: OPENSTEP on 95 is several times slower than on NT! Date: 8 Jul 1997 12:38:18 GMT Organization: Virginia Tech, Blacksburg, Virginia Message-ID: <5ptcbq$l1n$1@solaris.cc.vt.edu> References: <5pjcqj$mti$1@solaris.cc.vt.edu> <5psrlb$ksr$1@newshost.uni-koblenz.de> bresink@informatik.uni-koblenz.de (Marcel Bresink) wrote: > Isn't that a general problem of Windows 95? It's a well-known problem that > Windows applications which use lots of memory generally run much slower on 95 > that on NT. The cause seems to be an ill implementation of the multi-tasking > scheduler of 95 which counteracts the CPU's caching strategy, leading to a > bad memory performance. See http://www.heise.de/ct/english/9611015/ for > further information. > > So it's not NeXT to blame... I'm not sure about that. No other Win32 app I use has such major performance problems going from NT to 95. We're talking a slow down of 4 to 5. I will check out the site you mentioned. -- Chuck Esterbrook, Software Eng. http://www.orcacomputer.com/~chuck --------------------------------------------------------------------- chuck_esterbrook@orcacomputer.com / vo 540 231-3475 / fx 540 231-3480 Orca Computer, Inc. / 1800 Kraft Dr. Suite 111 / Blacksburg, VA 24060
From: Greg_Anderson@afs.com (Gregory H. Anderson) Newsgroups: comp.sys.next.programmer Subject: Re: OPENSTEP on 95 is several times slower than on NT! Date: 8 Jul 1997 14:48:18 GMT Organization: Anderson Financial Systems Inc. Message-ID: <5ptjvi$a89@shelob.afs.com> References: <5ptcbq$l1n$1@solaris.cc.vt.edu> Chuck_Esterbrook@orcacomputer.com writes > However, we have discovered a serious problem with OPENSTEP 4.2 on > Windows 95: our application runs 4 to 5 times slower than on Windows NT. As another data point, the beta release of WriteUp/OPENSTEP runs at about the same speed on equivalent hardware running 95 and NT (Compaq Presarios with Pentium/90mhz CPUs). I have not seen noticeable degradation on 95, even on deployment machines with "only" 32MB. Is it possible this is a network/TCP problem? At first, I couldn't get the app to run at all on a standalone machine, and it turned out to be the lack of an IP address. The pasteboard server needs to have TCP/IP services running and configured properly, or it hangs/crashes. And if DNS is not configured properly on a given machine, you can run into some hellacious waits for network activity. -- Gregory H. Anderson | "We're in the land of the blind, Visionary Ophthalmologist | selling working eyeballs, and they Anderson Financial Systems | balk at the choice of color." -- Tony greg@afs.com (NeXTmail OK) | Lovell, on Mac user reactions to NeXT
From: penrose@sfc.keio.ac.jp (Penrose Christopher) Newsgroups: comp.sys.next.programmer Subject: NSThread severely degrades event queue access Date: 08 Jul 1997 16:18:30 GMT Organization: Keio University Shonan Fujisawa Campus, Kanagawa Japan Distribution: world Message-ID: <PENROSE.97Jul9011830@ccs01.sfc.keio.ac.jp> Howdy Folks: I am developing an application, under Openstep 4.1 for intel, which is multithreaded. In my application, I have a single solitary method that is run in a seperate thread. It is spawned with: [NSThread detachNewThreadSelector:@selector(mySelector:) toTarget:self withObject:sender]; In this method (as referenced by mySelector) there are no AppKit objects manipulated, nor are any messages passed to any AppKit objects. Elsewhere in the application, in a custom view, mouseDown: is overridden to implement the dragging of another view. When this view is dragged on the screen before the application becomes mult-threaded (while [NSThread isMultiThreaded] returns NO) the dragging is smooth and continuous. If I execute mySelector: in the same thread as the application, the dragging performance is also smooth. However, as soon as the application becomes multithreaded, e.g. [NSThread isMultiThreaded] returns YES), the dragging behavior degrades drastically: the view stutters when dragged. A huge (sometimes as humungous as 500ms) latency is introduced. This latency remains in the application ever after, even though the detached thread has exitted. Why is this happening? Are there any ways to alleviate this unacceptable latency? I am tempted to use cthread_fork() to see if I can fool the AppKit into better performance, but I'd prefer this application to be Openstep compliant. Thanks for any help! Christopher Penrose penrose@sfc.keio.ac.jp penrose@silvertone.princeton.edu
From: penrose@sfc.keio.ac.jp (Penrose Christopher) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 08 Jul 1997 16:13:55 GMT Organization: Keio University Shonan Fujisawa Campus, Kanagawa Japan Message-ID: <PENROSE.97Jul9011355@ccs01.sfc.keio.ac.jp> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$ <19970705124505517300@dialup76-3-36.swipnet.se> <5pqjk0$ble$1@netty.york.ac.uk> <5prfv6$ck3@sjx-ixn4.ix.netcom.com> In-reply-to: mtrombin@ix.netcom.com's message of 7 Jul 1997 19:27:34 GMT mtrombin@ix.netcom.com (Mark Trombino): >However, the fact that a simple question about GC has >turned into such a long thread should raise a flag somewhere at Apple. It should raise a flag yes. However, I think that you know more about Openstep garbage collection than anyone at Apple outside of the "Apple Enterpise Software" division (formerly NeXT software inc.) You have had more to time to use it than they have. The lack of an xxx.next.com or xxx.apple.com in this thread indicates that no flag has been raised. Our problem is trivial, insignificant, and is being ignored in their ostrich eyes. This is a very bad sign as this thread is totally huge (50+ messages). It is very frustrating as a developer to have been given the promise of consistency yet only receive a voodoo reality. I have posted several times to the net with the hope that someone could answer me regarding multi-threaded applications (spawned with NSThread) and NSEvent event queue performance degradations. My problem is trivial in their (and your) eyes. In my eyes, my application is completely useless until I can find some solution. NSThread was a promise of redemption from mach c-threads. Instead it seems to have become a colossal waste of time and energy. I just can't wait to see what happens when we get an Openstep that Apple has monkey-wrenched. We are discussing garbage collection as designed by NeXT software. Who knows what Apple engineers will do (to) Next. They couldn't even modernize their kernel despite having many working os models to analyze and steal from. As an audio developer, like Mark, I wait in deep fear that the SoundKit will be replaced with an AIFF or Quicktime "solution", forcing me to rewrite huge amounts of code in hideously inelegant ways. I'll either be running linux at that point or running screaming off a builing in Tokyo. Software companies don't operate by tallying votes from its users or even developers, software companies decide how you are going to use and interact with their system for you. Christopher Penrose Audio DSP Researcher Keio University Shonan Fujisawa, Japan penrose@sfc.keio.ac.jp
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Tue, 8 Jul 1997 13:26:48 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <knkbVMS00iWW0465o0@andrew.cmu.edu> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> [ ...damned if you respond, damned if you don't... ] Excerpts from netnews.comp.sys.next.programmer: 7-Jul-97 Re: [Q] OpenStep Garbage Co.. by Lars Farm@ite.mh.se >> Java's semantics buy you more than that-- the Java GC is guaranteed that >> it can always free all unused memory because (for example) it cannot be >> fooled by integer aliases that appear to be valid pointers. A >> conservative GC is therefore likely to use more space than a >> non-conservative GC. > > A copying collector needs a heapsize of at least twice the allocated > size for copying. A conservative collector can make do with less. What you say is absolutely true for a situation without virtual memory. However, to understate the case, the presence of virtual memory interacts strongly with garbage collection. [ ... ] > As for speed, a copying collector must touch every allocated byte when > copying, but a conservative collector needs only touch pointers to > allocated blocks. You're disregarding VM again. Clever systems can take advantage of changing the page tables to copy and/or remap memory on a page-by-page basis very quickly compared to actually copying every byte. [ ... ] > Saying GC is ok, but not if it is conservative GC simply doesn't make > sense to me and I think that is what you're saying. Well, it's not. >> The problem comes in that unless you explicitly NULL old pointers to >> dynamic structures, they will continue to point to them and the GC will >> not free that memory. This means that such structures will stay around >> either forever if the pointer variable is global or automatic, or will >> stay around until after the call stack has recursed past the frame >> containing the local pointer variables. > > So don't use global variables;-) This "problem" is the same with any > kind of GC. It would be the same in Lisp, Smalltalk, Java or Eiffel > regardless of how the GC was implemented. It is also the same for > reference counting. What does that have to do with conservative GC > beeing an inferior kind of GC and therefor not worthy of using? It has nothing to do with such a claim. > Does this mean that you're arguing against GC in general, not against > conservative GC? No. I'm not arguing against GC at all, nor am I arguing against conservative GC. >> And if you're going to have to explicitly decide to NULL pointers to >> avoid leaking uncollectable memory, then there is no gain in convenience >> over something like NeXT's current reference counting scheme where you'd >> send a -release message. > > ... or say free() in C, delete in C++ or dispose in Pascal. What is your > point? I was refuting your claim that implementing a conservative GC for a C-like language somehow makes the language that much easier to use because you never have to worry about managing memory anymore. IMHO, a conservative GC scheme has about the same level of convenience to use that a good reference-counting scheme like NeXT's does. [ ...autorelease... ] Go look on NeXT's web site for documentation on autorelease. Since you didn't know what it is, there's no point in addressing your impression of what that thought it does. > Your argument is a variation of one of the standard arguments against > GC, not against conservative GC in particular: - "If it doesn't solve > programming problem X as well as reusing unused memory it's no good". That has not been my argument-- I understand what GC is, and that it can be very nice under some circumstances. The problem is that you don't seem to understand that GC is not a panacea, and that it is not a solution to all the world's programming ills. There are tradeoffs involved, but you appear to discount them. Also, your analyses (and the ones on the web you cited) claim to apply to the majority of common cases and thus are generalizable, but the simple fact that they largely fail to consider the impact of virtual memory, or the impact of the active working set of pages over a machine running multiple processes via PMT makes me question the accuracy of the claims being derived. A copying collector has the advantage over a mark-sweep collector that it improves locality by removing empty holes as a result of performing GC. NEXTSTEP's memory management system takes an alternate approach via the concept of memory allocation zones, which can also greatly improve locality when used correctly. [ ... ] > GC is often combined with 'finalization' and finalization can be used to > release other external resourses. Without pretending to know anything > about NeXT distributed objects I assume finalization could be used for > them. It's possible to implement finalization very easily in Obj-C for any NSObject classes you desire. >> Simply assigning pointers around does not change the reference count, so >> it is perfectly possible to create self-referential structures which >> will be autoreleased later without any manual memory management. > > Then it isn't reference counting. It doesn't count references. Perhaps > only "important" references and semi-manual counting or perhaps even > manual counting? The reference count involves explicit retain/release combinations, which are often handled automatically via inherited behavior from various classes in use. But yes, there is some level of explicit manual counting that one encounters, depending on what you're doing. > Doesn't that expose one to the usual pointer related problems with > dangling pointers, etc, etc? Yes. There are debugging utilities provided with the system and available as add-ons which allow one to track down dangling pointer references in an app under development and eliminate them. The 80-20 rule applies here-- NEXTSTEP's system provides about 80% of the convenience of full GC at about 20% of the cost. Trying to do better means that you run into the aforementioned problems and cost with full GC. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: mtrombin@ix.netcom.com (Mark Trombino) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 8 Jul 1997 20:22:58 GMT Organization: Egghead Billy, Inc. Message-ID: <5pu7j2$egf@dfw-ixnews9.ix.netcom.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$ <19970705124505517300@dialup76-3-36.swipnet.se> <5pqjk0$ble$1@netty.york.ac.uk> <5prfv6$ck3@sjx-ixn4.ix.netcom.com> <5prkc0$i0l$2@darla.visi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In <5prkc0$i0l$2@darla.visi.com> David Young wrote: > > For instance, I think that whenever you create an object, whether its by > > copy, alloc, or special class methods, and you want to keep it around > outside > > the scope of the method its created in, then you should have retain it. > > Then, its a simple matter of always matching a retain with a release. Its > > seems simple and consistant -- so there's probably a reason why its not > this > > way! > > That's not how it is. I know! That's how I think it *should* be though! > > See above. I hope this clears up some of your confusion. > > Thanks David, Marcel, Chris, and everyone else who have answered my posts! I will, after this, be quiet. I understand how it is *now*. I get it. But I still think its conceptually inconsistent. It would be consistent if whenever you get an object, whether you create it or not, but you get a pointer to it and you want to make sure that that object stays around until you are done with it, that you retain it. Then whenever you are through with it, you release it. Isn't that more elegant and simple? There's nothing to question - I retained it, so I have to release it. I want to keep it for a while, so I must retain it. That's all I'm saying.... Also, and this is very minor :), why don't these "convenience methods" like +dictionary or +array also leave the refcount at 1. Why do they autorelease them before returning? Not as "convenient" as they could be! What I was looking for was the reason why its not this way now. For example, maybe there's some performance hit by using autorelease. If every object created was added to the autorelease pool, maybe that would be a problem. I don't know this to be the case, I'm just talking out my ass... And that's it from me. Thanks for the help! -- |-------------------------------------------------------------------------| | Mark Trombino | J A M S o f t | | mtrombin@ix.netcom.com | Audio DSP Tools for Openstep & Rhapsody | |-------------------------------------------------------------------------|
Date: Tue, 08 Jul 1997 15:40:36 -0600 From: David Stes <stes@can.nl> Subject: Re: [Q] OpenStep Garbage Collection Newsgroups: comp.sys.next.programmer Message-ID: <868393273.25503@dejanews.com> Organization: Computer Algebra Netherlands References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <199707061639241455966@dialup156-4-13.swipnet.se> <5popa2$sfp$1@darla.visi.com> <1997070718480867755@dialup103-4-10.swipnet.se> In article <1997070718480867755@dialup103-4-10.swipnet.se>, lars.farm@ite.mh.se (Lars Farm) wrote: > > How come GC is acceptable for OpenStep in Java but GC is not acceptable > for OpenStep in other languages? That's a very good question. Your reaction is very much unlike what, I believe, is common among Apple developers : "Objective-C has auto-release pools : I hope they convert OpenStep to Java". It's the _OpenStep_ spec that has auto-release pools ... (OpenStep encourages you to convert NextStep apps (which don't use autorelease pools) to use the brand new NeXT Foundation Kit auto-release pool alternative of the new NSObject class instead of the +new method of the Object class). Brad Cox' 86 (!) and 91 books have a discussion, both of mark-and-sweep gc and refcount (automatic) gc, applied to Objective-C. In fact the Stepstone Object class has, in addition to the "isa" field, fields such as "attr" and "objID", which were used as identifier for double-indirection, for heap compaction, to fight fragmented memory, or for flags for static allocation etc. My compiler has a -gc option (using the Boehm package); the following test results might be of interest. stes@mix> vi boo.m "boo.m" 9 lines, 103 characters #include "Object.h" void main() { int i; for(i=0;i<10000000;i++) [Object new]; printf("done\n"); } stes@mix> objc --version Portable Object Compiler - objc1.4.5 - m68k-next-nextstep3 Copyright (c) 1996,97 by Stes & Lerman. All Rights Reserved. stes@mix> objc -gc boo.m stes@mix> /bin/time a.out done 326.2 real 281.5 user 1.9 sys During this time I was measuring (with ps -aux) the vm and real memory of the process; real size was stable at 504 KB, vm size stable at about 1.5 MB. Then I repeat the test without the -gc flag : stes@mix> objc boo.m stes@mix> /bin/time a.out I was going to save the output of the ps -aux test (which showed something like 28MB of fastly growing VM), but could not, because of "no space left on device" at the time I was going to save the results. I think it's because the swapfile is dynamically growing and was consuming all space on my disk; I had to reboot my machine. The same test on my zippy Pentium machine running FreeBSD (BSD 4.4) : psyche: {1} objc --version Portable Object Compiler - objc1.4.7 - i686-unknown-freebsd2.1.7 Copyright (c) 1996,97 by Stes & Lerman. All Rights Reserved. psyche: {5} objc boo.m psyche: {6} a.out Out of memory on call to OC_Malloc psyche: {7} objc -gc boo.m psyche: {8} /usr/bin/time a.out done 30.32 real 30.29 user 0.00 sys I did not have to reboot. David. -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 8 Jul 1997 22:23:30 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5puel2$af5$1@brachio.zrz.TU-Berlin.DE> References: <5pu7j2$egf@dfw-ixnews9.ix.netcom.com> In article <5pu7j2$egf@dfw-ixnews9.ix.netcom.com> mtrombin@ix.netcom.com (Mark Trombino) writes: > What I was looking for was the reason why its not this way now. > For example, maybe there's some performance hit by using autorelease. Autorelease actually has several performance hits: 1. The object has to be added to the autorelease pool and a release message has to be sent. 2. Accessing the current autorelease pool means accessing the current thread dictionary, not a fast operation. 3. Increasing the retain count to above 1 (for example when you retain an object that was previously autoreleased) is very expensive in the default NSObject implementation, the implementation uses an external hash table because there are no bits left in the object. Ugh! 4. Unless you have you manage all your own autorelease pool, you have no control over when the object gets released. So, use alloc/init and explicit release whenever possible, autorlease only when returning temporary objects from methods/functions. Marcel
From: thomas@zippy.sonoma.edu (Thomas Poff) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.programmer Subject: Re: Sun Sparc TOO! (was Re: Prelude2Rhapsody on black hardware!) Date: 8 Jul 1997 22:55:38 GMT Organization: Information Resources and Technology Message-ID: <5pugha$bbo$1@nuke.csu.net> References: <rbarris-ya023280001106971149270001@news.intelenet.com> <WEGMANN.97Jul7143948@talisker.linguistics.ruhr-uni-bochum.de> Hi. Actually I just said: boot cdrom at that funky BIOS prompt after hitting <stop-A> and awaaaaaaaaaaaaaaaaaaayyyyyyyyyyyyyyyyyyyyyyyyyyyyyy I went! I've been running Prelude on my Sparc 5 for a couple of weeks now. It was a no-brainer. I wonder what the hangup was with your Sparc 20. Did you have a genuine-like Sun CD-ROM player? Thomas Frank Wegmann (wegmann@talisker.linguistics.ruhr-uni-bochum.de) wrote: : In article <5p11qf$ar2$2@nuke.csu.net> thomas@zippy.sonoma.edu (Thomas Poff) writes: : > Prelude installed on a friend's Sparc trivially. No muss no fuss. : > It was a trivial installation. : Well, almost. You cannot boot from the PTR CD (at least on my SS20), so : you need a bootable NS3.3/OS4.x CD-ROM. Then change disks at the boot prompt : and continue. From that point, installation is a no-brainer. : Frank : -- : Frank Wegmann voice: +49 234 700 7677 / +49 234 700 2461 : Sprachwiss. Institut fax : +49 234 7094 137 : Ruhr-Universitaet Bochum email: wegmann@linguistics.ruhr-uni-bochum.de, : 44780 Bochum wegmann@acm.org (NeXTmail, MIME welcome) : Germany WWW : http://www.linguistics.ruhr-uni-bochum.de -- <>+<> ////// __v__ __\/__ `\|||/ /---\ """"""" | _ - | (_____) . / ^ _ \ . (q p) | o o | <^-@-@-^> (| o O |) .( O O ), |\| (o)(o) |/| _ooO_<_>_Ooo_ooO_U_Ooo_ooO__v__Ooo_ooO_u_Ooo_ooO__(_)__Ooa__oOO_()_OOo___ [_____}_____!____.}_____{_____|_____}_____i____.}_____!_____{_____}_____] __.}____.|_____{_____!____.}_____|_____{.____}_____|_____}_____|_____!___ [_____{_____}_____|_____}_____i_____}_____|_____}_____i_____{_____}_____] Thomas Poff 1308 Michele Ct. Rohnert Park, CA 94928 (707)664-1867 To see some interesting software for the Newton, please try: http://www.cs.sonoma.edu/Newton ftp://ftp.cs.sonoma.edu/pub/Newton
From: rmessner@tolkien.BLaCKSMITH.com (Rick Messner) Newsgroups: comp.sys.next.programmer Subject: Lex/Yacc in 4.1 PB Date: 8 Jul 1997 22:46:30 GMT Organization: BLaCKSMITH, Inc. Message-ID: <5pug06$iml$1@anvil.BLaCKSMITH.com> I have lex and yacc files I cannot get to compile in 4.1 PB. It seems as if the extension is not being recognized, but yet it is listed in the basicrules.make file. I have tried .lm, .lm.m, and .l.c and corresponding yacc extensions to no avail. Any help would be appreciated. Thanks, Rick Messner, BLaCKSMITH, Inc., rick_messner@blacksmith.com
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 8 Jul 1997 23:12:43 GMT Organization: 21st Century Software, New York City Sender: daver@ts1-1.nj.cnct.com Message-ID: <5puhhb$h4r$1@darla.visi.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$ <19970705124505517300@dialup76-3-36.swipnet.se> <5pqjk0$ble$1@netty.york.ac.uk> <5prfv6$ck3@sjx-ixn4.ix.netcom.com> <5prkc0$i0l$2@darla.visi.com> <5pu7j2$egf@dfw-ixnews9.ix.netcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 08 Jul 1997 18:12:43 CDT Cc: mtrombin@ix.netcom.com In <5pu7j2$egf@dfw-ixnews9.ix.netcom.com> Mark Trombino wrote: > I understand how it is *now*. I get it. But I still think its conceptually > inconsistent. It would be consistent if whenever you get an object, whether > you create it or not, but you get a pointer to it and you want to make sure > that that object stays around until you are done with it, that you retain it. > Then whenever you are through with it, you release it. Isn't that more > elegant and simple? There's nothing to question - I retained it, so I have > to release it. I want to keep it for a while, so I must retain it. That's > all I'm saying.... In order for this to happen, objects would have to begin their lifetime with a retainCount of zero. This doesn't really make any sense as unreferenced objects are freed! Also, if you ever _did_ add on mark-sweep GC over the retainCount system (there's a package that does this for GNUstep) (not that I'm suggesting this), this would be very confusing. > Also, and this is very minor :), why don't these "convenience methods" like > +dictionary or +array also leave the refcount at 1. Why do they autorelease > them before returning? Not as "convenient" as they could be! They _do_ leave the retainCount at one. They _also_ place the object in the current autorelease pool. They retainCount _stays_ at one until the current pool is freed. The "convenience" implied is that you don't have to type alloc/init and release. They're intended for temporary calculation and processing, and for the most part, once you're done with them, you can forget about them. > What I was looking for was the reason why its not this way now. For example, > maybe there's some performance hit by using autorelease. If every object > created was added to the autorelease pool, maybe that would be a problem. I > don't know this to be the case, I'm just talking out my ass... There is a performance hit. There's the addition of the object to the pool, processing involved in the pool, etc. Also, autorelease+retain is slower than just alloc/init. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: independant software and network guy ::::: new york, new york :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
Message-ID: <33C2FCE7.942F47D8@neteffects.com> Date: Tue, 08 Jul 1997 19:52:23 -0700 From: "Jack G. Bader" <jbader@neteffects.com> Organization: NetEffects Inc. MIME-Version: 1.0 Subject: **** Need Objective-C/OpenStep Programmer in Minneapolis **** Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Newsgroups: comp.sys.next.programmer,comp.lang.objective-c NetEffects has an immediate opening for a consultant with one of our Fortune500 clients in Minneapolis. You will join a small team of NeXtStep/OpenStep developers who are enhancing large customer service applications. If you have 1+ years of Objective-C background you are a candidate for this long-term position. Please email your resume and availablility as soon as possible. Thanks in advance. Jack Bader jbader@neteffects.com
From: scott@doubleu.com (Scott Hess) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 9 Jul 1997 02:00:31 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jul8215807@slave.doubleu.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$ <19970705124505517300@dialup76-3-36.swipnet.se> <5pqjk0$ble$1@netty.york.ac.uk> <5prfv6$ck3@sjx-ixn4.ix.netcom.com> <5prkc0$i0l$2@darla.visi.com> <5pu7j2$egf@dfw-ixnews9.ix.netcom.com> In-reply-to: mtrombin@ix.netcom.com's message of 8 Jul 1997 20:22:58 GMT In article <5pu7j2$egf@dfw-ixnews9.ix.netcom.com>, mtrombin@ix.netcom.com (Mark Trombino) writes: I understand how it is *now*. I get it. But I still think its conceptually inconsistent. It would be consistent if whenever you get an object, whether you create it or not, but you get a pointer to it and you want to make sure that that object stays around until you are done with it, that you retain it. Then whenever you are through with it, you release it. Isn't that more elegant and simple? There's nothing to question - I retained it, so I have to release it. I want to keep it for a while, so I must retain it. That's all I'm saying.... There are two reasons that come immediately to mind. One is that then you'd _always_ do -retain on +alloc'ed objects, making the -retain redundant. The other is that this would significantly change how +alloc works, meaning it'd need to change to a different method name, leaving an +alloc which didn't need the retain for compatibility reasons... Actually, I just thought of another reason. There _must_ be an implicit -retain for +alloc - otherwise, how does the object manage to exist with a retainCount of 0? Also, and this is very minor :), why don't these "convenience methods" like +dictionary or +array also leave the refcount at 1. Why do they autorelease them before returning? Not as "convenient" as they could be! They do leave the refCount at 1 :-). The real reason is that generally it's expected that you aren't going to hold onto the returned object. For instance, you might ask some other object for a dictionary, which you in turn ask for a particular value by key, which you in turn pass to some other method. The real base of the problem is a misunderstanding of what the scope of an object that some method returned to you. In the best world, the object's scope should be through the end of the calling method, and specifically until the stack is unwound and the autorelease pool is released. But sometimes you can get an object which is a direct reference to some other objects storage area. For instance, if an object is storing data "foo" in a string, and you ask for the "foo" data, it might just return the "foo" instance variable directly. If you then send a message to the object which causes the "foo" instance variable to be changed, then you have a dangling reference to a potentially deallocated object. In this particular case, if the method returning "foo" instead of doing "return foo" had done something like "return [[foo retain] autorelease]", or perhaps -copy/-autorelease, then you wouldn't have a problem. Likewise if the code that changed foo used -autorelease instead of -release. There are other more involved problems, but most of them are similar, -- scott hess <scott@doubleu.com> (606) 578-0412 http://www.doubleu.com/ <Favorite unused computer book title: The Idiots Guide to the Zen of Dummies in a Nutshell in Seven Days, Unleashed>
From: Damn Yankee<damnyankee@yankee.inc> Newsgroups: comp.sys.next.programmer Organization: Yankee Inc. Subject: !!!Hello!!! Message-ID: <33c0b845.0@nntp.kalnet.net> Date: 7 Jul 97 09:35:01 GMT Are you looking for a great place to find interesting things on the WWW??? Than look no more because you have found the right place!!! The place you are looking for is Yankee Inc.!!! A great Web Site full of interesting links to interesting places!!! Everything from Adult to Xzlyaphone it's here!!! There is something for everyone and then some!!! So give it a try what do you have to lose except some sleep??? Yankee Inc. http://www.kalnet.net/yank714 Yankee Inc. Your Alternative Web Solution!!! Yankee Inc. http://www.kalnet.net/yank714 If you would like to be removed from my mailing list - hit reply and type "REMOVE" and I will promptly remove you from my list!!! Thank You!!!
From: cjs@occs.cs.oberlin.edu (Miles Standish) Newsgroups: comp.sys.next.programmer Subject: Academic vs Developer: cheaper? Date: 9 Jul 1997 03:55:55 GMT Organization: Oberlin College Message-ID: <5pv24b$k18@news.cc.oberlin.edu> The adacemic bundle is only $319 but becoming an apple developer is only $250. Any idea which is a better deal? Will I get them same (or bettter) sutff as a develper as I get on the academic CD? And when is Rhapsody coming out? (It is July right?) Thanks in advance!
Newsgroups: comp.sys.next.programmer,comp.sys.psion.programmer,comp.sys.sgi.graphics,comp.unix.programmer,comp.unix.sco.programmer,comp.windows.ms.programmer,ctdl.lang.c++,ctdl.lang.pascal,dc.graphics.avs,de.alt.binaries.sounds,de.comp.lang.c,de.comp.lang. From: "Aaron Smith" <bluegras@iglou.com> Subject: AL'S SOFTWARE LOOKING FOR MEMBERS Message-ID: <01bc8c17$1f9f47c0$3ee8ffcc@s-1242> Sender: news@iglou.com (News) Organization: fy Date: Wed, 9 Jul 1997 03:23:44 GMT ***If you are a programmer/artist/composer this is an opportunity you don't want to miss!*** AL's SoftWare "Programming Group" (which currently has over 150 members) is now looking for new members! Go to http://www.gos.net/alsoft/ProgGroup/ and read all about it! If you think this is a "JUNK ADVERTISEMENT", let me tell you that the membership is free, and you will get a chance to make money. This is a real job, not a chain letter or any other "get rich scheme". As a member you will get free webspace, free advertising for your products, all the help/files you need and much more - all you pay is a super-tiny percentage (tax) of the money you make (after you make it). -- -- AL's Software =+= AARON SMITH =+= =+= Head Recruiter =+= SEND MAIL TO: Alex Libman president <alsoft@cybercomm.net>
Date: Wed, 09 Jul 1997 03:42:35 -0600 From: David Stes <stes@can.nl> Subject: Re: [Q] OpenStep Garbage Collection Newsgroups: comp.sys.next.programmer Message-ID: <868437336.29789@dejanews.com> Organization: Computer Algebra Netherlands References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <19970705124505517300@dialup76-3-36.swipnet.se> <gnjemhO00iWp03KEs0@andrew.cmu.edu> <199707061639241455966@dialup156-4-13.swipnet.se> <5popa2$sfp$1@darla.visi.com> <1997070718480867755@dialup103-4-10.swipnet.se> <868393273.25503@dejanews.com> In article <868393273.25503@dejanews.com>, David Stes <stes@can.nl> wrote: > > ... some facts supporting Lars' claims ... > About the program that I posted yesterday, psyche: {1} less boo.m #include <objpak.h> void main() { int i; for(i=0;i<10000000;i++) [Object new]; printf("done\n"); } I just realized that I missed a fine opportunity to demo the interesting -inlinecache option. psyche: {1} setenv OBJCOPT "-q -gc -inlinecache" psyche: {2} objc boo.m psyche: {3} /usr/bin/time a.out done 8.75 real 8.75 user 0.00 sys On the same Pentium Pro machine running FreeBSD, as yesterday. What would be most instructive, is to make a timing table of this simple test, using different memory management strategies. It's not just "faster" that is "better", I think you also have to take simplicity of the source code into account (maybe measured by source bulk : 'wc' ?) DAvid. -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: Kris Jacobs<jtsnake@serv01.net-link.net> Newsgroups: comp.sys.next.programmer Organization: Jake's Home Brew Subject: Try Me! Message-ID: <33c1fb15.2@nntp.kalnet.net> Date: 8 Jul 97 08:32:21 GMT Hello net world are you a beer drinker or maybe a home brewer than this web page is for you! My web page is dedicated to home brewing and beer on the net! If this interests you than go to Jake's Home (brew) Page it is located at http://www.net-link.net/~jtsnake/ I am looking forward to hearing from you soon!!! Kris Jacobs Jake's Home (brew) Page http://www.net-link.net/~jtsnake/ E-Mail To: jtsnake@net-link.net jtsnake@serv01.net-link.net mpinc@SERV01.NET-LINK.NET
From: Alex Blakemore <alex@genoa.com> Newsgroups: comp.sys.next.programmer Subject: Re: How to scroll from keyboard Date: 9 Jul 1997 05:15:27 GMT Organization: Genoa Software Systems Message-ID: <5pv6pf$f0a@saturn.genoa.com> References: <5pppd1$hhr@slip.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: emclean@slip.net In <5pppd1$hhr@slip.net> Emmett McLean wrote: > Is there a way to scroll to the bottom > of a terminal session without draging If you are using Mach, consider getting the shareware app Stuart to replace Terminal. Its much nicer in several respects. One of its strengths is that you can do many things from the keyboard, without using the mouse. scroll up is alt-up arrow. page up is shift-alt up arrow. goto top is control-alt up arrow (down is similar). You can also cycle through windows and icons with the keyboard. These sound small, but they eliminate many interuptions to reach for the mouse, and thus save time. -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: Alex Blakemore <alex@genoa.com> Newsgroups: comp.sys.next.programmer Subject: Re: Lex/Yacc in 4.1 PB Date: 9 Jul 1997 05:23:50 GMT Organization: Genoa Software Systems Message-ID: <5pv796$f0a@saturn.genoa.com> References: <5pug06$iml$1@anvil.BLaCKSMITH.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: rmessner@tolkien.BLaCKSMITH.com In <5pug06$iml$1@anvil.BLaCKSMITH.com> Rick Messner wrote: > I have lex and yacc files I cannot get to compile in 4.1 PB. It seems as > if the extension is not being recognized, but yet it is listed in the > basicrules.make file. I have tried .lm, .lm.m, and .l.c and corresponding > yacc extensions to no avail. Any help would be appreciated. Thanks, Rick, I'm not sure about Project Builder, but the standard Unix extensions for lex (or flex) files is .l, and the extension for yacc (or bison) files is .y. Hope this helps. Alex -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: maury@softarc.com (Maury Markowitz) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Some initial OpenStep notes Date: Wed, 09 Jul 1997 10:41:25 -0400 Organization: SoftArc Inc. Message-ID: <maury-0907971041250001@199.166.204.230> References: <maury-0307971643370001@199.166.204.230> <paxon-0307972126420001@ppp-206-170-3-31.okld03.pacbell.net> <5pj9cf$nc$1@ironhorse.plsys.co.uk> <maury-0407971712190001@198.133.37.101> <5plifk$o83$1@ironhorse.plsys.co.uk> <maury-0707971204360001@199.166.204.230> <ECyoG5.Ep4@gateway.ali.bc.ca> <maury-0707971700430001@199.166.204.230> <5pt2dh$pru$1@ironhorse.plsys.co.uk> In article <5pt2dh$pru$1@ironhorse.plsys.co.uk>, mmalcolm crawford <malcolm@plsys.co.uk> wrote: > Just in case you haven't found it yet, note that you can make the preview in > the Font Panel persistent by Shift-clicking the Preview button. How is this used? > We've also extended the Font Panel's functionality in Mesa so that you can > drag a Font "swatch" out of the preview well to drop on any cell. This is my point, all apps should allow this. > I do wonder, though, how you'd get a Ruler preview in a well? Unless you > have a floating Panel like the current Tabs view... I've been thinking about this. What you do is use a swatch of "greek"ed text and changing ruler settings make the picture change. Then you just pick up the greeked text and drop it on a paragraph. This preserves the idea of using the well as a preview area. Maury
From: Chuck_Esterbrook@orcacomputer.com Newsgroups: comp.sys.next.programmer Subject: Re: OPENSTEP on 95 is several times slower than on NT! Date: 9 Jul 1997 16:49:44 GMT Organization: Virginia Tech, Blacksburg, Virginia Message-ID: <5q0ff8$gu1$1@solaris.cc.vt.edu> References: <5ptcbq$l1n$1@solaris.cc.vt.edu> <5ptjvi$a89@shelob.afs.com> Greg_Anderson@afs.com (Gregory H. Anderson) wrote: > > Is it possible this is a network/TCP problem? At first, I couldn't get > the app to run at all on a standalone machine, and it turned out to be > the lack of an IP address. The pasteboard server needs to have TCP/IP > services running and configured properly, or it hangs/crashes. And if > DNS is not configured properly on a given machine, you can run into some > hellacious waits for network activity. > -- > Gregory H. Anderson | "We're in the land of the blind, > Visionary Ophthalmologist | selling working eyeballs, and they > Anderson Financial Systems | balk at the choice of color." -- Tony > greg@afs.com (NeXTmail OK) | Lovell, on Mac user reactions to NeXT I think I would have noticed this in other apps such as Explorer (Network Neighborhood), Netscape and FTP. That's not the problem, but thanks for the attempt. I'm guessing that there is some part of OPENSTEP that we are using that suffers performance problems on 95. We just can't tell what it is since Sampler doesn't work. -- Chuck Esterbrook, Software Eng. http://www.orcacomputer.com/~chuck --------------------------------------------------------------------- chuck_esterbrook@orcacomputer.com / vo 540 231-3475 / fx 540 231-3480 Orca Computer, Inc. / 1800 Kraft Dr. Suite 111 / Blacksburg, VA 24060
From: penrose@w09.sfc.keio.ac.jp Newsgroups: comp.sys.next.programmer Subject: CONTEST! solve the NSThread event queue performance bug and win! Date: 10 Jul 1997 02:07:57 +0900 Organization: Keio University Shonan Fujisawa Campus, Fujisawa Japan Message-ID: <wz7mf02lxu.fsf@w09.sfc.keio.ac.jp> 100 Yen goes to the first cogent reply to my earlier inquiries regarding the use of NSThread and NSEvent queue performance degradation. Update: I have checked the event queue to compare its contents during both cases: normal performance (pre-spawning of thread) and degraded performance (thread spawned AND exited). The event queue has a comparable list of mouseDown and mouseUp (1 of each) events during the modal loop, and a comparable number of dragging events for comparable drags. No other NSEvent types are in the queue. I haven't been able to trap where the delay occurs yet. It definitely doesn't seem to be my code. Looking at my thread-list in gdb before and after some interesting things can be seen. Before I spawn my thread there is only one thread present in the application (plus gdb's thread): There are 2 threads. Thread State Suspend Port PC Name Function 0 S 0 0x1501 0x504ffec main _msg_receive_trap 1 S 0 0x1701 0x1201afc4 gdb dynamic lib debugging thread During the thread execution we have 4 threads, including the mysterious newcomer, Run Loop Worker: There are 5 threads. Thread State Suspend Port PC Name Function 0 S 0 0x1501 0x504ffec main _msg_receive_trap 1 S 0 0x1701 0x1201afc4 gdb dynamic lib debugging thread 2 S 0 0x1b01 0x504ffec Run Loop Worker _msg_receive_trap 3 S 0 0x1c01 0x7108 Foundation-created thread -[MixDocument playMixThread:] 4 S 0 0x1d01 0x504ffec _t4 _msg_receive_trap And after my thread exits all four threads are still present: There are 5 threads. Thread State Suspend Port PC Name Function 0 S 0 0x1501 0x504ffec main _msg_receive_trap 1 S 0 0x1701 0x1201afc4 gdb dynamic lib debugging thread 2 S 0 0x1b01 0x504ffec Run Loop Worker _msg_receive_trap 3 S 0 0x1c01 0x504ffec _t3 _msg_receive_trap 4 S 0 0x1d01 0x504ffec _t4 _msg_receive_trap First of all, Run Loop Worker suggests that the AppKit is the cause of my problems. It suggests that certain AppKit functions are being handled in this thread in a sort of middle stage toward a thread-safe AppKit release. It would be really really nice if someone from Apple could comment on this. Demystify it just a tad, please? If you do, you'll be the winner of 100 yen. It would be nice if developers had access to the information that describes, in detail, what exactly happens when your application becomes multi-threaded with NSThread. Your silence is deafening. I have used mach threads just fine in other applications, but they are cumbersome, inelegant, and the biggest bummer, they are not Openstep compliant. I have thought also about implementing this with distributed objects, but I keep thinking to myself that threads should be easier than that. I should be able to run something in a seperate thread, and when the thread exits, I should be able to drag an object on the screen with full system performance. Right now, Openstep doesn't work that way. I am leaning towards mach-threads, but I must remind you in the community who actually managed to read this far, mach-threads are scorned all throughout the Openstep documentation. Christopher Penrose penrose@sfc.keio.ac.jp
From: lars.farm@ite.mh.se (Lars Farm) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Wed, 9 Jul 1997 23:38:41 +0200 Organization: pv Message-ID: <199707092338411585593@dialup96-5-11.swipnet.se> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <knkbVMS00iWW0465o0@andrew.cmu.edu> Cache-Post-Path: mn8!s-49817@dialup186-1-11.swipnet.se Charles William Swiger <cs4w+@andrew.cmu.edu> wrote: > > Your argument is a variation of one of the standard arguments against > > GC, not against conservative GC in particular: - "If it doesn't solve > > programming problem X as well as reusing unused memory it's no good". > > That has not been my argument-- I understand what GC is, and that it can > be very nice under some circumstances. The problem is that you don't > seem to understand that GC is not a panacea, and that it is not a > solution to all the world's programming ills. Interesting quoting technique and similarly interesting deduction. I said "useful tool, but no silver bullet". From that you deduce that "useful tool, but no panacea" is beyond my grasp... Hmmm... Please reread the rest of the quoted paragraph (snipped by you) and the next paragraph (snipped by you). Compare to what you said above. > Also, > your analyses (and the ones on the web you cited) claim to apply to the > majority of common cases and thus are generalizable, but the simple fact > that they largely fail to consider the impact of virtual memory, or the > impact of the active working set of pages over a machine running > multiple processes via PMT makes me question the accuracy of the claims > being derived. Well if you actually read it and then disregard the experiences of experts in the field like Hans Boehm, Paul Wilsson, ..., and the others as easily as that, then your line of argument suddenly makes sense. -- Lars Farm; lars.farm@ite.mh.se - Limt/Channelmatic: lars.farm@limt.se
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 9 Jul 1997 15:16:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFE9608B-A7701@206.165.44.26> To: HyperCardList@pogo.interedu.com, "GX-Talk" <gx-talk@aimed.org>, "Subscribers to" <semper.fi@private.lists.apple.com>, "AIMED-Talk" <aimed-talk@aimed.org> nntp://news.primenet.com/comp.sys.mac.programmer.games, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit I think that all Mac developers and other interested people should meet on AIMED-TALK (AIMED = Association of Independent Macintosh Engineers and Developers) to discuss how to handle this situation (Amelio and Hancock resigning). We need to act very quickly to respond to this in an Apple/Mac/Rhapsody-supportive way. If we can portray developers as taking a "business as usual" stance, maybe the press will at least mention this... Talk about ammo for more FUD... We gotta be as positive (in a non-finger-pointing way) as possible about this situation. And figure out how to convince dealers, customers and stockholders that this is not the end of the world for Apple/Macintosh/etc. ++++++++++++++ HOW DO I SUBSCRIBE TO AIMED-TALK? ------------------------------------- To subscribe to AIMED-TALK proceed as follows: Send E-mail to: <aimed-talk-request@aimed.org> with a blank subject line and the following command as the first (and only) line of the message body: SUBSCRIBE ++++++++++++++++++
Date: Wed, 09 Jul 1997 18:30:49 -0400 From: joe.ragosta@dol.net (Joe Ragosta) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Message-ID: <joe.ragosta-ya02408000R0907971830490001@news.dol.net> References: <AFE9608B-A7701@206.165.44.26> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-transfer-encoding: 8bit In article <AFE9608B-A7701@206.165.44.26>, "Lawson English" <english@primenet.com> wrote: > I think that all Mac developers and other interested people should meet on > AIMED-TALK > > (AIMED = Association of Independent Macintosh Engineers and Developers) > > to discuss how to handle this situation (Amelio and Hancock resigning). We > need to act very quickly to respond to this in an > Apple/Mac/Rhapsody-supportive way. > > If we can portray developers as taking a "business as usual" stance, maybe > the press will at least mention this... > > Talk about ammo for more FUD... > > We gotta be as positive (in a non-finger-pointing way) as possible about > this situation. And figure out how to convince dealers, customers and > stockholders that this is not the end of the world for Apple/Macintosh/etc. > > What happened? Lawson writing something supportive of Apple? That's bigger news than Amelio's resignation. Maybe we can get AP and Reuters to carry that news to put a positive spin on things. -- Regards, Joe Ragosta joe.ragosta@dol.net Visit the Complete Macintosh Web Site http://www.dol.net/~Ragosta/complmac.htm
Sender: Jonathan Hendry <jhendry@> From: jon@txenxe.com Subject: Re: Academic vs Developer: cheaper? Newsgroups: comp.sys.next.programmer References: <5pv24b$k18@news.cc.oberlin.edu> Message-ID: <15891725@NEWS.SAIC.COM> Date: Wed, 09 Jul 1997 22:33:11 GMT Miles Standish <cjs@occs.cs.oberlin.edu> wrote: >The adacemic bundle is only $319 but becoming an apple developer is only >$250. Any idea which is a better deal? Will I get them same (or >bettter) sutff as a develper as I get on the academic CD? And when is >Rhapsody coming out? (It is July right?) A) You cannot develop commercial applications with academic OpenStep liceneses. In other words, you can't sell what you write, even if it's just shareware B) OPENSTEP is probably a dead-end product, to be replaced with Rhapsody. (This is my impression, at least.) The developer program is a much better deal, IMHO, if you're willing to wait for Rhapdsody. -- "I am not a vegetarian because I love animals; I am a vegetarian because I hate plants." - A. Whitney Brown
From: ckoller@worldnet.att.net (Craig Koller) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Wed, 09 Jul 1997 15:31:36 -0700 Organization: MediActive Message-ID: <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> References: <AFE9608B-A7701@206.165.44.26> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit In article <AFE9608B-A7701@206.165.44.26>, "Lawson English" <english@primenet.com> wrote: > I think that all Mac developers and other interested people should meet on > AIMED-TALK > > (AIMED = Association of Independent Macintosh Engineers and Developers) > > to discuss how to handle this situation (Amelio and Hancock resigning). We > need to act very quickly to respond to this in an > Apple/Mac/Rhapsody-supportive way. > > If we can portray developers as taking a "business as usual" stance, maybe > the press will at least mention this... > > Talk about ammo for more FUD... > > We gotta be as positive (in a non-finger-pointing way) as possible about > this situation. And figure out how to convince dealers, customers and > stockholders that this is not the end of the world for Apple/Macintosh/etc. > Come on, Lawson, we all know Amelio got canned because he axed QDGX... ;@>
From: Michael Giddings <giddings_at_whitewater.chem.wisc.edu> Newsgroups: comp.sys.next.programmer Subject: PB CVS Adaptor questions Date: 9 Jul 1997 22:20:34 GMT Organization: University of Wisconsin - Madison Distribution: world Message-ID: <5q12ri$1f4g@news.doit.wisc.edu> I'm having difficulty setting up PB for CVS. I seem to be most of the way there, but the following tidbit from the very brief documentation in Librarian and NextAnswers has me confounded: "With the CVS Adaptor, you must have the CVSROOT environment variable set for Project Builder to inherit." Since ProjectBuilder is typically launched via the Workspace, I can't figure out how to do this. I suppose I could set this environment variable in a shell then always launch PB via the Terminal, but that seems kludgey. Is there some other way to inherit environment variables from the Workspace? Also, I'm wondering how this setup will deal with .nib, .tiff, and other non-text files? I downloaded CVS 1.9 from the archives (thanks Tom Hageman for putting these together). Previously I had used a version of CVS that had been modified to deal with such files by uuencoding/decoding them, but that was based on 1.3. Any info appreciated -- Michael Giddings nospam: please change "_at_" to "@" to reply, thanks giddings_at_whitewater.chem.wisc.edu giddings_at_barbarian.com (608)258-1699 or (608) 692-2851 http://smithlab.chem.wisc.edu/PersonalPages/giddings/giddings.html http://www.barbarian.com
Sender: Jonathan Hendry <jhendry@> From: jon@txenxe.com Subject: Re: Some initial OpenStep notes Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer References: <maury-0307971643370001@199.166.204.230> <paxon-0307972126420001@ppp-206-170-3-31.okld03.pacbell.net> <5pj9cf$nc$1@ironhorse.plsys.co.uk> <maury-0407971712190001@198.133.37.101> <5plifk$o83$1@ironhorse.plsys.co.uk> <maury-0707971204360001@199.166.204.230> <ECyoG5.Ep4@gateway.ali.bc.ca> <maury-0707971700430001@199.166.204.230> <5pt2dh$pru$1@ironhorse.plsys.co.uk> <maury-0907971041250001@199.166.204.230> Message-ID: <15891368@NEWS.SAIC.COM> Date: Wed, 09 Jul 1997 22:31:08 GMT Maury Markowitz <maury@softarc.com> wrote: >In article <5pt2dh$pru$1@ironhorse.plsys.co.uk>, mmalcolm crawford ><malcolm@plsys.co.uk> wrote: >> I do wonder, though, how you'd get a Ruler preview in a well? Unless you >> have a floating Panel like the current Tabs view... > I've been thinking about this. What you do is use a swatch of "greek"ed >text and changing ruler settings make the picture change. Then you just >pick up the greeked text and drop it on a paragraph. This preserves the >idea of using the well as a preview area. Let the user name them, and display a ruler in the well with the name on it. Use a default name, perhaps a letter. The user could give them a new name, like 'headline', or 'quoted text'. Actually, this opens up the possibility of storing them on disk, and having a palette of persistent rulers. People could create sets of rulers for their company documents and keep them in /LocalApps. -- "I am not a vegetarian because I love animals; I am a vegetarian because I hate plants." - A. Whitney Brown
From: gary-nospam-@screaming.org (Gary W. Longsine) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 9 Jul 1997 23:20:57 GMT Organization: Save the Skeet Foundation Message-ID: <5q16cp$abi$1@news.platinum.com> References: <AFE9608B-A7701@206.165.44.26> Cc: english@primenet.com In <AFE9608B-A7701@206.165.44.26> it appeared that "Lawson English" wrote: > I think that all Mac developers and other interested people should meet on > AIMED-TALK > > (AIMED = Association of Independent Macintosh Engineers and Developers) > > to discuss how to handle this situation (Amelio and Hancock resigning). We > need to act very quickly to respond to this in an > Apple/Mac/Rhapsody-supportive way. > > If we can portray developers as taking a "business as usual" stance, maybe > the press will at least mention this... > > We gotta be as positive (in a non-finger-pointing way) as possible about > this situation. And figure out how to convince dealers, customers and > stockholders that this is not the end of the world for Apple/Macintosh/etc. > HOW DO I SUBSCRIBE TO AIMED-TALK? > Send E-mail to: > <aimed-talk-request@aimed.org> > with a blank subject line and the following command as the > first (and only) line of the message body: > SUBSCRIBE Lawson, I could kiss you... or did you go away and leave yourself logged in again? ; ) I personally believe this is the best news since Dec. 21, 1996. /gary -- Gary W. Longsine, Systems Engineer | ____/| OpenStep MachOS PLATINUM technology, inc. | \ o.O| Objective-C l_o_n_gsine@platinum.com (NeXTmail | =(_)= the Dock (Can i have his spam?) & MIME) |. U Elegance is Relevant.
From: mvanalst@rbi.com (Mark Van Alstine) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Wed, 09 Jul 1997 16:08:30 -0800 Organization: rbi software systems Message-ID: <mvanalst-0907971608300001@rbi142.rbi.com> References: <AFE9608B-A7701@206.165.44.26> <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> In article <ckoller-ya023480000907971531360001@netnews.worldnet.att.net>, ckoller@worldnet.att.net (Craig Koller) wrote: > In article <AFE9608B-A7701@206.165.44.26>, "Lawson English" > <english@primenet.com> wrote: > > > I think that all Mac developers and other interested people should meet on > > AIMED-TALK > > > > (AIMED = Association of Independent Macintosh Engineers and Developers) > > > > to discuss how to handle this situation (Amelio and Hancock resigning). We > > need to act very quickly to respond to this in an > > Apple/Mac/Rhapsody-supportive way. > > > > If we can portray developers as taking a "business as usual" stance, maybe > > the press will at least mention this... > > > > Talk about ammo for more FUD... > > > > We gotta be as positive (in a non-finger-pointing way) as possible about > > this situation. And figure out how to convince dealers, customers and > > stockholders that this is not the end of the world for Apple/Macintosh/etc. > > > Come on, Lawson, we all know Amelio got canned because he axed QDGX... > > > > ;@> ROTFL! So cruel. };-> Mark --------------------------------------------------------------------- "Gradually it was disclosed to me that the line seperating good and evil passes not through states, nor between classes, nor between political parties--bit right through every human heart--and all human hearts." -- Alexander Solzhenitsyn, "The Gulag Archipelago" ---------------------------------------------------------------------
From: Damn Yankee<damnyankee@yankee.inc> Newsgroups: comp.sys.next.programmer Organization: Yankee Inc. Subject: I Am Very Sorry!!! Message-ID: <33c2597d.0@nntp.kalnet.net> Date: 8 Jul 97 15:15:09 GMT I would like to apologise to this newsgroup and everyone who reads this newsgroup!!! I promise never to post or send spam to this or any other newsgroup that does not pertain to my posting!!! Please accept my humble apology and again I will never post spam here again!!! Thank You!!! Andrew Schero yank714@kalnet.net
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 9 Jul 1997 21:33:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFE9B8D4-B9CF6@206.165.44.143> References: <joe.ragosta-ya02408000R0907971830490001@news.dol.net> nntp://news.primenet.com/comp.sys.mac.programmer.games, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Joe Ragosta <joe.ragosta@dol.net> said: > > > What happened? Lawson writing something supportive of Apple? That's > bigger > news than Amelio's resignation. Maybe we can get AP and Reuters to carry > that news to put a positive spin on things. When was the last time that I said "Apple Bad." I have had *SERIOUS* problems with Gil Amelio, the direction that he took Apple when he acquired NeXT, the lack of Apple Marketing, etc. But when did I ever say something bad about Apple? ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 9 Jul 1997 21:54:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFE9BDAA-CBFE5@206.165.44.143> References: <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> nntp://news.primenet.com/comp.sys.mac.programmer.games, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Craig Koller <ckoller@worldnet.att.net> said: > > Come on, Lawson, we all know Amelio got canned because he axed QDGX... Yes and no. Amelio got canned because he didn't improve Apple's bottom line and (I suspect) because Q3 results are going to be much worse than the concensus suggested: $120 million+ losses, I pontificate, maybe more. As to how this relates to GX? GX is a MacOS technology that puts everything else in its category (save possibly Taligent Graphics) to shame. Amelio's decision to not only go with the inferior Display PostScript in Rhapsody but to ALSO cripple MacOS graphics by killing GX's ability to print, is as good an example of the lack of consumer advocacy that has plagued Amelio's (and now, Steve Jobs', I fear) Apple as you can find. GX enables small developers to compete with large developers in the 2D graphics arena. OpenDOc does that in a more comprehensive way. Amelio's decision to embrace the enterprise market over Apple's traditional home/SOHO/educational markets by stagnating these home/SOHO/education-oriented technologies is EXACTLY why Apple is in the position that it is in: Amelio abandoned the customers that made Apple great and they, in turn, abandoned Apple. He did nothing to inspire the customers, the engineers, the developers or the stockholders. I mean, what can you say about Rhapsody? It is an NT-class OS that has some really nifty development tools which won't be available to the end-user for at least 6 months to a year. What can you say about GX? It is a PageMaker/FreeHand-level graphics engine that allows MacOS developers (including HyperCard programmers) to develop cool graphics solutions for MacOS, including transparently easy printing of ALL of GX transparency options, even to PS printers that is available NOW. Given equal marketing dollars, which would sell more MacOS computers and make more money for Apple? As I said, GX is a perfect example of Gil Amelio's Way of Doing Things. It is VERY distressing to hear that Steve Jobs' role is "expanded." ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: penrose@w09.sfc.keio.ac.jp Newsgroups: comp.sys.next.programmer Subject: A WINNER! NSThread event queue performance woes Date: 10 Jul 1997 14:43:18 +0900 Organization: Keio University Shonan Fujisawa Campus, Fujisawa Japan Message-ID: <wzrad78nt5.fsf@w09.sfc.keio.ac.jp> Alex Blakemore is the official winner of 100 yen for helping solve the NSThread performance riddle. He asked that the yen be donated to a worthy charity. Alex quoted the NeXTstep 3.3 release notes, mentioning that the objective-c run-time system is made thread-safe by invoking objc_setMultithreaded(BOOL value). The release notes indicate that this safety mechanism incurs a performance penalty for any objective-c message, increasing the overhead by a factor of 3, as a lock must be acquired each time a message is sent. My Openstep 4.1 didn't have these release notes or any mention of objc_setMultithreaded(BOOL value). I have thus solved my performance problems by using cthreads and disabling the runtime thread protection when the spawned thread exits. Unfortunately, NSThread has no mechanism for disabling the thread-safety features in the objective-c runtime once an application has "become multi-threaded". There is no equivalent to: objc_setMultithreaded(BOOL value). For NSThread to consider an application as irrevocably multi-threaded is a design flaw in my estimation, particularly when AppKit performance is so severely degraded by unnecessary thread protection. Using cthreads and objc_setMultithreaded(NO) I can disable thread safety features once my thread has exited. I have tested this, and it completely eliminates the performance problems I talked about. The performance problems persist only while the thread is actually executing, which is completely reasonable, and these performance problems are about 80%-90% of what I experienced with NSThread, suggesting that NSThread has some other overhead as well. If anyone at Apple cares to comment, it would be great. I suggest implementing a method in NSThread such as setMultiThreaded:(BOOL) aValue such that multi-threaded applications are not crippled for life once they use a thread. There may be reasons that this feature is not present in Openstep, such as the nature of thread implementations on other platforms. At the very least, Apple could add this as a simple Openstep extension. It was very easy to implement myself with cthreads. Christopher Penrose Audio DSP Researcher penrose@sfc.keio.ac.jp
From: arti@address.in.signature (Art Isbell) Newsgroups: comp.sys.next.programmer Subject: Re: Lex/Yacc in 4.1 PB Date: 10 Jul 1997 08:44:11 GMT Organization: LavaNet, Inc. Distribution: world Message-ID: <5q27cs$nr8@mochi.lava.net> References: <5pug06$iml$1@anvil.BLaCKSMITH.com> rmessner@tolkien.BLaCKSMITH.com (Rick Messner) wrote: > I have lex and yacc files I cannot get to compile in 4.1 PB. It seems as > if the extension is not being recognized, but yet it is listed in the > basicrules.make file. I have tried .lm, .lm.m, and .l.c and corresponding > yacc extensions to no avail. Any help would be appreciated. Thanks, The lex/yacc implicit rules in implicitrules.make are incomplete is OS 4.1. They seem to be fixed in OS 4.2. I added the following to $(LOCAL_MAKEFILEDIR)/common.make.postamble: .SUFFIXES: .ym .lm .h .m # When both LEX and YACC will be executed and the LEX source imports a header # file produced by YACC, add a dependencies to Makefile.postamble that force # YACC to execute before LEX: # lexer.m: parser.h # parser.h: parser.m # If .ym.h is also a target, the rule fires twice. .ym.m: $(CD) $(SFILE_DIR) && $(YACC) $(ALL_YFLAGS) $(shell pwd)/$< $(MV) $(SFILE_DIR)/y.tab.h $(SFILE_DIR)/$*.h $(MV) $(SFILE_DIR)/y.tab.c $(SFILE_DIR)/$*.m .lm.m: $(CD) $(SFILE_DIR) && $(LEX) $(ALL_LFLAGS) $(shell pwd)/$< $(MV) $(SFILE_DIR)/lex.yy.c $(SFILE_DIR)/$*.m Also, the 4.1 common.make includes a couple of DOS end-of-line characters (extra Carriage Return characters) that prevent LOCAL_MAKEFILEDIR from being defined :-( Open common.make in your favorite binary editor (emacs is a great choice) and remove these extraneous Carriage Returns (printed as ^M) to make things work properly. -- Art Isbell NeXT/MIME Mail: arti at lava dot net Trego Systems (for whom I don't speak) Voice/Fax: +1 808 394 0511 OPENSTEP/NT Voice Mail: +1 808 394 0495 managed care solutions US Mail: Honolulu, HI 96825-2638
From: Steve Watt <steve@watt.com.nospam> Organization: USENET spam abatement Sender: Damn Yankee<damnyankee@yankee.inc> Date: 10 Jul 97 07:44:58 GMT Message-ID: <cancel.33c2597d.0@nntp.kalnet.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33c2597d.0@nntp.kalnet.net> Control: cancel <33c2597d.0@nntp.kalnet.net> I have cancelled this article which had a BI of more than 20. Spam is spam, even when it's an apology. The original subject was: }Subject: I Am Very Sorry!!!
From: Steve Watt <steve@watt.com.nospam> Organization: USENET spam abatement Sender: Kris Jacobs<jtsnake@serv01.net-link.net> Date: 10 Jul 97 07:49:22 GMT Message-ID: <cancel.33c1fb15.2@nntp.kalnet.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33c1fb15.2@nntp.kalnet.net> Control: cancel <33c1fb15.2@nntp.kalnet.net> I have cancelled this article which had a BI of more than 20. The original subject was: }Subject: Try Me!
From: izidor.jerebic@select-tech.si Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Openstep examples OO design (was Re: Some initial OpenStep notes) Date: 10 Jul 1997 08:36:30 GMT Organization: SELECT Technology Message-ID: <5q26ue$ndb$1@lazar.select-tech.si> References: <5pmd4m$i21$2@news.xmission.com> In article <5pmd4m$i21$2@news.xmission.com> don@globalobjects.com (Don Yacktman) writes: > > Yes, it has a lot of neat stuff, but there is one very important thing it > does very BADLY. It has a terrible OO design, IMHO. As someone who once > made *extensive* modifications to build a whole new app, I will say that by > the time I was done, I had redesigned practically everything. The design in > Draw is far too cumbersome to extend easily, so I ended up replacing one If you compare Draw in Nextstep 3.x and Draw in 4.x you can see that the Draw in 4.x is completely new code and very improved. Actually, it conforms to ModelViewController OO design. The original Draw was one big hack, really. But the new one is rather good, as are all other new (4.x) examples, IMHO. I must confess I did not do any modification of the example code. I just briefly went through the code of examples and it looked OK to me (old examples were just plain hacks). I even recommended them to people as nice little examples of coding AND design. I'm really interested what are the big deficiencies you found in new examples, e.g. Draw? Are there any practices advised in there that we should definitely not follow? izidor
From: drifterusa@macconnect.com (John Bauer) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Thu, 10 Jul 1997 02:37:52 -0500 Organization: None Message-ID: <199707100237521485904@accs-as809-dp07.atln.grid.net> References: <AFE9608B-A7701@206.165.44.26> <5q16cp$abi$1@news.platinum.com> Gary W. Longsine wrote: > I personally believe this [Amelio and Hancock's departure] is the best > news since Dec. 21, 1996. Could you explain why? John Bauer
From: *johnnyc*@or.psychology.dal.ca (John Christie) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Thu, 10 Jul 1997 09:08:01 -0400 Organization: ISINet, Nova Scotia Message-ID: <*johnnyc*-1007970908020001@192.0.2.1> References: <AFE9608B-A7701@206.165.44.26> <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> In article <ckoller-ya023480000907971531360001@netnews.worldnet.att.net>, ckoller@worldnet.att.net (Craig Koller) wrote: > Come on, Lawson, we all know Amelio got canned because he axed QDGX... And I thought it was because Markkula owns a big piece of BBDO. -- You aren't free if you CAN choose - only if you DO choose. All you are is the decisions you make. Remove "*" and ohnny (i.e. jc@) to reply via email
From: mmalcolm crawford <malcolm@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 10 Jul 1997 13:09:29 GMT Organization: P&L Systems Message-ID: <5q2mu9$ppq$1@ironhorse.plsys.co.uk> References: <AFE9608B-A7701@206.165.44.26> In-Reply-To: <AFE9608B-A7701@206.165.44.26> <AD_HOMINEM> On 07/10/97, "Lawson English" wrote: >I think that all Mac developers and other interested people should meet on >AIMED-TALK > >(AIMED = Association of Independent Macintosh Engineers and Developers) > >to discuss how to handle this situation (Amelio and Hancock resigning). We >need to act very quickly to respond to this in an >Apple/Mac/Rhapsody-supportive way. > Yeah, and you're exactly the person to respond in a positive and supportive way... >If we can portray developers as taking a "business as usual" stance, maybe >the press will at least mention this... > Hmm, yes, business as usual: DPS SUKZ GX ROOLZ BRING ME THE HEAD OF GIL AMELIO >We gotta be as positive (in a non-finger-pointing way) as possible about >this situation. And figure out how to convince dealers, customers and >stockholders that this is not the end of the world for Apple/Macintosh/etc. > Some of us have been being positive for several months. </AD_HOMINEM> Best wishes, mmalc. --
From: maury@softarc.com (Maury Markowitz) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Some initial OpenStep notes Date: Thu, 10 Jul 1997 10:52:47 -0400 Organization: SoftArc Inc. Message-ID: <maury-1007971052470001@199.166.204.230> References: <maury-0307971643370001@199.166.204.230> <paxon-0307972126420001@ppp-206-170-3-31.okld03.pacbell.net> <5pj9cf$nc$1@ironhorse.plsys.co.uk> <maury-0407971712190001@198.133.37.101> <5plifk$o83$1@ironhorse.plsys.co.uk> <maury-0707971204360001@199.166.204.230> <ECyoG5.Ep4@gateway.ali.bc.ca> <maury-0707971700430001@199.166.204.230> <5pt2dh$pru$1@ironhorse.plsys.co.uk> <maury-0907971041250001@199.166.204.230> <15891368@NEWS.SAIC.COM> In article <15891368@NEWS.SAIC.COM>, jon@txenxe.com wrote: > Let the user name them, and display a ruler in the well with the name > on it. Use a default name, perhaps a letter. The user could give them > a new name, like 'headline', or 'quoted text'. > > Actually, this opens up the possibility of storing them on disk, and > having a palette of persistent rulers. People could create sets of > rulers for their company documents and keep them in /LocalApps. Yeah, and better yet they would work in most programs too. This is what I hate on the Mac, I've got something like four spell checkers (I only use one though) and god knows how many programs that have style sheets, yet I can't share any of it. Quoted text might be another issue, unless we adopt a standard like we use in our product. Maury
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: GDB and NSProxies... no good friends Date: 10 Jul 1997 15:09:24 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5q2tv4$c6i$1@concorde.ctp.com> Hi, is it a known "goodie" that you can not work with GDB using proxy objects ? e.g.: po [myServer description] works and reutrns the servers class (while isproxy reutns 1.. so we really have a distant object) but po [myserver doFunkyGeekySound] does not work since gdb complains the "Target does not respond to method" even if it really is implementing it. Is there a known workaround ? Since I suspect gdb does just check the mehtdos which are really implemented by the local class. Shouldn't it behave differently for NXProxies ? Aloha Tomi
From: acurylo@inmediapresents.nospam.com (Alex Curylo) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 10 Jul 1997 16:37:55 GMT Organization: InMedia Presentations Message-ID: <acurylo-1007970942180001@van0209.tvs.net> References: <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> <AFE9BDAA-CBFE5@206.165.44.143> In article <AFE9BDAA-CBFE5@206.165.44.143>, "Lawson English" <english@primenet.com> wrote: > Yes and no. Amelio got canned because he didn't improve Apple's bottom line > and (I suspect) because Q3 results are going to be much worse than the > concensus suggested: $120 million+ losses, I pontificate, maybe more. Heh. Try $300 million. #include <stddisclaimer.h> > possibly Taligent Graphics) to shame. Amelio's decision to not only go with > the inferior Display PostScript in Rhapsody but to ALSO cripple MacOS > graphics by killing GX's ability to print, is as good an example of the > lack of consumer advocacy that has plagued Amelio's (and now, Steve Jobs', > I fear) Apple as you can find. Didn't we go over this before? The major source of actual dollars for Apple in the COMPETITIVE marketplace is publishing. Publishers of all sorts go all glassy-eyed and drooling at the mere thought of being able to work in Display PostScript, while they scatter screaming at the mere mention of GX. So a realistic look at Apple's actual marketplace leads one inescapably to the conclusion that DPS is the truly consumer advocating position. QED. > He did nothing to inspire the customers, the engineers, the developers or > the stockholders. Hmmmm...I fall into three of those four categories, and Rhapsody inspired me :) ---- Alex Curylo -- alex@witty.com
From: ryno@teleport.com (Mitchell J Laurren-Ring) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Thu, 10 Jul 1997 10:16:46 -0700 Organization: La Cie Limited Message-ID: <ryno-1007971016470001@ip-pdx16-08.teleport.com> References: <AFE9608B-A7701@206.165.44.26> <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> In article <ckoller-ya023480000907971531360001@netnews.worldnet.att.net>, ckoller@worldnet.att.net (Craig Koller) wrote: > In article <AFE9608B-A7701@206.165.44.26>, "Lawson English" > <english@primenet.com> wrote: > > > I think that all Mac developers and other interested people should meet on > > AIMED-TALK > > > > (AIMED = Association of Independent Macintosh Engineers and Developers) > > > > to discuss how to handle this situation (Amelio and Hancock resigning). We > > need to act very quickly to respond to this in an > > Apple/Mac/Rhapsody-supportive way. > > > > If we can portray developers as taking a "business as usual" stance, maybe > > the press will at least mention this... > > > > Talk about ammo for more FUD... > > > > We gotta be as positive (in a non-finger-pointing way) as possible about > > this situation. And figure out how to convince dealers, customers and > > stockholders that this is not the end of the world for Apple/Macintosh/etc. > > > Come on, Lawson, we all know Amelio got canned because he axed QDGX... > > > > ;@> Uh oh! You just gave him a virtual dime... -Mick -- Mitchell J Laurren-Ring La Cie Limited
From: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.next.programmer Subject: Re: A WINNER! NSThread event queue performance woes Date: 10 Jul 1997 13:04:04 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5q34m4$b9r$1@sps1.phys.vt.edu> References: <wzrad78nt5.fsf@w09.sfc.keio.ac.jp> In article <wzrad78nt5.fsf@w09.sfc.keio.ac.jp>, penrose@w09.sfc.keio.ac.jp wrote: > Alex quoted the NeXTstep 3.3 release notes, mentioning that the > objective-c run-time system is made thread-safe by invoking > objc_setMultithreaded(BOOL value). The release notes indicate > that this safety mechanism incurs a performance penalty for any > objective-c message, increasing the overhead by a factor of 3, as > a lock must be acquired each time a message is sent. > If anyone at Apple cares to comment, it would be great. I suggest > implementing a method in NSThread such as > setMultiThreaded:(BOOL) aValue > such that multi-threaded applications are not crippled for life once > they use a thread. Or, Apple could use a better thread-safe runtime. IIRC, the gcc runtime does not suffer from that 3x performance hit (caused by throwing a mutex around the whole runtime, basically). (I also seem to recall that NeXT had fixed this.. but I looked at the 4.x release notes and didn't see anything about it. Was I hallucinating?)
From: Alex Blakemore <alex@genoa.com> Newsgroups: comp.sys.next.programmer Subject: Re: CONTEST! solve the NSThread event queue performance bug and win! Date: 10 Jul 1997 02:36:33 GMT Organization: Genoa Software Systems Message-ID: <5q1hrh$gar@saturn.genoa.com> References: <wz7mf02lxu.fsf@w09.sfc.keio.ac.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: penrose@w09.sfc.keio.ac.jp In <wz7mf02lxu.fsf@w09.sfc.keio.ac.jp> penrose@w09.sfc.keio.ac.jp wrote: > 100 Yen goes to the first cogent reply to my earlier inquiries regarding > the use of NSThread and NSEvent queue performance degradation. here's a hint from the old NS3.3 Objective C release notes. I don't know how much this applies to NSThread, but it probably still does. P.S. From experience, if you have mach threads without making the run time system thread safe, you will get random crashes. (if by some miracle this is the best answer you get, give the yen to charity please) Making Objective C Thread Safe The Objective C run-time system has been made safe for use in multi-threaded programs. Since complete thread-safety requires that a lock be acquired every time a message is sent (which increases the time required to send a message by a factor of approximately three), the thread safety features must be explicitly enabled using the new function objc_setMultithreaded(). -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: dbriggs@stem.com Newsgroups: comp.sys.next.programmer Subject: Draw.app (Re: Openstep examples OO design) Date: 10 Jul 1997 18:06:10 GMT Organization: Systemix, Inc. Message-ID: <5q38ai$kfh@iserver.stem.com> Hi, all -- About Draw.app as an example of Openstep OO design: In article <5pmd4m$i21$2@news.xmission.com> don@globalobjects.com (Don Yacktman) writes: > ... The design in Draw is far too cumbersome to extend easily, I agree. And izidor.jerebic@select-tech.si wrote: > I'm really interested what are the big deficiencies you found in > new examples, e.g. Draw? Are there any practices advised in there > that we should definitely not follow? Good question -- getting started on the right path is very important. I've found the examples generally very helpful, especially Draw.app. However, in my opinion, the developer community would be better served if Draw.app were repackaged as an example framework, served up with documentation, and had some bugs fixed. Also, I can identify a practice in Draw.app that does not scale well. There's a file called "draw.h" that imports "everything". A newbie building an application based on Draw.app might be tempted to extend draw.h to import all the important additional *.h files. Don't! You'll lose a kind of modularity. It becomes "cumbersome" rapidly: instead of #import "Banana.h" it'll be #import "WholeGorilla.h" instead. I recommend that you follow the more typical structure found in the other examples. One last point: Draw.app as it stands fails to "do the right thing" in some simple cases. Try this: [1] Fire up Draw.app. [2] Draw two rectangles; draw a large one, then a smaller one below the first. They shouldn't overlap. [3] Move the smaller rectangle inside the bounds of the larger one. [4] Drag-select both rectangles. Draw.app is now confused; try to move the selected rectangles. Pfft! In order to use Draw.app as a basis, we repackaged it as a framework and fixed a few such bugs (J. Renato Galvez). Don Briggs <standard disclaimer>
From: randy97 Subject: http://www.love.com Newsgroups: comp.sys.next.programmer Message-ID: <33c52b4f.1@pgh.nauticom.net> Date: 10 Jul 97 18:34:55 GMT Looking to find people in your area that enjoy the same things as this newsgroup? Check out http://www.love.com It's free, it's new, and it's awesome. Rand
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Thu, 10 Jul 1997 15:26:43 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <4nlHRna00iWW05apI0@andrew.cmu.edu> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <knkbVMS00iWW0465o0@andrew.cmu.edu> <199707092338411585593@dialup96-5-11.swipnet.se> In-Reply-To: <199707092338411585593@dialup96-5-11.swipnet.se> Excerpts from netnews.comp.sys.next.programmer: 9-Jul-97 Re: [Q] OpenStep Garbage Co.. by Lars Farm@ite.mh.se >> That has not been my argument-- I understand what GC is, and that it can >> be very nice under some circumstances. The problem is that you don't >> seem to understand that GC is not a panacea, and that it is not a >> solution to all the world's programming ills. > > Interesting quoting technique and similarly interesting deduction. I > said "useful tool, but no silver bullet". From that you deduce that > "useful tool, but no panacea" is beyond my grasp... Hmmm... Please > reread the rest of the quoted paragraph (snipped by you) and the next > paragraph (snipped by you). Compare to what you said above. Oh, I read it. The problem is, your earlier arguments were explicitly ignoring issues like VM and you have done a remarkable job of not discussing substantive issues such as the tradeoffs involved with conservative GC. If you were willing to describe your experiences with GC in a quantitative fashion-- ie, in this project we switched from using explicit malloc()/free() combinations to using a conservative Boehm GC, and the running time changed from T1 second to T2 seconds, and the average memory usage switched from a RSIZE of M1 megabytes to M2 megabytes, then this discussion would be much more productive. As it stands now, you appear to be determined to misinterpret my comments as a blanket criticism of GC when instead I am trying to clearly identify the tradeoffs encountered when using GC, and I don't see any value to continuing such a debate. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Michael Giddings <giddings_at_whitewater.chem.wisc.edu> Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: CVS unusable with 4.2 PB/Mach? Strange path mapping bug . . . Date: 10 Jul 1997 20:48:31 GMT Organization: University of Wisconsin - Madison Distribution: world Message-ID: <5q3hqv$1ifo@news.doit.wisc.edu> Has anyone else gotten the SCM module for CVS to work correctly with mach? I've run into what seems like a major bug. In trying to "Create Work Directory" under OS 4.2/mach, Project builder converts all pathnames I enter under the field "Project's Local Work Directory" and makes them into Windoze style paths, by making the substitution "/ -> \". It actually checks out a directory at root with a name such as \joes\garage when I give it /joes/garage, but then, in the last phase of creating this work directory, it bombs because it goes back and looks for the original /joes/garage to make some modifications, which was never created in the first place! I haven't yet found a way around this, and it seemingly makes the SCM on PB for Mach useless. Has anyone else seen this? Please replace "_at_" with @ to respond via e-mail (to help me avoid spam), thanks. -- Michael Giddings giddings_at_chem.wisc.edu giddings_at_barbarian.com (608)258-1699 or (608) 692-2851 http://smithlab.chem.wisc.edu/PersonalPages/giddings/giddings.html http://www.barbarian.com
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 10 Jul 1997 17:44:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFEAD48E-1812E9@206.165.44.28> References: <acurylo-1007970942180001@van0209.tvs.net> nntp://news.primenet.com/comp.sys.mac.programmer.games, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Alex Curylo <acurylo@inmediapresents.nospam.com> said: > Didn't we go over this before? The major source of actual dollars for > Apple in the COMPETITIVE marketplace is publishing. Publishers of all > sorts go all glassy-eyed and drooling at the mere thought of being able to > work in Display PostScript, while they scatter screaming at the mere > mention of GX. So a realistic look at Apple's actual marketplace leads one > inescapably to the conclusion that DPS is the truly consumer advocating > position. QED. Could it possibly be because Apple did an absolutely horrible job marketing GX? What if GX could be marketed properly? ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: "Michael J. Peck" <mjpeck@intex.net> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Thu, 10 Jul 1997 20:16:14 -0500 Organization: Intex.Net (http://www.intex.net/) Message-ID: <33C5895E.7A35627C@intex.net> References: <acurylo-1007970942180001@van0209.tvs.net> <AFEAD48E-1812E9@206.165.44.28> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lawson English wrote: > > Didn't we go over this before? The major source of actual dollars for > > Apple in the COMPETITIVE marketplace is publishing. Publishers of all > > sorts go all glassy-eyed and drooling at the mere thought of being able > to > > work in Display PostScript, while they scatter screaming at the mere > > mention of GX. So a realistic look at Apple's actual marketplace leads > one > > inescapably to the conclusion that DPS is the truly consumer advocating > > position. QED. > > Could it possibly be because Apple did an absolutely horrible job marketing > GX? > > What if GX could be marketed properly? I'm inclined to believe that GX was a good technology that was never given a chance. All exasperation aside over flogging a dead horse, from what I know (very little) about GX, it was a good technology that, because of its newness and lack of supporters (Apple is a weak supporter of anything), was buried in the onslaught of new technologies. DPS had/has a much stronger market position by far, and no matter how much better GX was, it never had a real chance to show what it was worth, especially amid all the snide jeers and jokes about memory requirements and such. MJP -- "Now you see, Lone Star, that evil will always triumph, because good is dumb." - Dark Helmet
From: dcoyle@ctp.com (David A. Coyle) Newsgroups: comp.sys.next.programmer Subject: NSTableView doesn't appear to know it's selected row Date: 10 Jul 1997 23:57:27 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5q3st7$les$1@concorde.ctp.com> Hi all: I'm having real trouble with an NSTableView. My tableView delegate, which also has an ivar pointing to the NSTV, implements the delegate method: - (void)tableViewSelectionDidChange:(NSNotification*)aNotification Cool. Fills with data, looks good... When I ask the notification's object for the selected row, it returns the correct selected row. HOWEVER if, anywhere else in this class, I attempt to get the selected row using [myIvarForTheNSTV selectedRow] I ALWAYS get 0. but I can, for example, check that my ivar really does point to that same NSTV (via myIvarForTheNSTV == [aNotification object] in aforementioned method) BUT WAIT! it gets stranger!. I create int myHackIndexIvar; and do myHackIndexIvar = [[aNotification object] selectedRow]; in that delegate method, then it is correctly set. But! when I try to use that ivar anywhere else, it has mysteriously become 0. Also, for what it's worth, this is all in a bundle, and thus a real pain to debug with gdb...( NSLog() debugging ) Any ideas? Dave ------- /\/\ Dave Coyle <dcoyle@ctp.com> / /_ \ Cambridge Technology Partners \ / / 118/119 Lower Baggot Street \/\/ Dublin 2, Republic of Ireland Tel: +353 1 607 9008 WWW: http://www.ctp.com>
From: LTaylor7@ix.netcom.com (Daniel L. Taylor) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Thu, 10 Jul 1997 18:52:59 -0700 Organization: Netcom Message-ID: <LTaylor7-1007971853010001@vic-ca2-13.ix.netcom.com> References: <acurylo-1007970942180001@van0209.tvs.net> <AFEAD48E-1812E9@206.165.44.28> In article <AFEAD48E-1812E9@206.165.44.28>, "Lawson English" <english@primenet.com> wrote: > > Didn't we go over this before? The major source of actual dollars for > > Apple in the COMPETITIVE marketplace is publishing. Publishers of all > > sorts go all glassy-eyed and drooling at the mere thought of being able > to > > work in Display PostScript, while they scatter screaming at the mere > > mention of GX. So a realistic look at Apple's actual marketplace leads > one > > inescapably to the conclusion that DPS is the truly consumer advocating > > position. QED. > > Could it possibly be because Apple did an absolutely horrible job marketing > GX? > > What if GX could be marketed properly? Right now Windows and DPS dominate their respective markets. May I suggest that we focus on competition with WINDOWS and just accept DPS since the former is our true competitive enemy and the latter could help us fight the former? Please can we focus? PLEASE? -- Daniel L. Taylor LTaylor7@ix.netcom.com ================================================================= The Heritage Foundation /// Town Hall http://www.heritage.org/heritage/ /// http://www.townhall.com/ =================================================================
From: Alex Blakemore <alex@genoa.com> Newsgroups: comp.sys.next.programmer Subject: How can you turn on/off the Windows NT wait cursor? Date: 11 Jul 1997 00:20:33 GMT Organization: Genoa Software Systems Message-ID: <5q3u8h$mt@saturn.genoa.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Does anyone know how to turn on and off the wait cursor under OS/NT4.2? It no longer happens automatically, but I heard you could make some function calls to some Microsoft stuff - say bracketed a long operation. Thanks in advance for any help. -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: Ian Russell Ollmann <iano@scripps.edu> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Thu, 10 Jul 1997 20:12:16 -0700 Organization: The Scripps Research Institute, La Jolla, CA Message-ID: <Pine.SGI.3.96.970710201114.13965B-100000@wong> References: <AFE9608B-A7701@206.165.44.26> <joe.ragosta-ya02408000R0907971830490001@news.dol.net> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <joe.ragosta-ya02408000R0907971830490001@news.dol.net> On Wed, 9 Jul 1997, Joe Ragosta wrote: > What happened? Lawson writing something supportive of Apple? That's bigger > news than Amelio's resignation. Maybe we can get AP and Reuters to carry > that news to put a positive spin on things. It would only be bigger news if Joe Ragosta wrote something unsupportive of Apple. :-) That would mean invoking the emergency broadcast system, I believe. Ian
From: "Frank Alviani" <alviani@ix.netcom.com> Newsgroups: comp.sys.next.programmer Subject: Prelude To Rhapsody - ToDo example application complaints Date: 10 Jul 97 20:43:25 -0500 Organization: Netcom Message-ID: <AFEAF9F1-43432@205.186.163.53> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Hello All, I am working through the examples in the P2R tutorial. While I found the first 2 examples quite adequate, I am pretty upset with the ToDo example. It seems more an example of bad editorial QA than a usable tutorial example. - The installed example code, compiled with no modifications, crashes due to a rather silly bug; I had already found it in the code I typed in from the book. - There are a significant number of differences between the files installed and the material in the book. Methods are overridden in the installed files, for one, that aren't even mentioned in the tutorial text. -Inter-object connections are made in the provided example that I cannot figure out. The connections, when viewed in IB, seem identical with those I have manually created. This is more obfuscation than clarification. Has anyone else run into this sort of thing? I have many years of OO programming experience in Object Pascal and C++, with and without frameworks, and consider myself pretty competent. I like Objective-C and the entire OpenStep system. But this is very bad! --------------------------------------------------- This message was created and sent using the Cyberdog Mail System ---------------------------------------------------
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 11 Jul 1997 06:09:09 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5q4im5$cs1$1@brachio.zrz.TU-Berlin.DE> References: <AFEAD48E-1812E9@206.165.44.28> In article <AFEAD48E-1812E9@206.165.44.28> "Lawson English" <english@primenet.com> writes: > Alex Curylo <acurylo@inmediapresents.nospam.com> said: > > > Didn't we go over this before? The major source of actual dollars for > > Apple in the COMPETITIVE marketplace is publishing. Publishers of all > > sorts go all glassy-eyed and drooling at the mere thought of being able > to > > work in Display PostScript, while they scatter screaming at the mere > > mention of GX. So a realistic look at Apple's actual marketplace leads > one > > inescapably to the conclusion that DPS is the truly consumer advocating > > position. QED. > > Could it possibly be because Apple did an absolutely horrible job marketing > GX? Nope. > What if GX could be marketed properly? Would have had very little impact on the marketplace. Now it is certainly too late. Marcel
From: arti@address.in.signature (Art Isbell) Newsgroups: comp.sys.next.programmer Subject: Re: GDB and NSProxies... no good friends Date: 11 Jul 1997 06:15:45 GMT Organization: LavaNet, Inc. Distribution: world Message-ID: <5q4j2h$mgt@mochi.lava.net> References: <5q2tv4$c6i$1@concorde.ctp.com> tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) wrote: > is it a known "goodie" that you can not work with GDB using proxy objects ? Unfortunately, this has been the case since D.O. was introduced. Maybe there's a basic problem making gdb work with proxies. I reported this as a bug to NeXT quite some time ago but nothing has changed (and maybe it's just not possible). -- Art Isbell NeXT/MIME Mail: arti at lava dot net Trego Systems (for whom I don't speak) Voice/Fax: +1 808 394 0511 OPENSTEP/NT Voice Mail: +1 808 394 0495 managed care solutions US Mail: Honolulu, HI 96825-2638
From: luomat@peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.programmer Subject: OpenStep command to make appwrapper out of shell script Date: 11 Jul 1997 06:18:50 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Message-ID: <5q4j8a$9fd$1@ha2.rdc1.nj.home.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii I'm gonna keep looking for this til I understand what I thought I heard, but anyway.... I thought I heard someone say that OpenStep had a command of some kind that would take a shell script and make it an app wrapper. That would be mighty handy. Does anyone know what I am talking about? TjL -- TjL <luomat@peak.org>
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: OpenStep command to make appwrapper out of shell script Date: 11 Jul 1997 07:48:41 GMT Organization: WolfWare Message-ID: <5q4ogp$65s$1@vader.wolfware.ipc.net> References: <5q4j8a$9fd$1@ha2.rdc1.nj.home.com> In-Reply-To: <5q4j8a$9fd$1@ha2.rdc1.nj.home.com> On 07/10/97, Timothy J. Luoma wrote: > >I'm gonna keep looking for this til I understand what I thought I heard, but >anyway.... > >I thought I heard someone say that OpenStep had a command of some kind that >would take a shell script and make it an app wrapper. That would be mighty >handy. Does anyone know what I am talking about? > No. I think what you heard was me saying that I had to manually turn some commands (shell scripts) into 1-line-of-code apps in order to wrap them together with a unique icon. Theoretically someone could create a utility to do this without requiring a recompile... would require a template app which looked for a shell script of a specific (hard-coded) name within the app wrapper. The utility could use the command line tools for manipulating Mach-O to stick an appropriate image in the app executable for the icon and then rename the app wrapper and the executable appropriately. >TjL > >-- >TjL <luomat@peak.org> > > -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: penrose@w09.sfc.keio.ac.jp Newsgroups: comp.sys.next.programmer Subject: Re: A WINNER! NSThread event queue performance woes Date: 11 Jul 1997 17:41:26 +0900 Organization: Keio University Shonan Fujisawa Campus, Fujisawa Japan Message-ID: <wzpvsq0ymh.fsf@w09.sfc.keio.ac.jp> References: <wzrad78nt5.fsf@w09.sfc.keio.ac.jp> <5q34m4$b9r$1@sps1.phys.vt.edu> nurban@sps1.phys.vt.edu (Nathan Urban) said: "Or, Apple could use a better thread-safe runtime. IIRC, the gcc runtime does not suffer from that 3x performance hit (caused by throwing a mutex around the whole runtime, basically). (I also seem to recall that NeXT had fixed this.. but I looked at the 4.x release notes and didn't see anything about it. Was I hallucinating?)" Unfortunately, you were probably hallucinating, unless the problem is fixed for 4.2 as I am developing in 4.1. I will reiterate for Apple's sake and hope that redundant discussion will spark a semi-official statement from Apple: Using NSThread spawned threads caused me a huge performance hit with the NSEvent queue. Mouse dragging was ridiculously slow and sluggish. The same performance problems exist with my mach cthread implementation only while objc_setMultithreaded(YES) is in effect, but the performance seems degradation is slightly less severe than NSThread (80-90%). The performance problems are not present with objc_setMultithreaded(NO). NSThread does not give similar developer access to the runtime thread safety as objc_setMultithreaded(). I suggest to Apple that they should, or as Nathan suggests, they improve the performance of the run-time thread-safety mechanism. Christopher Penrose penrose@sfc.keio.ac.jp
From: Wolfgang Baron <wbaron@ixpoint.de> Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: Re: CVS unusable with 4.2 PB/Mach? Strange path mapping bug . . . Date: 11 Jul 1997 08:58:03 GMT Organization: iXpoint Informationssysteme GmbH, www.ixpoint.de Distribution: world Message-ID: <5q4sir$24o$1@ixpoint.de> References: <5q3hqv$1ifo@news.doit.wisc.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Cc: giddings@whitewater.chem.wisc.edu In <5q3hqv$1ifo@news.doit.wisc.edu> Michael Giddings wrote: > Has anyone else gotten the SCM module for CVS to work correctly with mach? > I've run into what seems like a major bug. > > In trying to "Create Work Directory" under OS 4.2/mach, Project builder > converts all pathnames I enter under the field "Project's Local Work > Directory" and makes them into Windoze style paths, by making the > substitution "/ -> \". I came accross another problem, trying to set the CVSROOT variable for cvs (as described in the manual) to use under ProjectBuilder. I have tried all the startup scripts for users and root. I do not seem to be able to set this environment variable for any application, as applications are run by the workspace, which is run by the loginWindow and has never used a shell on the way. How can I avoid this problem? -- +++++++++++++++++++++++++++++++++++++++++++++++++ + Dipl.- Inform. Wolfgang Baron + + iXpoint Informationssysteme GmbH + + Daimlerstrasse 3, 76275 Ettlingen, Germany + + phone ++49 7243 3775-82 Fax ++49 7243 3775-77 + + email: wbaron@ixpoint.de (NeXTmail and MIME) + +++++++++++++++++++++++++++++++++++++++++++++++++ + pConnector - Network Printing for NEXTSTEP + + sca - the static code analyser + +++++++++++++++++++++++++++++++++++++++++++++++++
From: tiggr@ics.ele.tue.nl (Pieter Schoenmakers) Newsgroups: comp.sys.next.programmer,comp.lang.objective-c Subject: Re: ISO Objective C interface file grammar Date: 11 Jul 1997 12:27:00 +0200 Organization: Eindhoven University of Technology Sender: tiggr@tom.ics.ele.tue.nl Message-ID: <x7k9ixvq8b.fsf@tom.ics.ele.tue.nl> References: <5q3ucg$mt@saturn.genoa.com> In-reply-to: Alex Blakemore's message of 11 Jul 1997 00:22:40 GMT In article <5q3ucg$mt@saturn.genoa.com> Alex Blakemore <alex@genoa.com> writes: Does anyone no of an available yacc/bison grammar for Objective-C? If the GPL suits you, GCC comes with one. We just need to parse header files. Is a header file any simpler than a source file? --Tiggr
Newsgroups: comp.sys.next.programmer From: renkes@OneVision.de (Frank Renkes) Subject: Re: How to split classes hierarhy to bundles Message-ID: <ED5Dv3.Hw9@onevision.de> Sender: news@onevision.de Organization: OneVision Vertriebs-GmbH, Regensburg, Germany References: <33C0F8E2.273C@rtr.transit.ru> Date: Fri, 11 Jul 1997 09:18:39 GMT If you just export classes in your bundle you can use the "-suppress undefined" flag for the linker. Then you can ignore the warning for unresolved classes. Frank In article <33C0F8E2.273C@rtr.transit.ru> Iracly <iracly@rtr.transit.ru> writes: > Is anybody have idea how to divide class hierarhy to different loadable > bundles. Under Windows I can just declare class as __export/__import and > then link any other DLL (even with derived classes) with import library. > Is some analogues under OPENSTEP? When I just try to compile "derived > bundle" I got UNRESOLVED EXTERNAL ... message from linker, If I make > some dummy implementation of base class in derived bundle, than after I > load derived bundle, the original base class was replased with dummy > implementation from second one! > > PS I use OPENSTEP developer v4.2 prerelease 2 for NT > > Regards > > Iracly (iracly@com2com.ru or iracly@rat.radio-msu.net)
From: fpottier@pauillac.inria.fr (Francois Pottier) Newsgroups: comp.sys.next.programmer Subject: Re: CONTEST! solve the NSThread event queue performance bug and win! Date: 11 Jul 1997 13:46:52 GMT Organization: INRIA Rocquencourt, BP 105, 78153 Le Chesnay Cedex, France Message-ID: <5q5dgc$pb8@news-rocq.inria.fr> References: <wz7mf02lxu.fsf@w09.sfc.keio.ac.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit In article <wz7mf02lxu.fsf@w09.sfc.keio.ac.jp>, <penrose@w09.sfc.keio.ac.jp> wrote: > >100 Yen goes to the first cogent reply to my earlier inquiries regarding >the use of NSThread and NSEvent queue performance degradation. I don't have the answer (though I'm interested, since I'm designing an OpenStep app which will eventually be multi-threaded), but I'd like to commend you for this innovative way of getting advice on Usenet! Cheers, -- François Pottier Francois.Pottier@inria.fr http://pauillac.inria.fr/~fpottier/
From: dcoyle@ctp.com (David A. Coyle) Newsgroups: comp.sys.next.programmer Subject: Stupidity solved! Re: NSTableView ... Date: 11 Jul 1997 13:59:23 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5q5e7r$7cv$1@concorde.ctp.com> References: <5q3st7$les$1@concorde.ctp.com> In article <5q3st7$les$1@concorde.ctp.com> dcoyle@ctp.com (David A. Coyle) writes: > Hi all: > > I'm having real trouble with an NSTableView. <-- 8< ---> > if, anywhere else in this class, I attempt to get the selected row using > [myIvarForTheNSTV selectedRow] I ALWAYS get 0. Egad. I have to apologize to everyone who even wasted a minute reading my post. As it turned out, I had somehow created TWO instances of my object: one was connected to the table view, the other wasn't. Interestingly, they were strangely cross-connected to IB actions... Moral: ObjectAlloc is a REALLY USEFUL tool! Dave ------- /\/\ Dave Coyle <dcoyle@ctp.com> / /_ \ Cambridge Technology Partners \ / / 118/119 Lower Baggot Street \/\/ Dublin 2, Republic of Ireland Tel: +353 1 607 9008 WWW: http://www.ctp.com>
Date: Fri, 11 Jul 1997 08:07:03 -0400 From: joe.ragosta@dol.net (Joe Ragosta) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Message-ID: <joe.ragosta-ya02408000R1107970807030001@news.dol.net> References: <AFE9608B-A7701@206.165.44.26> <joe.ragosta-ya02408000R0907971830490001@news.dol.net> <Pine.SGI.3.96.970710201114.13965B-100000@wong> Organization: Graver Chemical Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-transfer-encoding: 8bit In article <Pine.SGI.3.96.970710201114.13965B-100000@wong>, Ian Russell Ollmann <iano@scripps.edu> wrote: > On Wed, 9 Jul 1997, Joe Ragosta wrote: > > > What happened? Lawson writing something supportive of Apple? That's bigger > > news than Amelio's resignation. Maybe we can get AP and Reuters to carry > > that news to put a positive spin on things. > > It would only be bigger news if Joe Ragosta wrote something unsupportive > of Apple. :-) That would mean invoking the emergency broadcast system, I > believe. Which, of course, proves that you don't actually _read_ anything I write. I have been openly critical of _many_ things Apple has done: Lack of marketing Dropping OpenDoc/Cyberdog Several UI issues Taking so long to get rid of Spindler Not correcting the press Lack of developer support Granted, I believe that Apple, on the whole, is on the right track. I suppose you have a problem with that, so you accuse me of being one sided and blind. -- Regards, Joe Ragosta See the Complete Macintosh Advocacy Site http://www.dol.net/~Ragosta/complmac.htm
From: Barry <Barry.nospam.Reuter@stanford.edu> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Fri, 11 Jul 1997 07:47:46 -0700 Organization: Stanford University Message-ID: <33C64791.1B7@stanford.edu> References: <AFE9608B-A7701@206.165.44.26> <5q16cp$abi$1@news.platinum.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Gary W. Longsine wrote: > > In <AFE9608B-A7701@206.165.44.26> it appeared that "Lawson English" wrote: > > I think that all Mac developers and other interested people should meet on > > AIMED-TALK > > > > (AIMED = Association of Independent Macintosh Engineers and Developers) > > > > to discuss how to handle this situation (Amelio and Hancock resigning). We > > need to act very quickly to respond to this in an > > Apple/Mac/Rhapsody-supportive way. > > > > > If we can portray developers as taking a "business as usual" stance, maybe > > the press will at least mention this... > > > > We gotta be as positive (in a non-finger-pointing way) as possible about > > this situation. And figure out how to convince dealers, customers and > > stockholders that this is not the end of the world for Apple/Macintosh/etc. > > > HOW DO I SUBSCRIBE TO AIMED-TALK? > > Send E-mail to: > > <aimed-talk-request@aimed.org> > > with a blank subject line and the following command as the > > first (and only) line of the message body: > > SUBSCRIBE > > Lawson, I could kiss you... or did you go away and leave yourself logged in > again? > Well, it appears Lawson was wrong. Instead of being positive and supportive, we should take this turn of events as a new opportunity to exchange insults and put-downs (in a finger-pointing way). Barry
From: "Jane Fabeck" <bmont@pacificrim.net> Newsgroups: comp.sys.next.programmer Subject: OpenStep App not running on Win95 Date: 11 Jul 1997 15:50:56 GMT Organization: Objects Alive Custom Software Message-ID: <01bc8e13$27b49480$02ac2ac0@jane> I created an application on WindowsNT and tried to run it on Windows95 without success. When I double-click the executable the application does'nt do anything. It just returns as if it was finished. I ran all the demo apps on the Windows95 without any problem. I only have the deployment part of OpenStep 4.2 installed on the Windows95 machine, Is this a problem? Do I have to recompile my application on the Windows95 machine? Are there different build options I have to take when compiling the application on the WindowsNT machine? Is there some sort of log file on Windows95 that I may be able to consult to find out what happened? Thanks Bruce Montegani bmont@telcomplus.com
From: acurylo@inmediapresents.nospam.com (Alex Curylo) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 11 Jul 1997 17:06:50 GMT Organization: InMedia Presentations Message-ID: <acurylo-1107971011180001@van0429.tvs.net> References: <acurylo-1007970942180001@van0209.tvs.net> <AFEAD48E-1812E9@206.165.44.28> In article <AFEAD48E-1812E9@206.165.44.28>, "Lawson English" <english@primenet.com> wrote: > Could it possibly be because Apple did an absolutely horrible job marketing > GX? > What if GX could be marketed properly? Yeah, well, the same could be said about AppleScript, the Telephone Manager, the Communications Toolbox, the PowerTalk collection, the Speech/Recognition collection, ScriptX, HyperCard, and I could go on and on for quite a while here but I'd get really depressed... What's actually the big missing piece is not the marketing, it's the lack of effort to integrate new technologies into the core OS and applications. How many Apple and Claris applications supported GX in any meaningful fashion? In ANY fashion? And how many were downright incompatible in one form or another? THERE is the core problem, not the marketing. As with pretty much everything I listed above. I personally suspect that GX is going to enjoy far greater success in its new incarnation of QT Vector Animation then it ever did as itself, anyway... ---- Alex Curylo -- alex@witty.com
From: jgrass@cs.umass.edu (Joshua Grass) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 11 Jul 1997 17:19:21 GMT Organization: CMPSCI Dept, UMass Amherst Message-ID: <5q5pup$hq0@kernighan.cs.umass.edu> References: <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> <AFE9BDAA-CBFE5@206.165.44.143> In article <AFE9BDAA-CBFE5@206.165.44.143>, Lawson English <english@primenet.com> wrote: >Craig Koller <ckoller@worldnet.att.net> said: > >> >> Come on, Lawson, we all know Amelio got canned because he axed QDGX... > >Yes and no. Amelio got canned because he didn't improve Apple's bottom line >and (I suspect) because Q3 results are going to be much worse than the >concensus suggested: $120 million+ losses, I pontificate, maybe more. > >As to how this relates to GX? > >GX is a MacOS technology that puts everything else in its category (save >possibly Taligent Graphics) to shame. Amelio's decision to not only go with >the inferior Display PostScript in Rhapsody but to ALSO cripple MacOS >graphics by killing GX's ability to print, is as good an example of the >lack of consumer advocacy that has plagued Amelio's (and now, Steve Jobs', >I fear) Apple as you can find. > >GX enables small developers to compete with large developers in the 2D >graphics arena. > >OpenDOc does that in a more comprehensive way. > I think the real leason here, and the one that Amelio understood, was that Apple doesn't have the marketing resources to "push" more then one technology at a time. And the one they picked was Quicktime, now everyone knows about quicktime, everyone loves quicktime, and it doesn't need to be pushed any more. I'd say that Quicktime is now pretty close to a standard, and Apple has made it clear what the new push will be, Rhapsody, and the fact that it will bring the Mac interface to intel. Apple doesn't have the resources to "push" anything else, and against Microsoft, if you don't push it it will die. Think about how much press ActiveX and DirectX have gotten, to name just two Microsoft technologies. Apple will survive long enough to make a really good show of Rhapsody, and that's were all of the resources have to be spent. Even if Apple dies it will do one good thing before then, it will kill Windows 95-97(8?) and all the DOS underpinnings it has. How? Because Rhapsody will get enough press that Microsoft will have to shift to full NT deployment. It's not the greatest of worlds, everyone using NT, but at least it's a world I can live in. Of course, I hope Rhapsody catches on, and as soon as the developer version is out, I'm going to start writing for it. But no matter what happens, windows 97 is a goner...and that makes me happy. Joshua -- If you want to know who you are, | jgrass@cs.umass.edu it's important to know who you've been. | http://anytime.cs.umass.edu/~jgrass
From: Alex Blakemore <alex@genoa.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Openstep examples OO design (was Re: Some initial OpenStep notes) Date: 11 Jul 1997 03:35:32 GMT Organization: Genoa Software Systems Message-ID: <5q49m4$mt@saturn.genoa.com> References: <5pmd4m$i21$2@news.xmission.com> <5q26ue$ndb$1@lazar.select-tech.si> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: izidor.jerebic@select-tech.si In <5q26ue$ndb$1@lazar.select-tech.si> izidor.jerebic@select-tech.si wrote: > If you compare Draw in Nextstep 3.x and Draw in 4.x you can see that the > Draw in 4.x is completely new code and very improved. Actually, it > conforms to ModelViewController OO design. The original Draw was one big > hack, really. But the new one is rather good Are you sure we're looking at the same app? As far as I can tell, 4.X draw is still one big hack. I tell you from experience, it is not a good foundation for building a custom app :-( > I must confess I did not do any modification of the example code. I just > briefly went through the code of examples and it looked OK to me... > what are the big deficiencies you found in new examples, e.g. Draw? Are there any > practices advised in there that we should definitely not follow? The undo system is atrocious for one. -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: How to split classes hierarhy to bundles Date: 11 Jul 1997 18:37:18 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5q5ugu$l16$1@owl.slip.net> References: <33C0F8E2.273C@rtr.transit.ru> Iracly <iracly@rtr.transit.ru> wrote: > Is anybody have idea how to divide class hierarhy to different loadable > bundles. This works fine for me when using frameworks. You might try using frameworks instead of bundles. -- Steve Dekorte - OpenStep consultant - San Francisco
From: drifterusa@macconnect.com (John Bauer) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Fri, 11 Jul 1997 11:58:45 -0500 Organization: None Message-ID: <1997071111584570467@accs-as804-dp09.atln.grid.net> References: <AFE9608B-A7701@206.165.44.26> <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> <*johnnyc*-1007970908020001@192.0.2.1> John Christie wrote: > Craig Koller wrote: > > > Come on, Lawson, we all know Amelio got canned because he axed QDGX... > > And I thought it was because Markkula owns a big piece of BBDO. Nah, the Board was getting concerned that Apple might actually have a coherent business strategy in place. John Bauer
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 11 Jul 1997 12:02:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFEBD615-9EE0E@206.165.44.66> References: <E0wmjBI-0005JS-00@redemption.uniserve.com> To: "Alex Curylo" <acurylo@inmediapresents.nospam.com> nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Alex Curylo <acurylo@inmediapresents.nospam.com> said: > I personally suspect that GX is going to enjoy far greater success in its > new incarnation of QT Vector Animation then it ever did as itself, > anyway... Except that it is NOT QT Vector Animation. Ironic, isn't it? ++++++++++++++++++++++++++++++ HOW DO I SUBSCRIBE TO GX-TALK? --------------------------------- To subscribe to GX-TALK proceed as follows: Send E-mail to: <gx-talk-request@aimed.org> with anything in the subject line and the following command as the first (and only) line of the message body: SUBSCRIBE +++++++++++++++++++++++++++++++++
From: tiggr@ics.ele.tue.nl (Pieter Schoenmakers) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 12 Jul 1997 00:11:01 +0200 Organization: Eindhoven University of Technology Sender: tiggr@tom.ics.ele.tue.nl Message-ID: <x7pvspfdg9.fsf@tom.ics.ele.tue.nl> References: <199707061639281456216@dialup156-4-13.swipnet.se> <5pq2l2$11i@mpaque.mpaque> Original-Sender: tiggr@tom.ics.ele.tue.nl In-reply-to: mpaque.spa-am@nospam.wco.com's message of 6 Jul 1997 23:34:10 -0700 to: mpaque@wco.com In article <5pq2l2$11i@mpaque.mpaque> mpaque.spa-am@nospam.wco.com (Mike Paquette) writes: > Reference counting is nothing but a very poor implementation of GC. It has the advantage of working over an allocation pool spread over multiple address spaces, or processes. The two are seperate mechanisms and the local GC can be a real GC while the distributed GC (which maintains the links across address spaces) can employ reference counting. --Tiggr
From: Wolfgang Baron <wbaron@ixpoint.de> Newsgroups: comp.sys.next.bugs,comp.sys.next.programmer Subject: Re: CVS unusable with 4.2 PB/Mach? Strange path mapping bug . . . Date: 11 Jul 1997 13:51:29 GMT Organization: iXpoint Informationssysteme GmbH, www.ixpoint.de Distribution: world Message-ID: <5q5dp1$2s8$1@ixpoint.de> References: <5q3hqv$1ifo@news.doit.wisc.edu> <5q4sir$24o$1@ixpoint.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Cc: wbaron@ixpoint.de Hi, In <5q4sir$24o$1@ixpoint.de> Wolfgang Baron wrote: > In <5q3hqv$1ifo@news.doit.wisc.edu> Michael Giddings wrote: Apart from the problems seen under Mach, has anyone successfully worked with cvs under NT? When I checkout a module, only the files and directories with the generated CVS subdirectories of the first level directory get extracted. The first level directory is named 'cvs' instead of the module name. When I run cvs from a shell, it works as expected. Has anyone successfully worked with cvs under NT ? Has anyone seen the same or other problems ? -- +++++++++++++++++++++++++++++++++++++++++++++++++ + Dipl.- Inform. Wolfgang Baron + + iXpoint Informationssysteme GmbH + + Daimlerstrasse 3, 76275 Ettlingen, Germany + + phone ++49 7243 3775-82 Fax ++49 7243 3775-77 + + email: wbaron@ixpoint.de (NeXTmail and MIME) + +++++++++++++++++++++++++++++++++++++++++++++++++ + pConnector - Network Printing for NEXTSTEP + + sca - the static code analyser + +++++++++++++++++++++++++++++++++++++++++++++++++
From: NNTP Component <nntpcomp@microsoft.com> Newsgroups: comp.sys.next.programmer,uswest.test,uswest.test,uswest.test Subject: Just a test Date: Fri, 11 Jul 1997 15:50:01 PST Organization: US WEST Advanced Technologies Message-ID: <5q69p0$4ec@news.advtech.uswest.com> Hey, this is only a test! About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes. About 3 kilobytes.
From: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Openstep examples OO design (was Re: Some initial OpenStep notes) Date: 11 Jul 1997 22:11:52 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5q6p58$uh0$1@sps1.phys.vt.edu> References: <5pmd4m$i21$2@news.xmission.com> <5q26ue$ndb$1@lazar.select-tech.si> <5q49m4$mt@saturn.genoa.com> In article <5q49m4$mt@saturn.genoa.com>, Alex Blakemore <alex@genoa.com> wrote: > In <5q26ue$ndb$1@lazar.select-tech.si> izidor.jerebic@select-tech.si wrote: > > what are the big deficiencies you found in new examples, e.g. Draw? > > Are there any practices advised in there that we should definitely not > > follow? > The undo system is atrocious for one. How would you implement an undo system?
From: gtr@cyberbundle.net Newsgroups: comp.sys.next.programmer Subject: >Entrepreneurs Needed Date: 11 Jul 1997 16:21:30 GMT Organization: AFE, LLP Message-ID: <5q5mia$n1r$1@its.hooked.net> >>>>>>>>>ENTREPRENEURS NEEDED!<<<<<<<<<< We are currently searching for motivated ENTREPRENEURS to assist us in selling our Internet Marketing Software. 60% COMMISSIONS. Great earning potential! Come check out the details! http://www.cyberbundle.com/dealer.html --------------->>>>>>>>>>> HOT, HOT, Summer Specials! For a limited time, become a DEALER for FREE! Thats right, up to $125.00 in savings!! When you print out the dealer form, just write NEWSGROUP SPECIAL in the total, and become a DEALER for FREE! All you have to do, is fax or mail it in and start making HUGE profits! So go check it out today! http://www.cyberbundle.com/dealer.html
From: Steve Watt <steve@watt.com.nospam> Organization: USENET spam abatement Sender: randy97 Date: 12 Jul 97 02:54:03 GMT Message-ID: <cancel.33c52b4f.1@pgh.nauticom.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33c52b4f.1@pgh.nauticom.net> Control: cancel <33c52b4f.1@pgh.nauticom.net> I have cancelled this article which had a BI of more than 20. The original subject was: }Subject: http://www.love.com
From: mcconent@cyberbundle.net Newsgroups: comp.sys.next.programmer Subject: ...Free Cash Grants Date: 11 Jul 1997 18:12:49 GMT Organization: Abc, LLC Message-ID: <5q5t31$r7d$5@its.hooked.net> >>>>>>>>>>FREE CASH GRANTS<<<<<<<<<<< Never pay back! $500-$50,000! Business, Medical, Debts, Education...FREE INFORMATION. Visit Our Web Site for FREE Details! http://www.cyberbundle.com/finacial.html
From: Dan Harley<netpro@op.net> Newsgroups: comp.sys.next.programmer Subject: Best price on the net CD-R 74 Minute Gold $2.99 Retail package - no rebates no gimmics no surcharges Date: 11 Jul 1997 12:23:52 GMT Organization: NetPro Computer Services, Inc. Message-ID: <5q58ko$284$6732@picasso.op.net> NetPro Computer Services, Inc. http://www.netprocs.com The sign of superior service since 1989! Weekly product specials If you don’t see if please email info@netprocs.com for a prompt quotation! SCSI subsystem specialists - let us build yours today Distributor of ACER OPEN SYSTEMS and Components Master distributor DRAM,INTEL processors we will not be undersold on geniune Intel products Dan the Disc and DAT man brings you the best prices on the net: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Wholly KAO ~ best price on the net! KAO 74 Minute CD-R 680MB 2x/4x compatible highest quality. The J & J of Japan! $2.99 kick off special (retail) $2.79 Spindle pack of 50 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Verbatim 74 Minute / 680 MB / 2x / 4x certified White / clear jewel case, white J card, individually wrapped, master case of 50, 5 inner boxes of 10. $3.69 each Memorex 74 Minute GOLD generic package - white j Card, individually wrapped in white/clear jewel case. 25 to a master carton (can break up) $ 3.39 Ricoh ReWritable 74 Minute CD-RW $22.00 HP 6020ES recorder (external SCSI 2x/6x) $529.00 HP 6020 recorder (internal SCSI 2x/6x) $429.00 JVC 2x/6x Internal kit $389.00 + free shipping!!! * Adaptec EZ CD Pro V. 2.11 PC & Mac, Dos Ver 3.5 software * kit includes PressIT CD labler software, 100 CD Blank lables, CD filled with templates, clipart - templates compt. with Pagemaker, Illus, P/S. * 16 bit SCSI controller, cables, drivers * 2 pieces blank media * Highest quality kit COMPAC pales in comparision * your satisfaction assured or you may return for full credit. * JVC Inventor of VHS makes a SUPERIOR mechanism - read the trade mags! Ricoh 6200 spi4 ReWritable Internal kit $519.00 * Authoring software now with variable packet writing - treats R/W as a floppy drive! * Adaptec EZ CD Pro V 2.11, Toast CD for Mac * 16 bit SCSI controller, cables, drivers * one CD-R, one rewritable CD-RW
From: Ian Russell Ollmann <iano@scripps.edu> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Sat, 12 Jul 1997 01:08:06 -0700 Organization: The Scripps Research Institute, La Jolla, CA Message-ID: <Pine.SGI.3.96.970712003021.28283C-100000@wong> References: <AFE9608B-A7701@206.165.44.26> <joe.ragosta-ya02408000R0907971830490001@news.dol.net> <Pine.SGI.3.96.970710201114.13965B-100000@wong> <joe.ragosta-ya02408000R1107970807030001@news.dol.net> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII To: Joe Ragosta <joe.ragosta@dol.net> In-Reply-To: <joe.ragosta-ya02408000R1107970807030001@news.dol.net> On Fri, 11 Jul 1997, Joe Ragosta wrote: > In article <Pine.SGI.3.96.970710201114.13965B-100000@wong>, Ian Russell > Ollmann <iano@scripps.edu> wrote: > > > On Wed, 9 Jul 1997, Joe Ragosta wrote: > > > > > What happened? Lawson writing something supportive of Apple? That's bigger > > > news than Amelio's resignation. Maybe we can get AP and Reuters to carry > > > that news to put a positive spin on things. > > > > It would only be bigger news if Joe Ragosta wrote something unsupportive > > of Apple. :-) That would mean invoking the emergency broadcast system, I > > believe. > > Which, of course, proves that you don't actually _read_ anything I write. > > I have been openly critical of _many_ things Apple has done: > > Lack of marketing > Dropping OpenDoc/Cyberdog > Several UI issues > Taking so long to get rid of Spindler > Not correcting the press > Lack of developer support > > Granted, I believe that Apple, on the whole, is on the right track. I > suppose you have a problem with that, so you accuse me of being one sided > and blind. No Joe, I agree with you 90% of the time. I'm a big Mac fan, but some of the things you've written to the advocacy groups are past what even I'm willing to believe...and no I don't read everything you write. Who could keep up? ;-) It was just a jab in fun in a similar spirit to your poke at Lawson, who also often has quite reasonable things to say (though once when I said he was well informed and trustworthy chap in another group, I was referred to his contributions to this one -- csm.programmer.misc) To be honest, I think you are often a bit one sided, but certainly not blind. You understand the mac a hell of a lot better than most. I think Apple was on pretty much the right track until yesterday. Now? Apple needs more change like they need another Q2 with a 700 million loss. Actually, it's funny how these things go hand in hand. So, sure Apple could do better marketing, and I'd like to be able to go to WWDC for free (the admission price is about equal to my disposable income for the year), and they could certainly put a better spin on negative events rather than hatching crazy stories about Armelio's political aspirations that no one believes, etc. I'll take your word that you've written such things and I'd have certainly agreed with you when you did, if I had read it. I'm sorry for not giving credit where it was due. Ian Ollmann
From: Ian Russell Ollmann <iano@scripps.edu> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Sat, 12 Jul 1997 01:57:57 -0700 Organization: The Scripps Research Institute, La Jolla, CA Message-ID: <Pine.SGI.3.96.970712010949.169B-100000@wong> References: <acurylo-1007970942180001@van0209.tvs.net> <AFEAD48E-1812E9@206.165.44.28> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <AFEAD48E-1812E9@206.165.44.28> On 10 Jul 1997, Lawson English wrote: > Alex Curylo <acurylo@inmediapresents.nospam.com> said: > > > Didn't we go over this before? The major source of actual dollars for > > Apple in the COMPETITIVE marketplace is publishing. Publishers of all > > sorts go all glassy-eyed and drooling at the mere thought of being able > to > > work in Display PostScript, while they scatter screaming at the mere > > mention of GX. So a realistic look at Apple's actual marketplace leads > one > > inescapably to the conclusion that DPS is the truly consumer advocating > > position. QED. > > Could it possibly be because Apple did an absolutely horrible job marketing > GX? > > What if GX could be marketed properly? I think the best way to market it is to just include it as a new set of features in Sys 8.1 or 8.2, features that cannot be removed. While a 3 MB larger System RAM footprint was unreasonable a year or two ago, when Apple still shipped machines with 8 MB total, these days, it is a little more reasonable, so long as the machine ships with 24-32 MB minimum. If there is some redundancy between Quickdraw and GX (don't know, haven't programmed for it) then maybe the old Quickdraw could be pulled out and calls to it redirected to GX, thereby saving some RAM. I think the problem was that you can't convince developers to use something if there is no installed base, and you wont convince users to give up extra RAM for something that so far as they can tell doesn't do any more than what is already there. However, I think that so long as it's candy coated, people will swallow the pill. The new appearance manager might work nicely with GX. Make the whole thing look really cool with lots of new GX effects, give the user the option of always running at high screen resolutions using GX to scale the image so it is not too small, etc. Backwards compatability issues aside, it could be really neat. It seems standard these days to give big releases a whole new look. MS does it with every Word release. Apple does it with the System. Is OS8 all that much different from the rest of the updates between 7.5 and 7.6.1 that it really needs the new monniker? No, not really. The new finder is cool and a time saver, but before we got redone error handling code, a lot more stability, maybe a 20% speed boost overall, faster boot times, better extensions handling, Open Transport, revamped virtual memory, PCI support, etc. To me the reason why OS 8 is OS8 and not 7.7 is that it looks noticeably different. People will pay big money (or big RAM) for big updates if they can be sure that they can tell the difference. GX is great, but also subtle. To sell it you need to use it in a way that wow's the public. Use those capabilities to make the mac look like something it wasn't before. Then people will want it and developers can develop for it. The GX implementation has to be impressive and system wide -- i.e. don't just ship the new system with Gerbils GX and call it a day. Allow new mouse pointer types that are transparent. Give the user the ability to make any window, or at least any modal dialog window translucent/transparent so that it doesn't obscure important data. Make the entire desktop scalable so that people with poor vision can make things three times normal size and people with 17" monitors that go to 1280x1024 can actually use that resolution and still read 12 pt text. Make windows individually scalable so that you can make a window small and still see all of its contents. How many times have you been able to fit one and three quarters pages on your screen, but not two? This could be fixed with scalable windows. Look at every standard interface feature and see if GX can enhance its operation in a way that adds value to the system and productivity to the user. Ian Ollmann
From: Steve Watt <steve@watt.com.nospam> Organization: USENET spam abatement Sender: gtr@cyberbundle.net Date: 12 Jul 97 08:39:47 GMT Message-ID: <cancel.5q5mia$n1r$1@its.hooked.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5q5mia$n1r$1@its.hooked.net> Control: cancel <5q5mia$n1r$1@its.hooked.net> I have cancelled this article which had a BI of more than 20. The original subject was: }Subject: >Entrepreneurs Needed
From: Steve Watt <steve@watt.com.nospam> Organization: USENET spam abatement Sender: mcconent@cyberbundle.net Date: 12 Jul 97 08:39:47 GMT Message-ID: <cancel.5q5t31$r7d$5@its.hooked.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5q5t31$r7d$5@its.hooked.net> Control: cancel <5q5t31$r7d$5@its.hooked.net> I have cancelled this article which had a BI of more than 20. The original subject was: }Subject: ...Free Cash Grants
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Sat, 12 Jul 1997 05:03:42 -0600 Organization: ErgoTech Message-ID: <33C7648E.6BB5@ergotech.com> References: <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> <AFE9BDAA-CBFE5@206.165.44.143> <acurylo-1007970942180001@van0209.tvs.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Alex Curylo wrote: > > In article <AFE9BDAA-CBFE5@206.165.44.143>, "Lawson English" > <english@primenet.com> wrote: > > He did nothing to inspire the customers, the engineers, the developers or > > the stockholders. > > Hmmmm...I fall into three of those four categories, and Rhapsody inspired me :) The man to inspire is Steve Jobs. His presentations are always great, with credit going to all the behind the scenes people who pull off the demos. From the "Wall Street Journal" interactive edition: >Apple Computer announced that Gilbert Amelio has resigned as chairman and >chief executive and that co-founder Steve Jobs will play an expanded >management role. Just so long as he isn't put in charge of the marketing department. The quality of NeXT and Apple marketing has been remarkable, in, as we all know, a negative sense. The one thing that you have to truly admire at Microsoft is the quality of their marketing. One way they do this is pre-annouce just about every feature of a product and generate some press hype long before the product is released, even the features that don't ever make it into the product. Traditionally NeXT's philosophy has been to keep everything under wraps until the big release day, which gets you one day of press coverage, instead of 6 months (or a year in the case of Win95). I'm intrigued to see what the Rhaposody philosophy will be. The pre-releases make me think that they may have learnt something from the masters, hopefully enough. Jim
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: initWithEditingContext:classDescription:globalID question Date: 12 Jul 1997 15:09:26 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5q86n6$4jq$1@concorde.ctp.com> Hi, I'm quite new in EOF and I am really amazed about the stuff that NeXT has put together. It takes some (a lot of ?) time to get used to it and to fully understnad what is going on behind the scenes. One thing that was not really clear form the docu of the EoGenericrecord is: initWithEditingContext:classDescription:globalID does that perfomr an [editicoContext insertObject:self] internally ? I know that I might track that down using the debugger...but my EOF project does not use EoGenericRecords and so if some knows the quick YES-NO answer it would keep me from setting up some special lines just to track this down. Thanx a lot. aloha Tomi
From: *johnnyc*@or.psychology.dal.ca (John Christie) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Sat, 12 Jul 1997 12:35:03 -0400 Organization: ISINet, Nova Scotia Message-ID: <*johnnyc*-1207971235030001@192.0.2.1> References: <acurylo-1007970942180001@van0209.tvs.net> <AFEAD48E-1812E9@206.165.44.28> <Pine.SGI.3.96.970712010949.169B-100000@wong> In article <Pine.SGI.3.96.970712010949.169B-100000@wong>, Ian Russell Ollmann <iano@scripps.edu> wrote: > I think the best way to market it is to just include it as a new > set of features in Sys 8.1 or 8.2, features that cannot be removed. While > a 3 MB larger System RAM footprint was unreasonable a year or two ago, If they just loaded the draw lbraries into the System (and maybe replaced some of QD in the process) it would only take a few hundred k. > However, I think that so long as it's candy coated, people will > swallow the pill. The new appearance manager might work nicely with GX. > Make the whole thing look really cool with lots of new GX effects, give > the user the option of always running at high screen resolutions using > GX to scale the image so it is not too small, etc. Backwards > compatability issues aside, it could be really neat. Excellent idea, just make it invisible to all but programmers who want to use it. Write as much of the OS as possible in GX. Port the draw libraries to run within DPS. Finish the job. Make it an option that is ALWAYS available. > The GX implementation has to be impressive and system wide -- i.e. Yes, but not as a replacement. The current draw libraries could be incorporated into the system with only a few hundred k of code. Put the PC version on the net and tote it as the standard for lightweight internet vector grphics. There is no single move that could be done to speed the internet than reduce graphics sizes by a factor of 10. -- You aren't free if you CAN choose - only if you DO choose. All you are is the decisions you make. Remove "*" and ohnny (i.e. jc@) to reply via email
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: NSMutableDictionary dictionaryWithContentsOfFile:aPath should different. Date: 12 Jul 1997 16:14:34 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5q8aha$5m3$1@concorde.ctp.com> Hi, I just ran accross this kind of annoying "feature" in NSMutableDictionary dictionaryWithContentsOfFile method. NSDcits and their ability to store data in ASCII format to disc is really nice and handy for tons of things. But the really big problem is that you always get in-mutable ionstances when you read the ASCII wil in. This is a pain since in order to manipulate a stored strucuter you have to make a deep copy first (using e.g. MiscKits deepCopy feature) But this really is a nightmare since it requires you too keep two copie of the potentially huge NSDict in memory. IMHO the NSMutablexxxxx classes should behave differently then they do right now. If I use a NSDict/Array to read in a structure I want it to be immutable. But fi I use a NSMutableXxxx with xxxxWithContentsOfFile then I want to get a entrier strucutre build out of mutable classes. This would be a lot more "productive" and a lot more useful then the current behavior. Did I miss some hidden feauter or would you agree ? Aloha Tomi
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection and the distributed aspect Date: 12 Jul 1997 16:37:34 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5q8bse$669$1@concorde.ctp.com> References: <199707061639281456216@dialup156-4-13.swipnet.se> <5pq2l2$11i@mpaque.mpaque> <x7pvspfdg9.fsf@tom.ics.ele.tue.nl> tiggr@ics.ele.tue.nl (Pieter Schoenmakers) wrote: >In article <5pq2l2$11i@mpaque.mpaque> mpaque.spa-am@nospam.wco.com (Mike Paquette) writes: > > > Reference counting is nothing but a very poor implementation of GC. > > It has the advantage of working over an allocation pool spread over > multiple address spaces, or processes. > >The two are seperate mechanisms and the local GC can be a real GC while >the distributed GC (which maintains the links across address spaces) can >employ reference counting. --Tiggr Ok...lets thing about this for a while. What does that imply. A local app can create object ref cylces since GC will detect them and remove them properly. But if your ref cycles happen to be between different processes over DO..it will never go away ? (I can't see a way for the "smart" local GC to fix that shortcoming of the "distributed" GC..) While I sure love the idea of real GC I can fully understand why NeXT went with the current solution since there is absolutly no difference between local and distributed objects. You have the same policies and restrictions all over the place which might cause less confusion while it always demands the same caution during the design of retain/release actions. I am not questioning the benefits of "real" GC...but I can see benefits of a unified "GC" mechanism as used under Openstep. Aloha Tomi
From: acurylo@inmediapresents.nospam.com (Alex Curylo) Newsgroups: comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 12 Jul 1997 17:27:21 GMT Organization: InMedia Presentations Message-ID: <acurylo-1207971031550001@van0409.tvs.net> References: <E0wmjBI-0005JS-00@redemption.uniserve.com> <AFEBD615-9EE0E@206.165.44.66> In article <AFEBD615-9EE0E@206.165.44.66>, "Lawson English" <english@primenet.com> wrote: > Except that it is NOT QT Vector Animation. Then why was Tom Dowd the one demonstrating it at WWDC? :) Not to mention why do the Lightning/Draw people export their animations as QT3 movies? Not because they spent a lot of time writing conversion code :) ---- Alex Curylo -- alex@witty.com
From: acurylo@inmediapresents.nospam.com (Alex Curylo) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 12 Jul 1997 17:33:13 GMT Organization: InMedia Presentations Message-ID: <acurylo-1207971037470001@van0409.tvs.net> References: <acurylo-1007970942180001@van0209.tvs.net> <AFEAD48E-1812E9@206.165.44.28> <Pine.SGI.3.96.970712010949.169B-100000@wong> In article <Pine.SGI.3.96.970712010949.169B-100000@wong>, Ian Russell Ollmann <iano@scripps.edu> wrote: > The GX implementation has to be impressive and system wide -- i.e. > don't just ship the new system with Gerbils GX and call it a day. Allow > new mouse pointer types that are transparent. Give the user the ability to > make any window, or at least any modal dialog window > translucent/transparent... This is PRECISELY the point I was trying to make about why GX/AppleScript/Telephone Manager/pick your failed technology of choice went nowhere in particular; Apple didn't integrate its whizzy features into the core OS in a sustained fashion. It seems rather unfashionable to say nice things about Amelio these days, but one thing you do need to give him credit for is doing a lot of good work towards breaking down the independent fiefdoms at Apple that led inescapably to this situation... ---- Alex Curylo -- alex@witty.com
From: acurylo@inmediapresents.nospam.com (Alex Curylo) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 12 Jul 1997 17:35:45 GMT Organization: InMedia Presentations Message-ID: <acurylo-1207971040190001@van0409.tvs.net> References: <acurylo-1007970942180001@van0209.tvs.net> <AFEAD48E-1812E9@206.165.44.28> <Pine.SGI.3.96.970712010949.169B-100000@wong> <*johnnyc*-1207971235030001@192.0.2.1> In article <*johnnyc*-1207971235030001@192.0.2.1>, *johnnyc*@or.psychology.dal.ca (John Christie) wrote: > Put the > PC version on the net and tote it as the standard for lightweight internet > vector grphics. There is no single move that could be done to speed the > internet than reduce graphics sizes by a factor of 10. <sigh> This has actually been done already. Another great coup for Apple marketing that was, I see. ---- Alex Curylo -- alex@witty.com
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 12 Jul 1997 12:14:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFED2A36-362CE@206.165.44.10> References: <E0wn5yj-0002Rw-00@redemption.uniserve.com> nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Alex Curylo <acurylo@inmediapresents.nospam.com> > (A copy of this message has also been posted to the following newsgroups: > comp.sys.mac.programmer.games, > comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.progr > ammer) > > In article <AFEBD615-9EE0E@206.165.44.66>, "Lawson English" > <english@primenet.com> wrote: > > > Except that it is NOT QT Vector Animation. > > Then why was Tom Dowd the one demonstrating it at WWDC? :) > He may have helped write it, but it is NOT GX, or so we GX-oriented folks have been told by both QT folk and GX folk. > Not to mention why do the Lightning/Draw people export their animations > as > QT3 movies? Not because they spent a lot of time writing conversion code > :) Excuse? They most certainly DID write a converter from GX to QT Vector Graphics. Want Mihail Lari's email address so that you can confirm what they already say on their web page? There is a LOT more to GX than what appears in QT. ++++++++++++++++++++++++++++++ HOW DO I SUBSCRIBE TO GX-TALK? --------------------------------- To subscribe to GX-TALK proceed as follows: Send E-mail to: <gx-talk-request@aimed.org> with anything in the subject line and the following command as the first (and only) line of the message body: SUBSCRIBE +++++++++++++++++++++++++++++++++
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: GX on HyperCard (Was Re: Amelio & Hancock Leave Apple Date: 12 Jul 1997 12:59:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFED34D6-5E1CF@206.165.44.10> References: <Pine.SGI.3.96.970712010949.169B-100000@wong> nntp://news.primenet.com/comp.sys.mac.programmer.games, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Ian Russell Ollmann <iano@scripps.edu> said: [GX would have potential if marketed right] > How many times have you been able to fit > one and three quarters pages on your screen, but not two? This could be > fixed with scalable windows. Look at every standard interface feature and > see if GX can enhance its operation in a way that adds value to the system > and productivity to the user. That's supposed to be a feature with Rhapsody. "GX competes with DPS and therefore must go" is the attitude that I have picked up. Some have suggested that this is pure paranoia, and perhaps it is. Not all decisions are based on greed and shortsightedness. Afterall, Amelio didn't set out to lose Apple money during Christmas last year, so the major blame that one can lay at his feet is what he has readily admitted: he is not a marketing guy. BTW, the GX HyperCard XFCN is going... Well, I just crashed the Mac while adding a new library function of my own to the list of available functions so I'm taking a break... HOPEFULLY, it is going well. THe current call that I'm working on tracks the mouse and draws an arbitrary GX shape on and off-screen simultaneously with the mouse-tracking loop in C for maximum speed. If it works as planned (I hope!), you will be able to create a simple-minded GX doodling space in any HyperCard stack merely by pasting a locked text field into the stack and adding an update routine to refresh the GX drawing when needed. The GXFCN has to be available, of course. Kewl, huh? ++++++++++++++++++++++++++++++ HOW DO I SUBSCRIBE TO GX-TALK? --------------------------------- To subscribe to GX-TALK proceed as follows: Send E-mail to: <gx-talk-request@aimed.org> with anything in the subject line and the following command as the first (and only) line of the message body: SUBSCRIBE +++++++++++++++++++++++++++++++++
From: netpro@op.net Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5q58ko$284$6732@picasso.op.net> Control: cancel <5q58ko$284$6732@picasso.op.net> Date: 12 Jul 1997 21:21:24 GMT Organization: MindSpring Enterprises Message-ID: <5q8sgk$er6@camel12.mindspring.com> ignore Article canceled by slrn 0.9.4.0
Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer From: slur@world.std.com (Scott Lahteine) Subject: Re: Amelio & Hancock Leave Apple Sender: news@world.std.com (Mr Usenet Himself) Message-ID: <slur-ya076180001207971849420001@news.std.com> Date: Sat, 12 Jul 1997 23:49:42 GMT Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=Macintosh References: <ckoller-ya023480000907971531360001@netnews.worldnet.att.net> <AFE9BDAA-CBFE5@206.165.44.143> <5q5pup$hq0@kernighan.cs.umass.edu> Mime-Version: 1.0 Organization: Hot Eggs In article <5q5pup$hq0@kernighan.cs.umass.edu>, jgrass@cs.umass.edu (Joshua Grass) wrote: |>Of course, I hope Rhapsody catches on, and as soon as the developer |>version is out, I'm going to start writing for it. But no matter what |>happens, windows 97 is a goner...and that makes me happy. I think you underestimate the power of reality in this matter.... -- scott lahteine <mailto:slur@world.std.com> <http://world.std.com/~slur> "No Universe is perfect which leaves no room for improvement."
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <737cd$1131d.d6@news.psrinc.com> Date: 4 Jul 1997 01:36:05 GMT Control: cancel <737cd$1131d.d6@news.psrinc.com> Message-ID: <cancel.737cd$1131d.d6@news.psrinc.com> Sender: Information Spam cancelled by sef@kithrup.com
From: NoSpa.Mgustilo@mail.med.upenn.edu Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Sun, 13 Jul 1997 00:33:30 -0400 Organization: University of Pennsylvania Message-ID: <NoSpa.Mgustilo-1307970033430001@ts3-11.upenn.edu> References: <AFE9608B-A7701@206.165.44.26> <5q16cp$abi$1@news.platinum.com> <199707100237521485904@accs-as809-dp07.atln.grid.net> In article <199707100237521485904@accs-as809-dp07.atln.grid.net>, drifterusa@macconnect.com (John Bauer) wrote: > Gary W. Longsine wrote: > > > I personally believe this [Amelio and Hancock's departure] is the best > > news since Dec. 21, 1996. > > Could you explain why? Apple bought NeXT? ----------- Bicycle Crash Test Dummy for Hire gustilo@mail.med.upenn.edu
From: drifterusa@macconnect.com (John Bauer) Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: Sun, 13 Jul 1997 02:29:20 -0500 Organization: None Message-ID: <19970713022920441835@accs-as20-dp01.dlls.grid.net> References: <AFE9608B-A7701@206.165.44.26> <5q16cp$abi$1@news.platinum.com> <199707100237521485904@accs-as809-dp07.atln.grid.net> <NoSpa.Mgustilo-1307970033430001@ts3-11.upenn.edu> <NoSpa.Mgustilo@mail.med.upenn.edu> wrote: > I wrote: > > > Gary W. Longsine wrote: > > > > > I personally believe this [Amelio and Hancock's departure] is the best > > > news since Dec. 21, 1996. > > > > Could you explain why? > > Apple bought NeXT? This is the second such answer I've gotten (though not from GWL) to my obviously ambiguous question. I wanted to know why a NeXT user thought the departure of Amelio and Hancock was such good news since without them, the NeXT purchase [apparently approved of above] might never have happened. Or is the whole statement sarcastic? John Bauer
From: lottery@sade.com Newsgroups: comp.sys.next.programmer Subject: Lottery FANS! Read This!!! Date: 13 Jul 1997 09:40:10 GMT Organization: Southwestern Bell Internet Services Message-ID: <5qa7pq$dkt@news1.rcsntx.swbell.net> Check this program out! Do you want to receive FREE instant scratcher lottery tickets up to the THOUSANDS EVERY month? then go here http://www.sunridge.net/index.phtml?id=22250 it will only take you a second! This is a great opportunity for lottery fans!!
From: jq@papoose.quick.com (James E. Quick) Newsgroups: comp.sys.next.programmer Subject: Re: mmap via vm_* ??? Date: 13 Jul 1997 06:53:34 -0400 Organization: Quick and Associates Message-ID: <5qac3e$30i@papoose.quick.com> References: <33B7AF69.167E@mpip-mainz.mpg.de> In article <33B7AF69.167E@mpip-mainz.mpg.de>, Stefan Ried <ried@mpip-mainz.mpg.de> wrote: >Hi, > > >I'd like to compile mSQL 2.0 for my Next 4.2 System. > >But there is no mmap included in mach. > >NeXTAnswer 1567 explains the corresponding vm_ routines to simulate >mmap. > >Has anybody written a mmap for NeXT using the vm_ routines already ? There actually is an mmap in the base system library, but it is undocumented for a reason. There is no corresponding munmap. I've spent some time working on the mmap problem, and though I was unable to find a satisfactory solution, I can at least explain the situation. The standard mmap allows mapping memory regions of arbitrary size. The mmap (and likewise the vm_* routines) in NS* and OS* only operate on memory that is aligned to page boundaries and whose size is a multiple of the page size. For many uses this is not a problem. For other uses (such as in msql) it is a show stopper. If mSQL wrote pages out as full native pages and implemented an offset+length reference scheme in its file IO, the current implementation would work fine. Unfortunately, when initializing a new database, msql writes only the header information to a file then mmaps that file. Since the total length of the file is less than 1 page, the mmap request fails. I think there may also be another problem, but it has been so long that I forgot. I am not sure whether a MAP_SHARED will actually flush pages to disk when information is modified. The data will be correctly shared among the participating processses but I forget whether or not one process will have to explicitly write some pages to make the new data persistent. Since mmap was a late addition in 4.3, but a core component of 4.4 I am hopeful that truly compatible versions of both mmap and munmap will be present in Rhapsody. -- ___ ___ | James E. Quick jq@quick.com / / / | Quick & Associates NeXTMail O.K. \_/ (_\/ | Apple, we know the song's not written yet, ) | but could you at least hum a few more bars?
From: Alan Dail <alandail@imperium.net> Newsgroups: comp.sys.next.hardware,comp.sys.next.programmer Subject: Prelude to Rhapsody on Virtual PC Date: Sun, 13 Jul 1997 11:07:06 -0400 Organization: Imperium Internet Message-ID: <33C8EF13.25FABA4E@imperium.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit Hi, I bought Virtual PC for the sole purpose of running the Prelude to Rhapsody release of OpenStep, but after 5 days of trying, I cannot get it to install. I created a 350 meg drive, run in a 60 meg partition, start the installation, after an hour or two I make it to the point where I eject the installation floppy and reboot so that I can configure my system and complete the installation. after annother two hours (which is MUCH longer than I expected this part to take), I get the message that the reboot is complete, but never get a login window or the configuration app. It just sits there forever saying the reboot is complete without doing anything else. Has anyone been able to install OpenStep with Virtual PC and does anyone have any clue as to what could be the problem. I tried calling Connectix and was first told that they don't offer support for OpenStep and was then told that they would refer my call to their level 2 support people, but that it will take at least a week for someone to get back to me. I would greatly appreciate any help that anyone could offer as I really need to get this installed before Wednesday. I have a PowerMac 8500/120 running system 7.6.1 with 80 megs of RAM, a PC Compatibility Card, a DAT tape drive (I mention this because it appears when I try to boot OpenStep in verbose mode that it is configuring a tape drive, which I didn't expect). Please respond via email as my internet provider loses more news postings than it recieves, so I will likely never see the reply if y ou only respond via the newsgroup. Thanks! Alan Dail, developer
From: Ken MacLeod <ken@bitsko.slc.ut.us> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: GX on HyperCard (Was Re: Amelio & Hancock Leave Apple Date: 11 Jul 1997 17:06:43 -0500 Organization: PSINet Message-ID: <m3iuyhp7kc.fsf@biff.bitsko.slc.ut.us> References: <Pine.SGI.3.96.970712010949.169B-100000@wong> <AFED34D6-5E1CF@206.165.44.10> "Lawson English" <english@primenet.com> writes: > "GX competes with DPS and > therefore must go" is the attitude that I have picked up. Some have > suggested that this is pure paranoia, and perhaps it is. Not all decisions > are based on greed and shortsightedness. "GX and DPS overlap in a lot of areas and DPS already exists in OpenStep/Rhapsody so we'll go with that and merge in features from GX" is both the attitude and public statement that I've picked up. And that makes a lot of sense when you think of the whole framework that already exists. It'll be interesting to see how Latitude handles some of the technologies like GX. GX, like QuickTime and OpenDoc, was developed in Apple's period when they were writing for two OSs (MacOS and Copland) and so isn't intricately tied into the lower level MacOS API and should be fairly easy to port to Rhapsody. > Well, I just crashed the Mac while adding a new library function of my own > to the list of available functions so I'm taking a break... Funny you should mention that, I was just thinking how much faster/more convenient writing MacOS programs will be in the Blue Box. Since you can have multiple Blue Box images, you can create tailored MacOS systems for use in development, put it to sleep, and then clone it every time you need to restart. You'd be recovered from a crash in seconds. -- Ken MacLeod ken@bitsko.slc.ut.us
From: jacob<jacob@friesl.net> Newsgroups: comp.sys.next.programmer Subject: Asian Ladies Date: 13 Jul 1997 19:27:03 GMT Organization: World Access/Planet Internet Message-ID: <5qba67$rtk@halley.pi.net> I've found the pefect site with nude Asian ladies. Much more than you can find in any newsgroup. http://home.pi.net/~sappie/playboy.htm
Sender: lottery@sade.com Control: cancel <5qa7pq$dkt@news1.rcsntx.swbell.net> Message-ID: <cancel.5qa7pq$dkt@news1.rcsntx.swbell.net> Subject: cmsg cancel <5qa7pq$dkt@news1.rcsntx.swbell.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 13 Jul 1997 10:29:37 GMT Distribution: rok EMP removed by jem@xpat.com. Original Headers: From: lottery@sade.com Subject: Lottery FANS! Read This!!! Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!news-peer.gsl.net!news.gsl.net!news.he.net!news.pagesat.net!news1.rcsntx.swbell.net!usenet Lines: 5
From: a18@a.a Subject: $$$$$ NEW SYSTEM, BETTER THAN "ADD ME TO YOUR MAILING LIST" $$$$$$ Newsgroups: comp.sys.next.programmer Organization: 1 Message-ID: <33c9247f.0@news.unibe.ch> Date: 13 Jul 97 18:54:55 GMT I have participated in the standard "Please put me on your mailing list" letter and found it to be worth my time (I get 1 or 2 handfulls or letters every day) but I also found that it's results were nowhere near what I expected due mainly to people not sending in the money but rather spreading the letter without paying for it. I participated in that mainly as an experiment and found that there are a tremendous amount of people willing to do it. I thought about how to, first, eliminate the "non-pay" problem and, second, to create a monthly income. I came upon the solution and decided to start a new program. I decided that there were 3 things that this new program needed in order to work for everyone and they were: 1.) It needed to be very simple and easy for anyone to do, and, 2.) It needed to be inexpensive enough for even the poorest of people, and, 3.) It needs to be DUPLICATABLE. I think you will find this program to meet those requirements. I have put lots of thought into it and I ask that you PLEASE do NOT modify it. This WILL WORK if you follow it. This system is based on the unconditional "loaning" of money to people. Simply say, "I am loaning you this $2 as an act of goodwill to help you in your financial need, you may pay me back if and when you can." You should find 5 or more people who will send $2 to the 5 needy people on this list AND MAINTAIN 5 or more people who will do the same. You should put your name on postition number 5 and move each of the other names up one position. The name originally on position number 1 gets removed. You should be able to contact each of your 5 or more people to see if they are going to be active this month. If not then you need to find one or more people to be active in order to maintain at least 5. I am not speaking about the 5 people on the list but rather the 5 new people you have found. I would highly suggest having more than 5 in any given month. Now I know that this would be extremely easy to do since I can think of at least 20 people myself who will do this consistently. The key is to maintain at LEAST 5 active people. If you don't then you can't expect for the rest of the people to do it either and you can't expect for this system to work. This system is a no-brainer, and if someone can't afford the $10 + stamps for this then they truly ARE in need! It is OK to use the internet to find people but I think it would be easier to find them through people that you know. This way it will be easier for you to contact them every month to ask about their being active, unless someone is willing to give you their e-mail address. Here are some numbers: ----------------------------------------------------------------------------------------------------- Monthly income model per level and total for maintaining 4 active members @ $2 each: Level 1: $2 x 4 people = $8, Total now $8 Level 2: $2 x 16 people = $32, Total now $8 + $32 = $40 Level 3: $2 x 64 people = $128, Total now $8 + $32 + $128 = $168 Level 4: $2 x 256 people = $512, Total now $8 + $32 + $128 + $512 = $680 Level 5: $2 x 1024 people = $2048, Total now $8 + $32 + $128 + $512 + $2048 = *** $2728 *** Yearly income: $2728 x 12 months = $32,736 ------------------------------------------------------------------------------------------------------ Monthly income model per level and total for maintaining 5 active members @ $2 each: Level 1: $2 x 5 people = $10, Total now $10 Level 2: $2 x 25 people = $50, Total now $10 + $50 = $60 Level 3: $2 x 125 people = $250, Total now $10 + $50 + $250 = $310 Level 4: $2 x 500 people = $1000, Total now $10 + $50 + $250 + $1000 = $1310 Level 5: $2 x 2500 people = $5000, Total now $10 + $50 + $250 + $1000 + $5000 = *** $6310 *** Yearly income: $6310 x 12 months = $75,720 ------------------------------------------------------------------------------------------------------ Monthly income model per level and total for maintaining 6 active members @ $2 each: Level 1: $2 x 6 people = $12, Total now $12 Level 2: $2 x 36 people = $72, Total now $12 + $72 = $84 Level 3: $2 x 216 people = $432, Total now $12 + $72 + $432 = $516 Level 4: $2 x 1296 people = $2592, Total now $12 + $72 + $432 + $2592 = $3108 Level 5: $2 x 7776 people = $15552, Total now $12 + $72 + $432 + $2592 + $15552 = *** $18660 *** Yearly income: $18660 x 12 months = $223,920 ------------------------------------------------------------------------------------------------------ Monthly income model per level and total for maintaining 7 active members @ $2 each: Level 1: $2 x 7 people = $14, Total now $14 Level 2: $2 x 49 people = $98, Total now $14 + $98 = $112 Level 3: $2 x 343 people = $686, Total now $14 + $98 + $686 = $798 Level 4: $2 x 2401 people = $4802, Total now $14 + $98 + $686 + $4802 = $5600 Level 5: $2 x 16807 people = $33614, Total now $14 + $98 + $686 + $4802 + $33614 = *** $39214 *** Yearly income: $39214 x 12 months = $470,568 ------------------------------------------------------------------------------------------------------ Keep in mind that it does not matter what day of the month that someone chooses to be active. It DOES matter that they are active on that day EVERY month. The key to this is DUPLICATION! You must treat this as a business. If you treat it like a hobby that is how it will treat you. You could even organize small meetings with your people and their prospects and work with your leaders. Think of how easy this would be for you, how reasonable this is, and of how good the chances are of it working for you. You may need to hire someone to open all the envelopes. NOTE: I decided on $2 instead of $1 because it is more feasible and it won't matter much for someone to send $2 as opposed to $1. Also I was against $5 as that becomes too expensive to duplicate. Mail $2 every month with a piece of paper saying "I am loaning you this $2 as an act of goodwill to help you in your financial need, you may pay me back if and when you can" to the following needy people: #1 Robert Jezil 114 Jefferson Ave. Slidell, LA 70460 #2 Phil Walther Jr. 9495 Annapolis Lane North Maple Grove, MN 55369 #3 C. E. Burkman 170 University Ave. W Suite 12-129 Waterloo, Ontario N2L 3E9 #4 A. Bailey 1207 Reeves Road Plainfield, IN 46168 #5 J. Martin P.O. Box 2292 Reston, Va. 20195
From: a16@a.a Subject: $$$$$ LOAN BUSINESS, EASY MONTHLY INCOME, NO BRAINER $$$$$ Newsgroups: comp.sys.next.programmer Organization: 9 Message-ID: <33c9248e.0@news.unibe.ch> Date: 13 Jul 97 18:55:10 GMT I have participated in the standard "Please put me on your mailing list" letter and found it to be worth my time (I get 1 or 2 handfulls or letters every day) but I also found that it's results were nowhere near what I expected due mainly to people not sending in the money but rather spreading the letter without paying for it. I participated in that mainly as an experiment and found that there are a tremendous amount of people willing to do it. I thought about how to, first, eliminate the "non-pay" problem and, second, to create a monthly income. I came upon the solution and decided to start a new program. I decided that there were 3 things that this new program needed in order to work for everyone and they were: 1.) It needed to be very simple and easy for anyone to do, and, 2.) It needed to be inexpensive enough for even the poorest of people, and, 3.) It needs to be DUPLICATABLE. I think you will find this program to meet those requirements. I have put lots of thought into it and I ask that you PLEASE do NOT modify it. This WILL WORK if you follow it. This system is based on the unconditional "loaning" of money to people. Simply say, "I am loaning you this $2 as an act of goodwill to help you in your financial need, you may pay me back if and when you can." You should find 5 or more people who will send $2 to the 5 needy people on this list AND MAINTAIN 5 or more people who will do the same. You should put your name on postition number 5 and move each of the other names up one position. The name originally on position number 1 gets removed. You should be able to contact each of your 5 or more people to see if they are going to be active this month. If not then you need to find one or more people to be active in order to maintain at least 5. I am not speaking about the 5 people on the list but rather the 5 new people you have found. I would highly suggest having more than 5 in any given month. Now I know that this would be extremely easy to do since I can think of at least 20 people myself who will do this consistently. The key is to maintain at LEAST 5 active people. If you don't then you can't expect for the rest of the people to do it either and you can't expect for this system to work. This system is a no-brainer, and if someone can't afford the $10 + stamps for this then they truly ARE in need! It is OK to use the internet to find people but I think it would be easier to find them through people that you know. This way it will be easier for you to contact them every month to ask about their being active, unless someone is willing to give you their e-mail address. Here are some numbers: ----------------------------------------------------------------------------------------------------- Monthly income model per level and total for maintaining 4 active members @ $2 each: Level 1: $2 x 4 people = $8, Total now $8 Level 2: $2 x 16 people = $32, Total now $8 + $32 = $40 Level 3: $2 x 64 people = $128, Total now $8 + $32 + $128 = $168 Level 4: $2 x 256 people = $512, Total now $8 + $32 + $128 + $512 = $680 Level 5: $2 x 1024 people = $2048, Total now $8 + $32 + $128 + $512 + $2048 = *** $2728 *** Yearly income: $2728 x 12 months = $32,736 ------------------------------------------------------------------------------------------------------ Monthly income model per level and total for maintaining 5 active members @ $2 each: Level 1: $2 x 5 people = $10, Total now $10 Level 2: $2 x 25 people = $50, Total now $10 + $50 = $60 Level 3: $2 x 125 people = $250, Total now $10 + $50 + $250 = $310 Level 4: $2 x 500 people = $1000, Total now $10 + $50 + $250 + $1000 = $1310 Level 5: $2 x 2500 people = $5000, Total now $10 + $50 + $250 + $1000 + $5000 = *** $6310 *** Yearly income: $6310 x 12 months = $75,720 ------------------------------------------------------------------------------------------------------ Monthly income model per level and total for maintaining 6 active members @ $2 each: Level 1: $2 x 6 people = $12, Total now $12 Level 2: $2 x 36 people = $72, Total now $12 + $72 = $84 Level 3: $2 x 216 people = $432, Total now $12 + $72 + $432 = $516 Level 4: $2 x 1296 people = $2592, Total now $12 + $72 + $432 + $2592 = $3108 Level 5: $2 x 7776 people = $15552, Total now $12 + $72 + $432 + $2592 + $15552 = *** $18660 *** Yearly income: $18660 x 12 months = $223,920 ------------------------------------------------------------------------------------------------------ Monthly income model per level and total for maintaining 7 active members @ $2 each: Level 1: $2 x 7 people = $14, Total now $14 Level 2: $2 x 49 people = $98, Total now $14 + $98 = $112 Level 3: $2 x 343 people = $686, Total now $14 + $98 + $686 = $798 Level 4: $2 x 2401 people = $4802, Total now $14 + $98 + $686 + $4802 = $5600 Level 5: $2 x 16807 people = $33614, Total now $14 + $98 + $686 + $4802 + $33614 = *** $39214 *** Yearly income: $39214 x 12 months = $470,568 ------------------------------------------------------------------------------------------------------ Keep in mind that it does not matter what day of the month that someone chooses to be active. It DOES matter that they are active on that day EVERY month. The key to this is DUPLICATION! You must treat this as a business. If you treat it like a hobby that is how it will treat you. You could even organize small meetings with your people and their prospects and work with your leaders. Think of how easy this would be for you, how reasonable this is, and of how good the chances are of it working for you. You may need to hire someone to open all the envelopes. NOTE: I decided on $2 instead of $1 because it is more feasible and it won't matter much for someone to send $2 as opposed to $1. Also I was against $5 as that becomes too expensive to duplicate. Mail $2 every month with a piece of paper saying "I am loaning you this $2 as an act of goodwill to help you in your financial need, you may pay me back if and when you can" to the following needy people: #1 Robert Jezil 114 Jefferson Ave. Slidell, LA 70460 #2 Phil Walther Jr. 9495 Annapolis Lane North Maple Grove, MN 55369 #3 C. E. Burkman 170 University Ave. W Suite 12-129 Waterloo, Ontario N2L 3E9 #4 A. Bailey 1207 Reeves Road Plainfield, IN 46168 #5 J. Martin P.O. Box 2292 Reston, Va. 20195
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: GX on HyperCard (Was Re: Amelio & Hancock Leave Apple Date: 13 Jul 1997 14:37:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFEE9D44-16B59@206.165.44.73> References: <m3iuyhp7kc.fsf@biff.bitsko.slc.ut.us> nntp://news.primenet.com/comp.sys.mac.programmer.games, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Ken MacLeod <ken@bitsko.slc.ut.us> said: > > Funny you should mention that, I was just thinking how much > faster/more convenient writing MacOS programs will be in the Blue Box. > Since you can have multiple Blue Box images, you can create tailored > MacOS systems for use in development, put it to sleep, and then clone > it every time you need to restart. You'd be recovered from a crash in > seconds. Very nice feature, but I'm using a 7100/66av... ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: luomat+next+usenet@luomat.peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.programmer Subject: Re: GX on HyperCard (Was Re: Amelio & Hancock Leave Apple Date: 13 Jul 1997 21:52:27 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Message-ID: <5qbimr$d1r$1@ha2.rdc1.nj.home.com> References: <Pine.SGI.3.96.970712010949.169B-100000@wong> <AFED34D6-5E1CF@206.165.44.10> <m3iuyhp7kc.fsf@biff.bitsko.slc.ut.us> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In <m3iuyhp7kc.fsf@biff.bitsko.slc.ut.us> Ken MacLeod wrote: > "Lawson English" <english@primenet.com> writes: Good lord, enough already..... Lawson must have his retirement portfolio entirely in GX stock, he won't give up, he won't listen to others, stick him in the kill file and let him argue to himself..... It's been a long time since I have seen anyone who simply refused to accept that they were not going to get their way... well, my nephew, but he's only 1.5yrs old..... TjL ps -- what we need is a Usenet filter that catches all the responses to posts made my people in the killfile.... -- TjL <luomat@peak.org>
From: nbsystems@usa.net (National Business Systems) Newsgroups: comp.sys.next.marketplace,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.northstar,comp.sys.novell,comp.sys.nsc.32k,comp.sys.oric Subject: << IMPROVE YOUR CREDIT - NOW! >> Date: Sun, 13 Jul 1997 20:58:27 GMT Organization: Pacific Bell Internet Services Message-ID: <33c941a9.0@blush.jps.net> DO YOU WANT GREAT CREDIT? 10 Incredible new secrets to improve your credit rating. Money back guarantee. Send $8 to NB Systems PO Box 12213, Oakland, CA 94604
From: Wesley Horner <wesman@azrael.uoregon.edu> Newsgroups: comp.sys.next.hardware,comp.sys.next.programmer Subject: Re: Prelude to Rhapsody on Virtual PC Date: 14 Jul 1997 00:53:15 GMT Organization: University of Oregon, Eugene Message-ID: <5qbt9r$m27@pith.uoregon.edu> References: <33C8EF13.25FABA4E@imperium.net> I think I know the problem that you are talking about. I have been dealing with the same problem for a couple of days. I have a 8500/120 with 256k cache. I got the same thing when I rebooted. Then I rebooted with the -v boot option and discovered that the thing boots entirely and then tries to start the display server. The problem is that it time out trying to open the port because my system is too slow. It was taking roughly an hour and a half for a reboot. I then tried it on a powercenter 150 with 512k cache. My boot time dropped to 3 minutes. I think the main constraint is the cache. You would do better by investing in cache than in a cpu. The other concern is ram. Give VPC at leat 45megs if you can. IT ran too slow with less than that and you need a fair amount to do color. I also wouldn't install the network driver in openstep because it is troublesome at startup. Hope this helps, wes In comp.sys.next.hardware Alan Dail <alandail@imperium.net> wrote: > Hi, > I bought Virtual PC for the sole purpose of running the Prelude to > Rhapsody release of OpenStep, but after 5 days of trying, I cannot get > it to install. I created a 350 meg drive, run in a 60 meg partition, > start the installation, after an hour or two I make it to the point > where I eject the installation floppy and reboot so that I can configure > my system and complete the installation. after annother two hours > (which is MUCH longer than I expected this part to take), I get the > message that the reboot is complete, but never get a login window or the > configuration app. It just sits there forever saying the reboot is > complete without doing anything else. Has anyone been able to install > OpenStep with Virtual PC and does anyone have any clue as to what could > be the problem. > I tried calling Connectix and was first told that they don't offer > support for OpenStep and was then told that they would refer my call to > their level 2 support people, but that it will take at least a week for > someone to get back to me. > I would greatly appreciate any help that anyone could offer as I really > need to get this installed before Wednesday. I have a PowerMac 8500/120 > running system 7.6.1 with 80 megs of RAM, a PC Compatibility Card, a DAT > tape drive (I mention this because it appears when I try to boot > OpenStep in verbose mode that it is configuring a tape drive, which I > didn't expect). > Please respond via email as my internet provider loses more news > postings than it recieves, so I will likely never see the reply if y ou > only respond via the newsgroup. > Thanks! > Alan Dail, developer -- ~~~~wesman@gladstone.uoregon.edu~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Vax a viscious creature known to eat 110AC and quotes through its *DCL*. Vax are usually found in groups of Vaxen called clusters where they lay in wait to ravage thier prey known as users.
Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer From: nagle@netcom.com (John Nagle) Subject: Re: GX on HyperCard (Was Re: Amelio & Hancock Leave Apple Message-ID: <nagleEDA9sy.Ms@netcom.com> Organization: Netcom On-Line Services References: <m3iuyhp7kc.fsf@biff.bitsko.slc.ut.us> <AFEE9D44-16B59@206.165.44.73> Date: Mon, 14 Jul 1997 00:38:58 GMT Sender: nagle@netcom6.netcom.com "Lawson English" <english@primenet.com> writes: >Ken MacLeod <ken@bitsko.slc.ut.us> said: >> Funny you should mention that, I was just thinking how much >> faster/more convenient writing MacOS programs will be in the Blue Box. >> Since you can have multiple Blue Box images, you can create tailored >> MacOS systems for use in development, put it to sleep, and then clone >> it every time you need to restart. You'd be recovered from a crash in >> seconds. Or you could use Executor under NT or the Mac emulator on the BeBox to get the same result right now. John Nagle
From: thrall@serv.net (Dean Johnson) Newsgroups: comp.sys.next.programmer Subject: Project Builder and C++? Date: 14 Jul 1997 02:16:57 GMT Organization: Not Sure Yet Message-ID: <5qc26p$1no$1@brockman.serv.net> Is it possible to write straight old c++ code in ProjectBuilder? I am taking a c++ class and we are just writing console type programs for data structures and drivers (linked lists, new math functions, etc). I would like to use ProjectBuilder instead of Edit and I definetely don't want to resort to using Visual C++. Any tips - suggestions for this. I did try but I received some 200 errors in some c++ code that I stuck into ProjectBuilder :( Thanks for any info. dean johnson
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33c9248e.0@news.unibe.ch> Date: 14 Jul 1997 00:43:11 GMT Control: cancel <33c9248e.0@news.unibe.ch> Message-ID: <cancel.33c9248e.0@news.unibe.ch> Sender: a16@a.a MMF cancelled by sef@kithrup.com
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33c9247f.0@news.unibe.ch> Date: 14 Jul 1997 00:43:10 GMT Control: cancel <33c9247f.0@news.unibe.ch> Message-ID: <cancel.33c9247f.0@news.unibe.ch> Sender: a18@a.a MMF cancelled by sef@kithrup.com
Newsgroups: comp.sys.next.programmer From: wolfram@iontichy.freinet.de (Wolfram Baumann) Subject: DO with Nextstep 3.3 and Openstep 4.1 Message-ID: <ED9y5x.1M3@iontichy.freinet.de> Keywords: DO, distributed objects Sender: wolfram@iontichy.freinet.de (Wolfram Baumann) Organization: private Date: Sun, 13 Jul 1997 20:27:33 GMT I have a program compiled under Nextstep 3.3 acting as a Server for distributed objects ( DOs ) and a program compiled under Openstep 4.1 which acts as a client. When the Openstep 4.1 client tries to connect via "[ NSConnection rootProxyForConnectionWithRegisteredName: ... ]" to the Nexstep 3.3 server the following message is shown on the console : *** Uncaught exception: <NSInvalidReceivePortException> connection is invalid A simple Nextstep 3.3 client could connect to the 3.3 server without an error. Does anybody know a solution to this problem, e.g. using an other coding mechanism on the Openstep 4.1 ?
Sender: jacob<jacob@friesl.net> Control: cancel <5qba67$rtk@halley.pi.net> Message-ID: <cancel.5qba67$rtk@halley.pi.net> Subject: cmsg cancel <5qba67$rtk@halley.pi.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 13 Jul 1997 19:36:19 GMT Distribution: rok EMP removed by jem@xpat.com. Original Headers: From: jacob<jacob@friesl.net> Subject: Asian Ladies Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!europa.clark.net!dispatch.news.demon.net!demon!bullseye.news.demon.net!demon!newsgate.unisource.nl!halley.pi.net!news Lines: 10
Sender: nbsystems@usa.net (National Business Systems) Control: cancel <33c941a9.0@blush.jps.net> Message-ID: <cancel.33c941a9.0@blush.jps.net> Subject: cmsg cancel <33c941a9.0@blush.jps.net> From: jem@xpat.com Newsgroups: comp.sys.next.marketplace,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.northstar,comp.sys.novell,comp.sys.nsc.32k,comp.sys.oric Date: 13 Jul 1997 23:38:12 GMT Distribution: rok EMP removed by jem@xpat.com. Original Headers: From: nbsystems@usa.net (National Business Systems) Subject: << IMPROVE YOUR CREDIT - NOW! >> Newsgroups: comp.sys.next.marketplace,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.northstar,comp.sys.novell,comp.sys.nsc.32k,comp.sys.oric Path: ...!newsfeed.direct.ca!ais.net!su-news-hub1.bbnplanet.com!news.bbnplanet.com!news.pbi.net!news.pacbell.net!blush.jps.net!ptest Lines: 4
From: dix.lorenz@hamburg.netsurf.de (Dix Lorenz) Newsgroups: comp.sys.next.programmer Subject: Re: Project Builder and C++? Date: Mon, 14 Jul 1997 09:21:40 +0200 Organization: EDV - Beratung Message-ID: <199707140921407230500@dip099-2.hamburg.netsurf.de> References: <5qc26p$1no$1@brockman.serv.net> Dean Johnson <thrall@serv.net> wrote: > Is it possible to write straight old c++ code in ProjectBuilder? > > I am taking a c++ class and we are just writing console type programs > for data structures and drivers (linked lists, new math functions, > etc). I would like to use ProjectBuilder instead of Edit and I > definetely don't want to resort to using Visual C++. > > Any tips - suggestions for this. I did try but I received some 200 > errors in some c++ code that I stuck into ProjectBuilder :( > > Thanks for any info. > > dean johnson I have used C++ in OpenStep programs using ProjectBuilder, no big problem. But I wrote the whole Interface using IB and Objective-C, only using C++ for the core calculations. I don't know about writing "pure" ANSI-C++ programs. Greetings, Dix
From: alanf@izzy.net Newsgroups: comp.sys.next.programmer Subject: inter-nib messaging? Date: 14 Jul 1997 16:29:22 GMT Organization: "Comshare, Inc." Message-ID: <5qdk52$iei$2@inet-prime.comshare.com> Greetings ethereal composite mind, I'm working on an app, using the 3.2 dev tools on black, trying to send a message between nib's, and I'm getting: unrecognized -updateLocalNumberDisplay message sent to instance (0x3e028) of PhoneController Class The sending object is messaging Files Owner. Any thoughts? Thanks in advance, Alan Frabutt (alanf@izzy.net)
From: "cii Web Services" <advertising@ciiweb.com> Newsgroups: comp.programming,comp.programming.threads,comp.sys.next.programmer,cz.comp.lang.basic.visual,es.comp.lenguajes.visual-basic,microsoft.public.activex.programming.control.dev,microsoft.public.activex.programming.scripting.engines,microsoft.public.act Subject: The ULTIMATE Visual Basic page! Date: Mon, 14 Jul 1997 13:11:57 -0400 Message-ID: <5qdmro$df8@newsfep2.sprintmail.com> Visit the biggest and best Visual Basic web page at http://www.ciiweb.com/vb This page has everything you've ever wanted to know about Visual Basic and related programming concepts and techniques! Please remember to bookmark the site, as it will update often!
From: Žoloft@hegel1.cs.chalmers.se.cs.chalmers.se (Olof Torgersson) Newsgroups: comp.sys.next.programmer Subject: cut copy and paste toolbuttons Date: 14 Jul 1997 16:34:35 GMT Organization: Chalmers University of Technology Message-ID: <5qdker$1h2$1@nyheter.chalmers.se> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi I'm trying to create an interface with toolbuttons to cut copy and paste text. Toolbuttons like this can be found in many windows applications like MS Word and Excel. I tried to connect the actions of the buttons in IB to the First Responders cut: copy: and paste: actions respectively. However nothing happens as it seems like the buttons become the first responder instead of the view with the selected text. (I'm using 4.0 but will be upgrading to 4.2 soon) I tried the same thing in NextStep 3.3 and it seems to work fine. Has anyone tried to do this on NT? Olof Torgersson --- Olof Torgersson oloft@cs.chalmers.se Department of Computing Science +46 31 772 54 06 Göteborg University & Chalmers University of Technology S-412 96 GÖTEBORG, SWEDEN
From: rog@ohm.york.ac.uk (Roger Peppe) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 14 Jul 1997 16:31:59 GMT Organization: Department of Electronics, University of York, UK. Sender: rp9@york.ac.uk Message-ID: <5qdk9v$ghg$1@netty.york.ac.uk> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup <5pk1a1$ngr@buggy.news.easynet.net> On 4 Jul 1997 23:34:25 GMT, Frederic Stark <stark@easynet.fr> wrote: > While we are at it, I hope they loose the source of the integration of gdb > in project builder (hint, hint, guys at apple, "rm -rf"). Terminal does a > much better job at handling text/selections/Alt-k. gdb window in PB is > soooo slooooow. And is often inacurate if a lot of text is dumped on it > (some output is just lost. Nice isn't it?). And when you click on the > middle of the text and type, nothing happends. Ghosh. *Who* did that ? > Or try to use the keyboard in gdb window: type 'n', and this half-dumb > project builder make your source key window, so you can happilly override > your source with gdb commands, Happends to me 3 or 4 times per debugging > session. [Ok at least, now we can undo...] i've found that a good way to get around most of the problems with Project Builder is to use it *only* for adding and removing files from the project (i'd avoid doing even that if it were possible...). i define a shell function: fn mk {@{cd .. && time make 'CC=cc3'} && `{echo gdb '-d '^$gdbpath ./MyApplication}}; gdbpath=(.. ../*.subproj) (or something similar, in whatever shell i'm using [in this case rc]) and then just use a terminal window, which provides me with infinite history, debugging output to the same terminal, and all the other advantages that Terminal gives. for editing of source files i use a terminal based editor (vi); i'm usually editing many files at the same time. using the above allows me almost complete immunity from Project Builder upgrades - i hardly even noticed the update for 4.0... better integration would be nice though - i've often thought about building an acme/wily clone under nextstep; has anyone else ? rog.
From: alanf@izzy.net Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 14 Jul 1997 16:22:18 GMT Organization: "Comshare, Inc." Message-ID: <5qdjnq$iei$1@inet-prime.comshare.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <knkbVMS00iWW0465o0@andrew.cmu.edu> <199707092338411585593@dialup96-5-11.swipnet.se> Cc: lars.farm@ite.mh.se In <199707092338411585593@dialup96-5-11.swipnet.se> Lars Farm wrote: <snip> I've been following this discussion on and off, and haven't seen the issue of Distributed Objects addressed. Forgive me if I missed a post, but the scope of the "Conservative GC" discussion seems to be limited to one machines memory space. Did I miss something? Regards, Alan Frabutt (alanf@izzy.net) disclamer: my opinions, not my employers.
From: lars.farm@ite.mh.se (Lars Farm) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Mon, 14 Jul 1997 20:28:33 +0200 Organization: pv Message-ID: <19970714202833592514@dialup169-2-2.swipnet.se> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <knkbVMS00iWW0465o0@andrew.cmu.edu> <199707092338411585593@dialup96-5-11.swipnet.se> <5qdjnq$iei$1@inet-prime.comshare.com> NNTP-Posting-User: s-49817 <alanf@izzy.net> wrote: > In <199707092338411585593@dialup96-5-11.swipnet.se> Lars Farm wrote: > <snip> > I've been following this discussion on and off, and haven't seen the issue of > Distributed Objects addressed. Forgive me if I missed a post, but the scope > of the "Conservative GC" discussion seems to be limited to one machines > memory space. Did I miss something? No, you didn't miss much. Others have mentioned DO briefly as a reason to avoid automatic GC, but have not explained why in a way that makes sense to an experienced Mac developer and future Rhapsody beginner. 'Garbage' in 'Garbage Collection' refers to unused RAM from the free store within your own process, not arbitrary external resource or worse arbitrary unsolved programming problem. GC solves one problem, one problem only and it does it very well. I consider that a good thing. It's quite clear what it does and what it does not do. At least to me;-) -- Lars Farm; lars.farm@ite.mh.se - Limt/Channelmatic: lars.farm@limt.se
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: How to save Apple Date: 14 Jul 1997 10:09:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFEFAFFC-20EF8@206.165.44.6> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Trial balloon time here: Many people have said "Buy Apple stock" as a way of showing support for Apple. That's all very well, but given that large investors control so much of it, I don't see it as doing much good because the needs of large investors are often radically different than the needs of Apple customers (shouldn't be that way, but it often is -if Larry Ellison bought 25% of Apple, would he be motivated by the needs of educational users unless they wanted to buy his Apple-branded net computers?). So... I've come up with an unusual idea for how customers that want to see MacOS thrive can more effectively spend their money: the "MacOS R&D Foundation," which makes low-interest loans/grants to startups, individuals and companies that are working on MacOS-oriented technologies. Possible targets for such loans (and/or grants) would be companies that have gathered back together the OpenDOc/Cyberdog teams, or are willing to work with Apple to recreate GX Printing on MacOS in a way that does NOT interfere with Classic Printing but still provides the features that make GX Printing unique and worth using. Enhancements of the GX Graphics and Typography would also be nice goals to support. The details of how this would work are left as an exercise for the reader, but the fact is that Apple currently has only 25% of the R&D budget that it used to have, and certain factions within Apple appear to be driven completely by their own internal vision/version of reality, rather than by what the MacOS customer-base might be interested in using. Personally, I think educational users would be more interested in Cocoa than Web Objects or Java. Home users would be more interested in OpenDoc than Interface Builder or JavaBeans. Internet users would be more interested in a robust version of Cyberdog than in upgrading to Rhapsody to use the less-than-equivalent on faster hardware. Developers would be more interested in using GX and GX printing (or their Taligent Graphics equivalents) than in programming via Display PostScript. Hobbyist and semi-professional programmers would find HyperCard 3.0 of more use than Objective C. Etc. An *independent* R&D foundation (profit-oriented? Not-for-profit?) set up to be responsive to Mac user's wants and needs as determined by those users who bother to support the foundation (as well as via the more traditional measures of users's desires, such as marketing surveys) would be far more responsive to the needs of the market than Apple has been and would off-load some of the burden that Apple currently has to carry by itself, R&D-wise. The details of how this would be set up, whether it would be profit/not-for-profit, etc., obviously need to be worked on, but the basic idea is, I firmly believe, essential to the short- and long-term well-being of Apple and MacOS. Comments? Criticisms? Critiques? Suggestions? Etc? ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: charlesa@hyperspace (Charles Ashley) Newsgroups: comp.sys.next.programmer Subject: mmap()? Date: 14 Jul 1997 19:06:14 GMT Organization: MatriX Publishing Network Message-ID: <5qdtb6$657@nebula.mpn.com> Hi - Basic question: Does OS-4.1/I have mmap()? If not, is there anywhere to locate it? I'm trying to do a trial install of mSQL 2, but get the following: >mSQL 2 needs a working mmap(), it cannot work without it. >You cannot run mSQL 2 on this machine. Sorry. Thoughts appreciated, C.
From: LTaylor7@ix.netcom.com (Daniel L. Taylor) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Mon, 14 Jul 1997 13:00:29 -0700 Organization: Netcom Message-ID: <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> References: <AFEFAFFC-20EF8@206.165.44.6> In article <AFEFAFFC-20EF8@206.165.44.6>, "Lawson English" <english@primenet.com> wrote: > Trial balloon time here: > > Many people have said "Buy Apple stock" as a way of showing support for > Apple. > > That's all very well, but given that large investors control so much of > it, I don't see it as doing much good because the needs of large investors > are often radically different than the needs of Apple customers (shouldn't > be that way, but it often is -if Larry Ellison bought 25% of Apple, would > he be motivated by the needs of educational users unless they wanted to buy > his Apple-branded net computers?). OK. > So... > > I've come up with an unusual idea for how customers that want to see MacOS > thrive can more effectively spend their money: the "MacOS R&D Foundation," > which makes low-interest loans/grants to startups, individuals and > companies that are working on MacOS-oriented technologies. With you so far. > Possible targets for such loans (and/or grants) would be companies that > have gathered back together the OpenDOc/Cyberdog teams, or are willing to > work with Apple to recreate GX Printing on MacOS in a way that does NOT > interfere with Classic Printing but still provides the features that make > GX Printing unique and worth using. Enhancements of the GX Graphics and > Typography would also be nice goals to support. If the "group" is going to have this "save the old OS" slant, then forget it. An R&D group set up to fund Rhapsody development would be great! If it funded consultants or start up small cloners, fine. If this R&D group also funded Mac OS development to the technologies Apple has endorsed, that's OK to, although the money is better spent on Rhapsody support. But if it's going to try and save OpenDoc and QDGX, forget it. *THESE ARE NO MORE!* They were great demos. They never caught on. OpenDoc as a paradigm had some flaws (there, I said it). Or, rather, it didn't fit today's computers well (10-20 years from now might be different). It's not going to make Apple any money, nor is it going to help Mac OS or Rhapsody fend off Windows much less fight back. The last thing we need is an Apple-only component software model to royally screw up the Yellow Box. QDGX was a great imaging model that never caught on. The publishing world would rather have DPS. So give it to them! > The details of how this would work are left as an exercise for the reader, > but the fact is that Apple currently has only 25% of the R&D budget that > it used to have, and certain factions within Apple appear to be driven > completely by their own internal vision/version of reality, rather than by > what the MacOS customer-base might be interested in using. The Mac OS customer base is interested in having a fully PMT, PM, OO OS with DPS. They are interested in seeing developers support said OS so they know it will grow and be around, and they can once again choose the superior OS rather than sell their soul to MS where they can't even choose their own browser. The Mac world desperately needs Rhapsody developers, including the start ups such a fund could really help. IT DOES NOT NEED OPENDOC OR GX! > Personally, I think educational users would be more interested in Cocoa > than Web Objects or Java. Cocoa without Web Objects/Java is a dead platform. Cocoa on Rhapsody is a good application. Fund a port of Cocoa to Rhapsody! Better yet, fund a port of its native tongue, Prograph, to Rhapsody. Then you can click-compile Cocoa to Rhapsody, and you have a revolutionary, industrial-strength iconic language on Rhapsody (one that is gasping for life due to stupid management decisions). THAT would help Apple. And I can name several people, myself included, who would be more than happy to come together and pick up the pieces and go for it given the funds. > Home users would be more interested in OpenDoc > than Interface Builder or JavaBeans. Home users weren't interested in OpenDoc when it was out and supported. They are far more interested in whether or not the latest PC game will run on Mac OS and Rhapsody. "I like Mac, but little Johnny's friends play all these games that I don't see on the Mac shelf...". Fund a port of that to Mac OS and Rhapsody! (Damn, now I'm sounding like Aristophanes...he over plays this card, but he's right, to a degree, on this none-the-less.) > Internet users would be more > interested in a robust version of Cyberdog than in upgrading to Rhapsody to > use the less-than-equivalent on faster hardware. Internet users would be far more interested in having all of the latest plug-ins and never having to apologize to any Wintel sites. They would also be far more interested in having a Yellow Box native browser which blows the socks off of Navigator and MSIE. An OpenDoc browser is, well, dead in the water. > Developers would be more > interested in using GX and GX printing (or their Taligent Graphics > equivalents) than in programming via Display PostScript. What developers? Certainly not those writing for Apple's core markets of content creation and media production. Rhapsody is targeted square at these guys. Right at their foreheads. It's targeted so well if they don't duck they might get hurt as the shrink-wrapped Rhapsody box flies through their front door and slams them right in the head. They don't give a damn about OpenDoc or Cyberdog or GX. They do give a big damn about a 100% native, PMT, PM OS running on Arthur chips which gives them PostScript ON THE SCREEN! Hey, maybe if Apple had handled things differently GX would be the cat's meow in publishing. But they didn't, and now's not the time to delude ourselves thinking that OpenDoc or GX is going to "save Apple" or do one thing to prevent that guy walking into Best Buy from buying Wintel. DPS can take Apple's 80% color publishing share and turn it into 100%. GX can take that 80% and turn it into 0%. I don't know about you, but Apple fans can't afford to spend cash losing customers from the Mac OS platform right now! > Hobbyist and > semi-professional programmers would find HyperCard 3.0 of more use than > Objective C. Hypercard would be an OK thing to fund, but don't even try to compare it to OpenStep for the semi-pro...OpenStep has them drooling just as much as the pros. > An *independent* R&D foundation (profit-oriented? Not-for-profit?) set up > to be responsive to Mac user's wants and needs as determined by those users > who bother to support the foundation (as well as via the more traditional > measures of users's desires, such as marketing surveys) would be far more > responsive to the needs of the market than Apple has been and would > off-load some of the burden that Apple currently has to carry by itself, > R&D-wise. You don't get it, do you? Apple was ***dying*** under the items you claim Mac users wanted. What Mac users want is what's sitting in Apple's labs right now. But they need developer support. Lots of it. There can't be too much given the press Apple is getting. Spend your funds sending people to stores to train salespeople about Macintosh. Spend your funds setting up consultants who can recommend Mac OS when it's the best choice (far more often than most so-called "consultants" recommend it). Spend your funds getting developers to write to Rhapsody. Spend your funds getting developers to write to Rhapsody. Repeat: Spend your funds getting developers to write to Rhapsody. Spend your funds porting a vertical market app or the next big game that will put every Mac kid at a disadvantage at his grade school if it's not ported. Spend your funds on anything other than a delusional attempt to save the "old glory" that nearly killed Apple. Right now Apple needs help in the areas I've mentioned. They don't need OpenDoc to crawl out of a grave somewhere only to have a home buyer say "What's that? I don't understand that...but they say Java's the future, and besides, I want this game over here for little Mikey...". > Comments? > > Criticisms? > > Critiques? > > Suggestions? > > Etc? The basic idea may be fine, if you can gather support. But the things to which you want to throw your support are worthless. They will not help Apple one bit. They probably won't hurt Apple...but right now mere "benevolence" is not what Apple needs from Mac fans spinning their wheels. -- Daniel L. Taylor LTaylor7@ix.netcom.com ================================================================= The Heritage Foundation /// Town Hall http://www.heritage.org/heritage/ /// http://www.townhall.com/ =================================================================
From: d89cb@efd.lth.se (Christian Brunschen) Newsgroups: comp.sys.next.programmer Subject: libg++ under OS4.2 for Mach Date: 14 Jul 1997 20:04:11 GMT Organization: Lund Institute of Technology, Sweden Message-ID: <5qe0nr$qm$1@news.lth.se> NNTP-Posting-User: d89cb I am trying to compile some software which uses libg++, under OS4.2 for Mach (on black hardware if that matters). However, I can't seem to find the header files. The libraries are there, in /usr/lib -- static and dynamic -- but the header files seem to be ... missing. I checked NeXTAnswers, but didn't find anything, and the documentation in DL seems to be rather, um old :) Doe sanyone know anything more ... ? (I have also made one attempt at compiling libg++ 2.7.2 myself, but I can't get that to quite work, either ... Best regards // Christian Brunschen
From: ga@ed4u.com (G. Apple) Newsgroups: comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 14 Jul 1997 23:55:33 GMT Organization: Advanced Communications Engineering, Inc. Message-ID: <5qee9l$fko$1@news7.gte.net> References: <19970713022920441835@accs-as20-dp01.dlls.grid.net> <EDBGup.BI3@gateway.ali.bc.ca> In article <EDBGup.BI3@gateway.ali.bc.ca>, Michael C. Cam <mike@ali.bc.ca> wrote: > In the same breath, Ellen might have been a likely successor to the > position because she had been working very closely with Gil and was > the main motivator for the merger with Next. Because Apple board > made no indication to appoint her (no surprise since Steve is part of > the group to find a new CEO) she found herself becomming a 5th wheel > in the whole management structure. Without Gil she has no purpose in > the organization ... so off she goes too. > > Amelio leaving is good news. Ellen not even being considered for the > new CEO position is bad news. Apple not having a CEO at the moment > is bad news as they do need someone to guide the company at this > transitionary period where it is likely that many users and > developers will be very confused with the technology and product > offerings. IMHO, it is good news, not bad, that Ellen Hancock was not even considered, largely because she was "the main motivator for the merger with Next". Many of us considered and still consider that to be a very bad move, or at best, think that Apple computer paid an order of magnitude more for NeXT than it was worth. I could say a lot more, but choose to stifle myself and hopefully avoid getting engaged in flame war that I don't have time for. -- G. Gordon Apple, PhD The Ed4U Project Advanced Communications Engineering, Inc. Redondo Beach, CA
From: jpalmer@best.com (Joseph Palmer) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Mon, 14 Jul 1997 17:26:25 -0800 Organization: VideoS2 Video Storage Systems Message-ID: <jpalmer-1407971726250001@jpalmer.vip.best.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> In article <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com>, LTaylor7@ix.netcom.com (Daniel L. Taylor) wrote: > Rhapsody is targeted square at these guys. Right at their foreheads. It's > targeted so well if they don't duck they might get hurt as the > shrink-wrapped Rhapsody box flies through their front door and slams them > right in the head. In a few months we'll have to warn them to duck. LOOKOUT! IT'S COMMING! ;) > They don't give a damn about OpenDoc or Cyberdog or GX. > They do give a big damn about a 100% native, PMT, PM OS running on Arthur > chips which gives them PostScript ON THE SCREEN! > PostScript on the screen is a perfect match for ink publishers, but it doesn't matter to - Web publishers & Web users (HTML) - Music Creation (There is no Postscript in MIDI) - Audio Editing - Video Editing - Film Editing - CAD (PCB CAD is gerber, Mechanical CAD is DXF and others) - 3D - Image editing & enhancement - Industrial monitoring and control There are thousands of applications for computers that would receive no benefit from the added overhead of DPS. PostScript can be used for a number of things, but its only mission critical to print publishing programs, and anyway it is possible to place DPS in the publishing program itself, it does not need to be in the OS. J. ______________________________________________________________ Joseph Palmer | Never join a titanic battle unless CEO | you get to be the iceberg. -- J. VideoS2 | Personal web site: http://www.videos2.com | http://www.best.com/~jpalmer
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: Mon, 14 Jul 1997 17:37:33 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <0nmdkRq00Uzx825xox@andrew.cmu.edu> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> <5pb22f$na6$1@oxygen.technet.net> <5ph9gd$5dp$1@owl.slip.net> <5phqhq$ied$1@vader.wolfware.ipc.net> <5pi60t$o16$1@news2.digex.net> <AnjN1ai00iWl05Awc0@andrew.cmu.edu> <knkbVMS00iWW0465o0@andrew.cmu.edu> <199707092338411585593@dialup96-5-11.swipnet.se> <5qdjnq$iei$1@inet-prime.comshare.com> Excerpts from netnews.comp.sys.next.programmer: 14-Jul-97 Re: [Q] OpenStep Garbage Co.. by alanf@izzy.net > I've been following this discussion on and off, and haven't seen the issue > of Distributed Objects addressed. Forgive me if I missed a post, but the > scope of the "Conservative GC" discussion seems to be limited to one > machines memory space. Did I miss something? Not at all-- a conservative GC only tracks references within a single address space, and won't be aware of external references like (P)DO, shared memory, and similar things. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 15 Jul 1997 00:02:30 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5qeemm$45g$1@owl.slip.net> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> In comp.sys.next.programmer Daniel L. Taylor <LTaylor7@ix.netcom.com> wrote: > Better yet, fund a port of its native tongue, Prograph, to Rhapsody. Then > you can click-compile Cocoa to Rhapsody, and you have a revolutionary, > industrial-strength iconic language on Rhapsody (one that is gasping for > life due to stupid management decisions). THAT would help Apple. And I can > name several people, myself included, who would be more than happy to come > together and pick up the pieces and go for it given the funds. I agree that a more visual language would be great, but as a data-flow system Prograph is the wrong direction. I'd like to see a pure-OO prototype-based language and dev environment closer to Self. Then we could do all development at runtime. I've tried to pitch this idea to people at Apple, but I get the feeling that inovation may not be as valued as it was in the early days of NeXT. -- Steve Dekorte - OpenStep consultant - San Francisco
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: cut copy and paste toolbuttons Date: 14 Jul 1997 23:08:34 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5qebhi$3a5$1@owl.slip.net> References: <5qdker$1h2$1@nyheter.chalmers.se> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit Olof Torgersson <Žoloft@hegel1.cs.chalmers.se.cs.chalmers.se> wrote: > Hi > I'm trying to create an interface with toolbuttons to cut copy and paste > text. Toolbuttons like this can be found in many windows applications like > MS Word and Excel. > I tried to connect the actions of the buttons in IB to the > First Responders cut: copy: and paste: actions respectively. > However nothing happens as it seems like the buttons become the first > responder instead of the view with the selected text. I don't know how the edit menus do this. It's interesting that they don't work when copied into a popup menu. Here's a hack though(it avoids switching the responder to the button): @implementation ResponderSmartWindow : NSWindow - makeFirstResponder:anObject { if ( [anObject respondsTo:@selector(action)] ) { if ( [anObject action] == @selector(cut:) || [anObject action] == @selector(copy:) || [anObject action] == @selector(paste:) ) { return self; } } return [self makeFirstResponder:anObject]; } @end -- Steve Dekorte - OpenStep consultant - San Francisco
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: Project Builder and C++? Date: 15 Jul 1997 00:11:05 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5qef6p$45g$2@owl.slip.net> References: <5qc26p$1no$1@brockman.serv.net> Dean Johnson <thrall@serv.net> wrote: > Is it possible to write straight old c++ code in ProjectBuilder? >.. > Any tips - suggestions for this. I did try but I received some 200 > errors in some c++ code that I stuck into ProjectBuilder :( It should work fine. Did you get it do compile on gcc first? Maybe your errors have to do with compiler differences (C++ is notorious for this) and not with ProjectBuilder. -- Steve Dekorte - OpenStep consultant - San Francisco
From: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 14 Jul 1997 22:04:06 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5qelqm$csf$1@sps1.phys.vt.edu> References: <19970713022920441835@accs-as20-dp01.dlls.grid.net> <EDBGup.BI3@gateway.ali.bc.ca> <5qee9l$fko$1@news7.gte.net> In article <5qee9l$fko$1@news7.gte.net>, ga@ed4u.com (G. Apple) wrote: > IMHO, it is good news, not bad, that Ellen Hancock was not even > considered, largely because she was "the main motivator for the merger > with Next". Many of us considered and still consider that to be a very > bad move, or at best, think that Apple computer paid an order of magnitude > more for NeXT than it was worth. Pardon me while I laugh hysterically. Apple should have bought NeXT a long time ago, but it was too busy digging its own grave. I'd _really_ like to know what you think Apple _should_ have done.
From: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 15 Jul 1997 00:15:36 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5qeth8$gb1$1@sps1.phys.vt.edu> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <5qeemm$45g$1@owl.slip.net> In article <5qeemm$45g$1@owl.slip.net>, Steve Dekorte <dekorte@slip.net> wrote: > I agree that a more visual language would be great, but as a data-flow > system Prograph is the wrong direction. I'd like to see a pure-OO > prototype-based language and dev environment closer to Self. Or better yet, Cecil. > Then we could do all development at runtime. The nirvana that Smalltalk programmers have enjoyed for decades.. > I've tried to pitch this > idea to people at Apple, but I get the feeling that inovation may not > be as valued as it was in the early days of NeXT. Well, they are somewhat busy trying to put out an OS right now. But I do tend to doubt that they'll put much priority on this sort of thing. Quite unfortunate. The NEXTSTEP development environment needs to regain its competitive edge.
From: Stefan Ried <ried@mpip-mainz.mpg.de> Newsgroups: comp.sys.next.programmer Subject: Re: mmap()? Date: Tue, 15 Jul 1997 08:58:12 +0200 Organization: Max Plank Institut for Polymer Research Message-ID: <33CB1F84.167E@mpip-mainz.mpg.de> References: <5qdtb6$657@nebula.mpn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Charles Ashley wrote: > > Hi - > > Basic question: > > Does OS-4.1/I have mmap()? If not, is there anywhere to locate it? I'm > trying to do a trial install of mSQL 2, but get the following: > > >mSQL 2 needs a working mmap(), it cannot work without it. > >You cannot run mSQL 2 on this machine. Sorry. > > Thoughts appreciated, > > C. I've just compiled msql on NeXT 4.2. Use this mmap below. I send email to the autors of msql to include it in their future release and got no reply. stefan ------------------------------------- /* * @(#)map.c 1.0 of 20 December 1996 * * Copyright (c) 1996 by Fabien Roy. * Written by Fabien Roy and Robert Ehrlich. * Fabien_Roy@free.fdn.fr Robert.Ehrlich@inria.fr * Not derived from licensed software. * * Permission is granted to anyone to use this software for any * purpose on any computer system, and to redistribute it freely, * subject to the following restrictions: * * 1. The author is not responsible for the consequences of use of * this software, no matter how awful, even if they arise * from defects in it. * * 2. The origin of this software must not be misrepresented, either * by explicit claim or by omission. * * 3. Altered versions must be plainly marked as such, and must not * be misrepresented as being the original software. * */ #include <sys/types.h> #include <sys/mman.h> #include <stdlib.h> #include <syscall.h> caddr_t mmap(caddr_t addr, size_t len, int prot, int flags, int fd, off_t off) { int pagelessone = getpagesize() -1; int size; caddr_t pageaddress; /* round to next page size */ size = (len + pagelessone) & ~pagelessone; /* allocate aligned pages */ if (!(pageaddress = (caddr_t) valloc(size))) return (caddr_t) -1; /* map it */ if (syscall(SYS_mmap, pageaddress, size, prot, flags, fd, off)){ free(pageaddress); return (caddr_t) -1; } return pageaddress; } void munmap(caddr_t addr, size_t len) { syscall(SYS_munmap,addr,len); free(addr); } -- ______________________________________________________________________ /Stefan Ried, MPI f. Polymerforschung, Postf.3148, 55021 Mainz, F.R.G. \ | ... openstep, the biggest step | | E-Mail ried@mpip-mainz.mpg.de (MIME welcome) ...since the invention | | Telefon ++49 6131 379 267 Fax:++49 6131 379 340 ...of the __/___/ | | Project working on pattern-formation in liquid crystals /./\__/\\| | WWW http://www-theory.mpip-mainz.mpg.de/~ried ...wheel\_/ \_/| \______________________________________________________________________/
From: Stefan Ried <ried@mpip-mainz.mpg.de> Newsgroups: comp.sys.next.programmer Subject: msql Adaptor Date: Tue, 15 Jul 1997 09:16:26 +0200 Organization: Max Plank Institut for Polymer Research Message-ID: <33CB23CA.41C6@mpip-mainz.mpg.de> References: <5qdtb6$657@nebula.mpn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: charlesa@ncast.co.uk Hi, Is anybody out there working on a EOF Adaptor to mSQL 2.0 ??? See the home of the msql. Version 2.0 was just released and has many new features: http://www.Hughes.com.au/software/msql2/ stefan ______________________________________________________________________ /Stefan Ried, MPI f. Polymerforschung, Postf.3148, 55021 Mainz, F.R.G. \ | ... openstep, the biggest step | | E-Mail ried@mpip-mainz.mpg.de (MIME welcome) ...since the invention | | Telefon ++49 6131 379 267 Fax:++49 6131 379 340 ...of the __/___/ | | Project working on pattern-formation in liquid crystals /./\__/\\| | WWW http://www-theory.mpip-mainz.mpg.de/~ried ...wheel\_/ \_/| \______________________________________________________________________/ leted the main that ProjectBuilder sets up automatically 3. Imported my main and .cc files into Other Sources 4. Imported the .h files into Headers 5. Set the Linker option in the Inspector to -lg++ Build went just fine after that Dean Johnson
Subject: Re: How to split classes hierarhy to bundles Newsgroups: comp.sys.next.programmer References: <33C0F8E2.273C@rtr.transit.ru> <5q5ugu$l16$1@owl.slip.net> In-Reply-To: <5q5ugu$l16$1@owl.slip.net> From: marco@sente.ch (Marco Scheurer) Organization: Sen:te Message-ID: <33cb38a7.0@epflnews.epfl.ch> Date: 15 Jul 97 08:45:27 GMT On 07/11/97, Steve Dekorte wrote: >Iracly <iracly@rtr.transit.ru> wrote: >> Is anybody have idea how to divide class hierarhy to different loadable >> bundles. > >This works fine for me when using frameworks. >You might try using frameworks instead of bundles. > Yes, but then you loose the ability of dynamically loading code in your application. To have the best of two worlds, we put our classes in frameworks, and we make a bundle which - uses the framework, and - describes its "capabilities" (for instance what classes are available) in a property list. Then your application can, through the bundle, dynamically load code from the framework. (Frankly, I don't understand why you cannot dynamically load code from a framework in the first place, or maybe you can an I'm missing something) -- Marco Scheurer (marco@sente.ch) Sen:te
Sender: "gaviota" <gaviota@telcel.net.ve> Control: cancel <5qelf7$js1$155@smarty.telcel.net.ve> Message-ID: <cancel.5qelf7$js1$155@smarty.telcel.net.ve> Subject: cmsg cancel <5qelf7$js1$155@smarty.telcel.net.ve> From: jem@xpat.com Newsgroups: comp.sys.amiga.audio,comp.sys.amiga.games,comp.sys.amiga.applications,comp.sys.amiga.marketplace,comp.sys.amiga.programmer,comp.sys.amiga.multimedia,comp.sys.amiga.introduction,comp.sys.amiga.cd32,comp.sys.amiga.uucp,comp.sys.amiga.reviews,comp.sys.amiga.announce,comp.sys.unisys,comp.sys.next,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.sysadmin,comp.sys.next.software,comp.sys.next.misc,comp.sys.next.marketplace,comp.sys.next.bugs,comp.sys.next.programmer Date: 15 Jul 1997 02:09:18 GMT EMP article removed by jem@xpat.com. Original Headers: From: "gaviota" <gaviota@telcel.net.ve> Subject: Add this success to you portfolio Venture.exe (1/1) Newsgroups: comp.sys.amiga.audio,comp.sys.amiga.games,comp.sys.amiga.applications,comp.sys.amiga.marketplace,comp.sys.amiga.programmer,comp.sys.amiga.multimedia,comp.sys.amiga.introduction,comp.sys.amiga.cd32,comp.sys.amiga.uucp,comp.sys.amiga.reviews,comp.sys.amiga.announce,comp.sys.unisys,comp.sys.next,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.sysadmin,comp.sys.next.software,comp.sys.next.misc,comp.sys.next.marketplace,comp.sys.next.bugs,comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!news-peer.sprintlink.net!news-sea-19.sprintlink.net!news-in-west.sprintlink.net!news.sprintlink.net!Sprint!206.48.41.100!news.telcel.net.ve!not-for-mail Lines: 3297
From: LTaylor7@ix.netcom.com (Daniel L. Taylor) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Tue, 15 Jul 1997 05:03:10 -0700 Organization: Netcom Message-ID: <LTaylor7-1507970503100001@vic-ca1-13.ix.netcom.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <5qeemm$45g$1@owl.slip.net> In article <5qeemm$45g$1@owl.slip.net>, Steve Dekorte <dekorte@slip.net> wrote: > In comp.sys.next.programmer Daniel L. Taylor <LTaylor7@ix.netcom.com> wrote: > > Better yet, fund a port of its native tongue, Prograph, to Rhapsody. Then > > you can click-compile Cocoa to Rhapsody, and you have a revolutionary, > > industrial-strength iconic language on Rhapsody (one that is gasping for > > life due to stupid management decisions). THAT would help Apple. And I can > > name several people, myself included, who would be more than happy to come > > together and pick up the pieces and go for it given the funds. > > I agree that a more visual language would be great, but as a data-flow > system Prograph is the wrong direction. I BEG your pardon ;-) > I'd like to see a pure-OO > prototype-based language and dev environment closer to Self. > Then we could do all development at runtime. I've tried to pitch this > idea to people at Apple, but I get the feeling that inovation may not > be as valued as it was in the early days of NeXT. Let's see both. A fund like Lawson suggested could do that. But not if it's going to try and save OpenDoc and GX.... -- Daniel L. Taylor LTaylor7@ix.netcom.com ================================================================= The Heritage Foundation /// Town Hall http://www.heritage.org/heritage/ /// http://www.townhall.com/ =================================================================
From: LTaylor7@ix.netcom.com (Daniel L. Taylor) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Tue, 15 Jul 1997 05:09:25 -0700 Organization: Netcom Message-ID: <LTaylor7-1507970509260001@vic-ca1-13.ix.netcom.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> In article <jpalmer-1407971726250001@jpalmer.vip.best.com>, jpalmer@best.com (Joseph Palmer) wrote: > > They don't give a damn about OpenDoc or Cyberdog or GX. > > They do give a big damn about a 100% native, PMT, PM OS running on Arthur > > chips which gives them PostScript ON THE SCREEN! > > > > PostScript on the screen is a perfect match for ink publishers, > but it doesn't matter to > > - Web publishers & Web users (HTML) > - Music Creation (There is no Postscript in MIDI) > - Audio Editing Granted. But do these people care about OpenDoc or GX? Or would they rather have some cool new Rhapsody native tools? > - Video Editing > - Film Editing > - CAD (PCB CAD is gerber, Mechanical CAD is DXF and others) > - 3D > - Image editing & enhancement > - Industrial monitoring and control Again, they may not care if DPS is the imaging model or not (though I think it might offer some advantages to some of the above despite the fact that the output is not PS). But do they care about GX? OpenDoc? Let's fund tools for these people, Rhapsody tools that take full advantage of PMT and PM and the nature of Rhapsody. Let's not try and breath life into those things that failed to "save Apple" in the past. > There are thousands of applications for computers that > would receive no benefit from the added overhead of DPS. Lawson has told all of us many times that DPS slows things down. But you know what? If it does, I certainly haven't seen it. NextStep runs better on an '040 than Windows does on a 200 MHz Pentium. And I dare say better than Mac OS on an '040. > PostScript can be used for a number of things, but its only > mission critical to print publishing programs, and anyway > it is possible to place DPS in the publishing program itself, > it does not need to be in the OS. Lawson wants to fund GX development. Does GX matter to these people anymore than DPS? No. The people who care about the imaging model want DPS...so why waste money trying to give them GX? They're never going to use it. If we are going to help Apple, let's fund native Rhapsody development. -- Daniel L. Taylor LTaylor7@ix.netcom.com ================================================================= The Heritage Foundation /// Town Hall http://www.heritage.org/heritage/ /// http://www.townhall.com/ =================================================================
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer,comp.sys.next.bug Subject: Cascading windows..hidden feature Date: 15 Jul 1997 12:17:39 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5qfpp3$iaq$1@concorde.ctp.com> Hi, if you ever wondered how cascading windows works when you open multiple ocuments inside your app...the you should be ware of an "undocumented" feature of NSWindow's cascadeTopLeftFromPoint method. position.x = position.y = 0.0; position = [previousWin cascadeTopLeftFromPoint:position]; [newWindow setFrameTopLeftPoint:position]; Passing the point zero forces a window to return a point cascaidng from its own top-left point...which is quite neat. BUT...this conflicts with the official documentation of NSWindows behavior where topLeftPoint is screen coordinates! IMHO the documentation has a "bug" and should be updated to properly "document" this features (which I consider a feature) Aloha Tomi
From: Thomas Harte <tph1001@CL.cam.ac.uk> Newsgroups: comp.sys.next.programmer Subject: uname Date: 15 Jul 1997 13:59:12 GMT Organization: Cambridge University Message-ID: <5qfvng$ki0@lyra.csx.cam.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii totally new to NEXTSTEP and even to GNU cc! problem with uname. i get a compile error: bash-2.00$ cc -g -o f f.c f.c: In function `main': f.c:7: storage size of `name' isn't known and think that i may not be loading in the correct library (or libraries?) i have tried to look for uname with : find / -name \*lib\* -print | xargs nm -r | grep -in uname| more as a rather crude measure (!) but no joy. full C code below. what library am i omitting to link? any other siggestions? -- thomas harte @ computer laboratory, cambridge university; tph1001@CL.cam.ac.uk; phone: +44 1223 334628; fax: 334678; http://www.CL.cam.ac.uk/users/tph1001 for zero entropy. #include <stdio.h> #include <stdlib.h> #include <sys/utsname.h> int main(){ struct utsname name; if (!uname(&name)) exit (1); /* error in uname */ printf("\n sysname %s", name.sysname); printf("\n nodename %s", name.nodename); printf("\n release %s", name.release); printf("\n version %s", name.version); printf("\n machine %s \n", name.machine); exit(0); }
From: Christian Neuss <neuss@informatik.th-darmstadt.de.NOSPAM> Newsgroups: comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Followup-To: comp.sys.next.advocacy Date: 15 Jul 1997 15:38:47 GMT Organization: Technische Hochschule Darmstadt Message-ID: <5qg5i7$aid$3@news.th-darmstadt.de> References: <19970713022920441835@accs-as20-dp01.dlls.grid.net> <EDBGup.BI3@gateway.ali.bc.ca> <5qee9l$fko$1@news7.gte.net> <5qelqm$csf$1@sps1.phys.vt.edu> nurban@sps1.phys.vt.edu (Nathan Urban) wrote: >In article <5qee9l$fko$1@news7.gte.net>, ga@ed4u.com (G. Apple) wrote: > >> IMHO, it is good news, not bad, that Ellen Hancock was not even >> considered, largely because she was "the main motivator for the merger >> with Next". Many of us considered and still consider that to be a very >> bad move, or at best, think that Apple computer paid an order of magnitude >> more for NeXT than it was worth. > >Pardon me while I laugh hysterically. Apple should have bought NeXT a >long time ago, but it was too busy digging its own grave. I'd _really_ >like to know what you think Apple _should_ have done. Well, license Solaris of course :-P <grinning, ducking and running *<:o)> Chris (Followups redirected to c.s.n.a) -- // Christian Neuss "static typing? how quaint.." // http://www.nexttoyou.de/~neuss/ // fax: (+49) 6151 16 5472
Sender: Jonathan Hendry <jhendry@> From: jon@txenxe.com Subject: Re: How to save Apple Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> Message-ID: <16819017@NEWS.SAIC.COM> Date: Tue, 15 Jul 1997 15:59:59 GMT Joseph Palmer <jpalmer@best.com> wrote: >PostScript on the screen is a perfect match for ink publishers, >but it doesn't matter to >- Web publishers & Web users (HTML) If Postscript on the screen weren't important in this market, there'd be no Acrobat. Just think: Instead of using Acrobat, websites will now have the option of including actual EPS images for Rhapsody users. >- Music Creation (There is no Postscript in MIDI) Some composers might like using Adobe's Sonata music font for WYSIWYG scoring. >- Audio Editing Okay... >- Video Editing >- Film Editing Device-independant text could be useful when outputting to video or film, perhaps. >- CAD (PCB CAD is gerber, Mechanical CAD is DXF and others) Seems to me that DPS would be useful to CAD users. If nothing else, it should make CAD apps a lot easier to write. >- 3D Okay. Of course, most 3D work is output at some point. At which point, DPS's device-independance becomes quite useful. >- Image editing & enhancement DPS's device-independance has benefits for on-screen images. Also, a lot of people use both bitmap-editing software and line-art apps to create images. Even Fractal Design Painter, a heavily bitmap-oriented application, added a set of vector drawing tools in 4.0. (Which, incidentally, were *I think* implemented by Glenn Reid, Postscript guru and former NeXT community cool guy.) >- Industrial monitoring and control Okay, but Rhapsody would have other interesting features for that market, like DO or Unix. > There are thousands of applications for computers that > would receive no benefit from the added overhead of DPS. And many of them can run as daemons, and not have to deal with DPS. >PostScript can be used for a number of things, but its only >mission critical to print publishing programs, and anyway >it is possible to place DPS in the publishing program itself, >it does not need to be in the OS. That would be stupid. DPS has features that make the entire OpenStep experience very nice. For instance, 8-bit color on OpenStep doesn't use color tables. It dithers the screen, and looks gorgeous. As far as the apps care, it's a 24 bit screen. On a high res screen (1152x864 and up) it's sometimes hard to tell it's dithered. You overestimate the 'overhead' of DPS. You also underestimate the value of having a powerful imaging engine available to all applications. -- "I am not a vegetarian because I love animals; I am a vegetarian because I hate plants." - A. Whitney Brown
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: How to split classes hierarhy to bundles Date: 15 Jul 1997 18:28:32 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5qgfgg$4e0$1@owl.slip.net> References: <33C0F8E2.273C@rtr.transit.ru> <5q5ugu$l16$1@owl.slip.net> <33cb38a7.0@epflnews.epfl.ch> Marco Scheurer <marco@sente.ch> wrote: > On 07/11/97, Steve Dekorte wrote: > >This works fine for me when using frameworks. > >You might try using frameworks instead of bundles. > Yes, but then you loose the ability of dynamically loading code in > your application. To have the best of two worlds, we put our classes > in frameworks, and we make a bundle which I don't understand - frameworks (at least in 4.2) are themselves dynamically loaded. If you're talking about having the app use new frameworks it didn't know about at compile time, I would guess you could use the same NSBundle calls with the type set to "framework" to load them. -- Steve Dekorte - OpenStep consultant - San Francisco
From: krispos@cris.com Newsgroups: comp.sys.next.programmer Subject: LISP on Next Date: Tue, 15 Jul 1997 14:35:45 -0400 Organization: God Machine Beta .98 Message-ID: <33CBC301.247BC1D@cris.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I remember reading somewhere that the NeXT cubes had a built in LISP compiler. However apparently the color slabs do not, and I was wondering if anyone knew of any LISP compilers/interpreters for the NeXT color system (Mach 3.2). I don't currently own a NeXT, but am thinking about getting one. However if I can't use it to program than its usefullness would decrease dramatically. Thanks in advance for the help. Please e-mail any responses Krispos@cris.com
From: krispos@cris.com Newsgroups: comp.sys.next.programmer Subject: LISP on Next Date: Tue, 15 Jul 1997 14:35:29 -0400 Organization: God Machine Beta .98 Message-ID: <33CBC2F1.D1EEA930@cris.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I remember reading somewhere that the NeXT cubes had a built in LISP compiler. However apparently the color slabs do not, and I was wondering if anyone knew of any LISP compilers/interpreters for the NeXT color system (Mach 3.2). I don't currently own a NeXT, but am thinking about getting one. However if I can't use it to program than its usefullness would decrease dramatically. Thanks in advance for the help. Krispos@cris.com
From: drifterusa@macconnect.com (John Bauer) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Tue, 15 Jul 1997 13:25:25 -0500 Organization: None Message-ID: <1997071513252527572@accs-as31-dp01.dlls.grid.net> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <5qeemm$45g$1@owl.slip.net> <5qeth8$gb1$1@sps1.phys.vt.edu> Nathan Urban wrote: > The NEXTSTEP development environment needs to regain its competitive edge. You mean it's lost it? The naysayers are right? I feel like I've been duped. John Bauer
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 15 Jul 1997 13:33:02 -0700 Organization: Primenet Services for the Internet Message-ID: <AFF13158-69336@206.165.44.50> References: <E0woCDg-0007ZF-00@redemption.uniserve.com> To: "Alex Curylo" <acurylo@inmediapresents.com> nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Alex Curylo <acurylo@inmediapresents.nospam.com> said: > In article <AFED2A36-362CE@206.165.44.10>, "Lawson English" > <english@primenet.com> wrote: > > > Excuse? They most certainly DID write a converter from GX to QT Vector > > Graphics. Want Mihail Lari's email address so that you can confirm what > > they already say on their web page? > > This is a person who knows something about QT3 and GX? Sure, I'd like > somebody who knows both to explain the difference. Even better, ask him > to > post a comparative review to the qt3-dev list, I'm on that too. You mentioned LariSoft in your original article as a software house that would NOT be wasting its time in writing a GX to QT Vector converter [because the QT Vector Format was already GX]. Mihail *LARI* is an engineer at *LARI*Soft... I'll ask him on the GX mailing list if he can furnish any details on what he and Humiyan Lari have had to do to make Electrifier output work with QuickTime. I don't believe that they are finished yet, to be honest, although I am wrong more often than right. ++++++++++++++++++++++++++++++ HOW DO I SUBSCRIBE TO GX-TALK? --------------------------------- To subscribe to GX-TALK proceed as follows: Send E-mail to: <gx-talk-request@aimed.org> with anything in the subject line and the following command as the first (and only) line of the message body: SUBSCRIBE +++++++++++++++++++++++++++++++++
From: kamran@tybrin3.hsv.tybrin.com (Kamran Talai) Newsgroups: comp.sys.next.programmer Subject: Application Path Date: 15 Jul 1997 20:57:54 GMT Organization: HiWAAY Information Services Message-ID: <5qgo8i$shu$1@parlor.HiWAAY.net> Is there a way to get the actual path for the application being executed? For reasons too long to mention here, the particular application I am working on must be executed from a link. "argv" and NXArgV give the link name that was typed in but not the actual path where the application was launched from. I would greatly appreciate any help. -- Kamran Talai TYBRIN Corporation (205) 837-2027 FAX 837-3472 kamran@hsv.tybrin.com
From: Greg_Anderson@afs.com (Gregory H. Anderson) Newsgroups: comp.sys.next.programmer Subject: Re: Application Path Date: 15 Jul 1997 21:45:57 GMT Organization: Anderson Financial Systems Inc. Message-ID: <5qgr2l$io8@shelob.afs.com> References: <5qgo8i$shu$1@parlor.HiWAAY.net> Kamran Talai writes > Is there a way to get the actual path for the application being > executed? For reasons too long to mention here, the particular > application I am working on must be executed from a link. "argv" > and NXArgV give the link name that was typed in but not the actual > path where the application was launched from. The answer to your first question is "yes," but it's impossible to say anything further until you tell us which OS and which version you're programming in. I have written the author privately, but wanted to take this opportunity to remind everyone who asks for advice here in CSNP that you should ALWAYS include information about which environment you are programming in. There are many permutations at this point, and the answers are not always the same for all of them. -- Gregory H. Anderson | "We're in the land of the blind, Visionary Ophthalmologist | selling working eyeballs, and they Anderson Financial Systems | balk at the choice of color." -- Tony greg@afs.com (NeXTmail OK) | Lovell, on Mac user reactions to NeXT
From: don@globalobjects.com (Don Yacktman) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 16 Jul 1997 01:06:57 GMT Organization: Global Objects Inc. Message-ID: <5qh6rh$og7$1@news.xmission.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> jon@txenxe.com wrote: > Joseph Palmer <jpalmer@best.com> wrote: > >PostScript on the screen is a perfect match for ink publishers, > >but it doesn't matter to > > >- Web publishers & Web users (HTML) > > If Postscript on the screen weren't important in this market, > there'd be no Acrobat. Just think: Instead of using Acrobat, > websites will now have the option of including actual > EPS images for Rhapsody users. Like http://www.misckit.com/don/opusEPS.html for NEXTSTEP and OPENSTEP users. Of course, for the .eps impaired, there's also http://www.misckit.com/don/opusGIF.html but it isn't quite as nice if the images are scaled. EPS scales a lot better than GIF (no surprise, there). And just imagine how compact a nice background EPS gradation is--it could really improve bandwidth requirements...and look a heck of a lot better than any .gif ever would! > >- Music Creation (There is no Postscript in MIDI) > > Some composers might like using Adobe's Sonata music font > for WYSIWYG scoring. Like Calliope.app, on NEXTSTEP/OPENSTEP. Darn fine WYSIWYG music notation editor. And it is free, except for the cost of the Sonata font (which I have of course). I've used the app a lot, and the musician's that have seen the output have been *very* impressed with its quality. > >- CAD (PCB CAD is gerber, Mechanical CAD is DXF and others) > > Seems to me that DPS would be useful to CAD users. If nothing else, > it should make CAD apps a lot easier to write. Yup. Having written a drawing app of my own, I'll heartily agree with that. (And, yes, I've done some cad stuff, too, so I do know the difference.) > >- 3D > > Okay. Of course, most 3D work is output at some point. At which > point, DPS's device-independance becomes quite useful. This is something OPENSTEP could do better; even when Renderman was integrated into NEXTSTEP, you had to treat the 3D scenes with RIB, which worked in parallel to EPS to define the output. Rendering a RIB always produces a bitmap, not an EPS, so the two really are working in parallel. That worked, but it did seem to have a few cracks around the seams. I would expect (hope) that Rhapsody will work more nicely with QD3D. > >- Industrial monitoring and control > > Okay, but Rhapsody would have other interesting features for > that market, like DO or Unix. Actually, I know of one application where DPS was crucial to making it work. I'm under NDA, so I can't be too explicit about what I saw, but if you were selling some large machine and wanted to be able to customize the display and controls for a particular customer, the combination of OO and a good graphics engine would be very important. The application I saw used DPS heavily. In this case, DPS was drawing onto custom hardware (a special LCD display). Granted that monitoring and control is primarily not at all involved with displays of any type, but there does have to be a control panel somewhere... And don't forget the Canon color laser copiers that effectively had a NeXTStation embedded in them... > >PostScript can be used for a number of things, but its only > >mission critical to print publishing programs, and anyway > >it is possible to place DPS in the publishing program itself, > >it does not need to be in the OS. > > That would be stupid. DPS has features that make the entire > OpenStep experience very nice. For instance, 8-bit color on > OpenStep doesn't use color tables. It dithers the screen, > and looks gorgeous. As far as the apps care, it's a 24 bit > screen. On a high res screen (1152x864 and up) it's sometimes > hard to tell it's dithered. > > You overestimate the 'overhead' of DPS. You also underestimate > the value of having a powerful imaging engine available to > all applications. Now there is an understatement if I've ever seen one! -- Later, -Don Yacktman don@misckit.com <a href="http://www.misckit.com/don.html">My home page</a>
From: jpalmer@best.com (Joseph Palmer) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Tue, 15 Jul 1997 21:44:05 -0800 Organization: VideoS2 Video Storage Systems Message-ID: <jpalmer-1507972144050001@jpalmer.vip.best.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> In article <5qh6rh$og7$1@news.xmission.com>, don@globalobjects.com wrote: > jon@txenxe.com wrote: > > Joseph Palmer <jpalmer@best.com> wrote: > > >PostScript on the screen is a perfect match for ink publishers, > > >but it doesn't matter to > > > > >- Web publishers & Web users (HTML) > > > > If Postscript on the screen weren't important in this market, > > there'd be no Acrobat. Just think: Instead of using Acrobat, > > websites will now have the option of including actual > > EPS images for Rhapsody users. > > Like http://www.misckit.com/don/opusEPS.html for NEXTSTEP > and OPENSTEP users. Of course, for the .eps impaired, there's also > http://www.misckit.com/don/opusGIF.html but it isn't quite as nice if the > images are scaled. EPS scales a lot better than GIF (no surprise, there). > And just imagine how compact a nice background EPS gradation is--it could > really improve bandwidth requirements...and look a heck of a lot better than > any .gif ever would! I've wished for a simple gradient descriptor for backgrounds, but you wouldn't need DPS for that, just a simple descriptor. Of course what I'd really like is <p> tags that allow me to do indented paragraphs, like <p indent = "12"> for an indent, and <p indent = "-12"> for a hanging paragraph. Oh well. > > > >- Music Creation (There is no Postscript in MIDI) > > > > Some composers might like using Adobe's Sonata music font > > for WYSIWYG scoring. > > Like Calliope.app, on NEXTSTEP/OPENSTEP. Darn fine WYSIWYG music notation > editor. And it is free, except for the cost of the Sonata font (which I have > of course). I've used the app a lot, and the musician's that have seen the > output have been *very* impressed with its quality. <Palmer holds a peice of Sheet music to his ear.> "Nope, no sound" I'm talking about _making_ music. In multichannel. With effects, and real time input. Music notation is another ink publishing application, Music creation is not. > > >- CAD (PCB CAD is gerber, Mechanical CAD is DXF and others) > > > > Seems to me that DPS would be useful to CAD users. If nothing else, > > it should make CAD apps a lot easier to write. > > Yup. Having written a drawing app of my own, I'll heartily agree with that. > (And, yes, I've done some cad stuff, too, so I do know the difference.) Since there are no hardware DPS accelerators, I'm not sure DPS helps. I really don't need DPS for doing a board with 9 Ball grid array parts on it, since that design need never see a postscript printer. I'm not talking about Drawing apps to print to PS printers, I'm talking about apps that work in engineering units and keep a database that is related to the design (Mechanical or electrical) that is created. = > > >- 3D > > > > Okay. Of course, most 3D work is output at some point. At which > > point, DPS's device-independance becomes quite useful. > > This is something OPENSTEP could do better; even when Renderman was > integrated into NEXTSTEP, you had to treat the 3D scenes with RIB, which > worked in parallel to EPS to define the output. Rendering a RIB always > produces a bitmap, not an EPS, so the two really are working in parallel. > That worked, but it did seem to have a few cracks around the seams. I would > expect (hope) that Rhapsody will work more nicely with QD3D. > I don't do much 3D myself, But I'm looking forward to the openGL on the BeOS, and to those wicked fast hardware acceleration cards. > > >- Industrial monitoring and control > > > > Okay, but Rhapsody would have other interesting features for > > that market, like DO or Unix. > > Actually, I know of one application where DPS was crucial to making it work. > I'm under NDA, so I can't be too explicit about what I saw, but if you were > selling some large machine and wanted to be able to customize the display and > controls for a particular customer, the combination of OO and a good graphics > engine would be very important. The application I saw used DPS heavily. In > this case, DPS was drawing onto custom hardware (a special LCD display). You can always find 1 application, and anyway, are you really sure that quickdraw or the windows equ. would not have done the job? > > Granted that monitoring and control is primarily not at all involved with > displays of any type, but there does have to be a control panel somewhere... > > And don't forget the Canon color laser copiers that effectively had a > NeXTStation embedded in them... Sorry. I had forgotten that. <bonks head> Oh well. > > > >PostScript can be used for a number of things, but its only > > >mission critical to print publishing programs, and anyway > > >it is possible to place DPS in the publishing program itself, > > >it does not need to be in the OS. > > > > That would be stupid. Then why is it DPS had been around for 8 or 10 years and has not been picked up by IBM, Apple, Microsoft, etc.... for their OS efforts? > > DPS has features that make the entire > > OpenStep experience very nice. For instance, 8-bit color on > > OpenStep doesn't use color tables. It dithers the screen, > > and looks gorgeous. As far as the apps care, it's a 24 bit > > screen. On a high res screen (1152x864 and up) it's sometimes > > hard to tell it's dithered. Yeah, the BeOS has that, it does look cool. But, they didn't need DPS for that. > > You overestimate the 'overhead' of DPS. You also underestimate > > the value of having a powerful imaging engine available to > > all applications. No more than I underestimate the value of driving arround in a 10 passenger vehicle, just in case I might need all those seats. > > Now there is an understatement if I've ever seen one! I think the "Bill Gates is rich" thread beats anything. ______________________________________________________________ Joseph Palmer | Never join a titanic battle unless CEO | you get to be the iceberg. -- J. VideoS2 | Personal web site: http://www.videos2.com | http://www.best.com/~jpalmer
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 15 Jul 1997 22:27:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFF1AE90-29E54@206.165.44.12> References: <5qh6rh$og7$1@news.xmission.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Don Yacktman <don@globalobjects.com> said: > on@txenxe.com wrote: > > Joseph Palmer <jpalmer@best.com> wrote: > > >PostScript on the screen is a perfect match for ink publishers, > > >but it doesn't matter to > > > > >- Web publishers & Web users (HTML) > > > > If Postscript on the screen weren't important in this market, > > there'd be no Acrobat. Just think: Instead of using Acrobat, > > websites will now have the option of including actual > > EPS images for Rhapsody users. > Er, GX's Portable Digital Documents are smaller than pdf files, in general. Also, for languages that use LARGE amounts of memory, there is no comparison. GX, as part of the printing process, is able to determine which characters are used in each page and only make use of those specific characters when creating a page description. For languages, like Japanese, which have fonts with as much as 1.5 MB per, the ability to only download those characters which are needed would be a BIG time/bandwidth savings for the internet. Not my fault that Apple executives were/are stupid. > Like http://www.misckit.com/don/opusEPS.html for NEXTSTEP > and OPENSTEP users. Of course, for the .eps impaired, there's also > http://www.misckit.com/don/opusGIF.html but it isn't quite as nice if the > images are scaled. EPS scales a lot better than GIF (no surprise, there). > And just imagine how compact a nice background EPS gradation is--it > could > really improve bandwidth requirements...and look a heck of a lot better > than > any .gif ever would! GX unfortunately, does not include gradiated bitmaps. That is something that could easily be added as a new GX shape type, of course. > > > >- Music Creation (There is no Postscript in MIDI) > > > > Some composers might like using Adobe's Sonata music font > > for WYSIWYG scoring. > > Like Calliope.app, on NEXTSTEP/OPENSTEP. Darn fine WYSIWYG music > notation > editor. And it is free, except for the cost of the Sonata font (which I have > of course). I've used the app a lot, and the musician's that have seen the > output have been *very* impressed with its quality. GX works wonders here, I'm told by those that create GX-based WYSIWYG music notators. > > > >- CAD (PCB CAD is gerber, Mechanical CAD is DXF and others) > > > > Seems to me that DPS would be useful to CAD users. If nothing else, > > it should make CAD apps a lot easier to write. > > Yup. Having written a drawing app of my own, I'll heartily agree with that. > (And, yes, I've done some cad stuff, too, so I do know the difference.) > Never done any GX drawing, have you? GX is *ideal* for most CAD/CAM work. > > >- 3D > > > > Okay. Of course, most 3D work is output at some point. At which > > point, DPS's device-independance becomes quite useful. > > This is something OPENSTEP could do better; even when Renderman was > integrated into NEXTSTEP, you had to treat the 3D scenes with RIB, which > worked in parallel to EPS to define the output. Rendering a RIB always > produces a bitmap, not an EPS, so the two really are working in parallel. > That worked, but it did seem to have a few cracks around the seams. I > would > expect (hope) that Rhapsody will work more nicely with QD3D. > GX and QD3D were on a collision course, API-wise. I don't see that as happening with DPS. Interceptor technology just doesn't allow it -at least not easily. > > >- Industrial monitoring and control > > > > Okay, but Rhapsody would have other interesting features for > > that market, like DO or Unix. > > Actually, I know of one application where DPS was crucial to making it > work. > I'm under NDA, so I can't be too explicit about what I saw, but if you were > selling some large machine and wanted to be able to customize the display > and > controls for a particular customer, the combination of OO and a good > graphics > engine would be very important. The application I saw used DPS heavily. > In > this case, DPS was drawing onto custom hardware (a special LCD display). > You are aware that the GX Print Driver manual has an explicit API for Vector Plotters, no? > Granted that monitoring and control is primarily not at all involved with > displays of any type, but there does have to be a control panel somewhere... > > And don't forget the Canon color laser copiers that effectively had a > NeXTStation embedded in them... > Not my fault if Apple management has been (and likely always will be) a pack of idiots... > > >PostScript can be used for a number of things, but its only > > >mission critical to print publishing programs, and anyway > > >it is possible to place DPS in the publishing program itself, > > >it does not need to be in the OS. > > > > That would be stupid. DPS has features that make the entire > > OpenStep experience very nice. For instance, 8-bit color on > > OpenStep doesn't use color tables. It dithers the screen, > > and looks gorgeous. As far as the apps care, it's a 24 bit > > screen. On a high res screen (1152x864 and up) it's sometimes > > hard to tell it's dithered. > > You're aware, of course, that GX allows dithering levels to be set on a per viewport basis, and that one can turn off dithering/halftoning for specific objects and groups of objects... > > You overestimate the 'overhead' of DPS. You also underestimate > > the value of having a powerful imaging engine available to > > all applications. > > Now there is an understatement if I've ever seen one! > GX was/is available on a system-wide basis. Not my fault if Apple management was/is a pack of idiots and theives (e.g.: Amelio's going-away present). Of course, the fact that GX fits so nicely with scripted languages and provides features that even the most advanced version of the AppKit apparently doesn't have (e.g. 3D perspective, multiple rendering, seperate channel transparency/copying, etc), doesn't mean squat. Non-Apple engineers have taken over the company and in nearly every case, only non-Apple technology has survived the invasion. The Vandals and Visigoths apparently don't recognize what they are destroying. ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: Lars Immisch <lars@ibp.de> Newsgroups: comp.sys.next.programmer Subject: timed wait on condition Date: Tue, 15 Jul 1997 12:18:53 +0200 Organization: Immisch, Becker & Partner Message-ID: <33CB4E8D.224A@ibp.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I am considering how to best implement a timed wait on a condition - this is Mach terminology - #import <cthreads.h> - so I am after a condition_wait(condition_t c, mutex_t m, time_t timeout) instead of the plain Mach condition_wait(condition_t c, mutex_t m) If I use setitimer, can I have my own signal handler for any thread, or will signal handling by synchronous only to thread 1? Alternatively, I consider using msg_receive/msg_send to emulate a condition. Any advice is very welcome, Lars -- mailto:immisch@pobox.com http://pobox.com/~immisch Yesterdays yellow yoyo can make you yawn today
From: LTaylor7@ix.netcom.com (Daniel L. Taylor) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Tue, 15 Jul 1997 23:57:12 -0700 Organization: Netcom Message-ID: <LTaylor7-1507972357180001@vic-ca2-13.ix.netcom.com> References: <5qh6rh$og7$1@news.xmission.com> <AFF1AE90-29E54@206.165.44.12> In article <AFF1AE90-29E54@206.165.44.12>, "Lawson English" <english@primenet.com> wrote: > Of course, the fact that GX fits so nicely with scripted languages and > provides features that even the most advanced version of the AppKit > apparently doesn't have (e.g. 3D perspective, multiple rendering, seperate > channel transparency/copying, etc), doesn't mean squat. No, it doesn't. Not any more. Once again, DTP wants DPS. Perhaps they would have been better off using GX. But it didn't happen. Blame it on the old Apple. And if we want to prevent the same thing from happening to Apple, we had better pick our fights more wisely. > Non-Apple engineers have taken over the company and in nearly every case, > only non-Apple technology has survived the invasion. What is this, bitterness? Apple got a great deal in getting the NeXT engineers. As to only non-Apple tech surviving, that doesn't explain the Mac UI, AppleScript, QTML, and all of Mac OS which was dead before Amelio and NeXT arrived. > The Vandals and Visigoths apparently don't recognize what they are > destroying. I bet they do. I bet they also realize that they can't have both the whole suite of old Apple technologies and jobs next year.... -- Daniel L. Taylor LTaylor7@ix.netcom.com ================================================================= The Heritage Foundation /// Town Hall http://www.heritage.org/heritage/ /// http://www.townhall.com/ =================================================================
Sender: 987655555453@compuserve.com Control: cancel <5qi0a9$emg@news1-alterdial.uu.net> Message-ID: <cancel.5qi0a9$emg@news1-alterdial.uu.net> Subject: cmsg cancel <5qi0a9$emg@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 16 Jul 1997 08:22:08 GMT EMP article removed by jem@xpat.com. Original Headers: From: 987655555453@compuserve.com Subject: Pa$$Word TO 1200 sex sites only $25! Newsgroups: comp.sys.next.programmer Path: ...!hammer.uoregon.edu!csulb.edu!gatech!nntprelay.mathworks.com!news.mathworks.com!uunet!not-for-mail Lines: 99
Sender: 18734757@compuserve.com Control: cancel <5qi0vg$d9n@news1-alterdial.uu.net> Message-ID: <cancel.5qi0vg$d9n@news1-alterdial.uu.net> Subject: cmsg cancel <5qi0vg$d9n@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 16 Jul 1997 08:34:00 GMT EMP article removed by jem@xpat.com. Original Headers: From: 18734757@compuserve.com Subject: Get FREE PASSWORD SEX SITE..password is... Newsgroups: comp.sys.next.programmer Path: ...!hammer.uoregon.edu!csulb.edu!gatech!nntprelay.mathworks.com!news.mathworks.com!cam-news-hub1.bbnplanet.com!news.bbnplanet.com!news-feed1.tiac.net!uunet!not-for-mail Lines: 99
From: 987655555453@compuserve.com Newsgroups: comp.sys.next.programmer Subject: Pa$$Word TO 1200 sex sites only $25! Date: 16 Jul 1997 08:21:29 GMT Organization: .PASSWORD INC... Message-ID: <5qi0a9$emg@news1-alterdial.uu.net> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary=_-=314 This is a multi-part message in MIME format. --_-=314 Content-type: text/html; name="Sex~1.htm" Content-transfer-encoding: base64 PCFET0NUWVBFIEhUTUwgUFVCTElDICItLy9JRVRGLy9EVEQgSFRNTC8vRU4iPg0KDQo8aHRt bD4NCg0KPGhlYWQ+DQo8dGl0bGU+U0VDUkVUIFBBU1NXT1JEIEZPUiBBRFVMVCBTSUdIVFM8 L3RpdGxlPg0KPG1ldGEgbmFtZT0iR0VORVJBVE9SIiBjb250ZW50PSJNaWNyb3NvZnQgRnJv bnRQYWdlIDEuMSI+DQo8L2hlYWQ+DQoNCjxib2R5IGJnY29sb3I9IiNDMEMwQzAiIHRleHQ9 IiM4MEZGRkYiIGxpbms9IiMwMDAwRkYiIHZsaW5rPSIjMDAwMEZGIiBhbGluaz0iI0ZGMDAw MCI+DQo8ZGl2IGFsaWduPWNlbnRlcj48Y2VudGVyPg0KPHRhYmxlIGJvcmRlcj01IGNlbGxw YWRkaW5nPTI+DQo8dHI+PHRkIGFsaWduPWNlbnRlciB3aWR0aD0xMDAlPjxmb250IGNvbG9y PSIjRkZGRjAwIj48Zm9udCBzaXplPTc+PGI+RlJFRSEgRlJFRSEgRlJFRSE8YnI+DQo8L2I+ PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT01PjxiPlRI QVQnUyBSSUdIVC4uLiBDT01QTEVURUxZIDwvYj48L2ZvbnQ+PGZvbnQgc2l6ZT02PjxiPkZS RUUhITwvYj48L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjMDAwMEZGIj48Zm9udCBzaXpl PTU+PGI+PGk+PGJyPg0KRU5URVIgVEhST1VHSCBBRFVMVFNJR0hUUyAmIzE2OTsgPC9pPjwv Yj48L2ZvbnQ+PGZvbnQgc2l6ZT02Pjxicj4NCjwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9 IiNGRkZGMDAiPjxmb250IHNpemU9Nz48Yj5JVCdTIEZSRUUhISEhITwvYj48L2ZvbnQ+PC9m b250PjwvdGQ+PC90cj4NCjwvdGFibGU+DQo8L2NlbnRlcj48L2Rpdj4NCjxwIGFsaWduPWNl bnRlcj48Zm9udCBGQUNFPSJDb21pYyBTYW5zIE1TIj48Zm9udCBjb2xvcj0iIzIyMTY1QSI+ PGZvbnQgc2l6ZT03PiA8L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjRkYwMDAwIj48Zm9u dCBzaXplPTc+T05FIFBBU1NXT1JEIDxmb250IEZBQ0U9IkNvbWljIFNhbnMgTVMiPjxicj4N CjwvZm9udD48L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjMjIxNjVBIj48Zm9udCBzaXpl PTU+IDwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9IiNGRkZGMDAiPjxmb250IHNpemU9NT48 c3Ryb25nPkdFVFMgWU9VIFVOTElNSVRFRCBGUkVFIEFDQ0VTUyBUTyBPVkVSIA0KMTgwMCsg T0YgVEhFIEhPVFRFU1QgQURVTFQgU0lURVMgPC9zdHJvbmc+PC9mb250PjwvZm9udD48L3A+ DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29sb3I9IiMwMDAwMDAiPjxmb250IHNpemU9NT48 c3Ryb25nPklGIFlPVSBBUkUgVElSRUQgT0YgUEFZSU5HIFRPIEdFVCBJTlRPIFNFWCANClNJ VEVTLi4uPC9zdHJvbmc+PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGRkZGRiI+PGZv bnQgc2l6ZT01PjxzdHJvbmc+IDwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9 IiMwMEZGRkYiPjxmb250IHNpemU9NT5PbmUgUGFzc3dvcmQgaXMgYWxsIHlvdSBuZWVkISEh PGJyPg0KPC9mb250PjwvZm9udD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgRkFDRT0i Q29taWMgU2FucyBNUyI+PGZvbnQgY29sb3I9IiMyMjE2NUEiPjxmb250IHNpemU9Nj4gPC9m b250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT02PjxzdHJvbmc+ TE9PSyBXSEFUIFlPVSBHRVQhPC9zdHJvbmc+PC9mb250PjwvZm9udD48L3A+DQo8cCBhbGln bj1jZW50ZXI+PGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PGZvbnQgY29sb3I9IiMyMjE2 NUEiPjxmb250IHNpemU9NT4gPC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGRkYwMCI+ PGZvbnQgc2l6ZT01PjxzdHJvbmc+VU5MSU1JVEVEIEZSRUUgQUNDRVNTIFRPIE9WRVIgMTgw MCsgT0YgDQpUSEUgSE9UVEVTVCBBRFVMVCBTSVRFUzwvc3Ryb25nPiA8Zm9udCBGQUNFPSJD b21pYyBTYW5zIE1TIj48L2ZvbnQ+PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iIzIyMTY1 QSI+PGZvbnQgc2l6ZT01PiA8L2ZvbnQ+PC9mb250PjwvcD4NCjxoMSBhbGlnbj1jZW50ZXI+ PGZvbnQgY29sb3I9IiNGRjAwMDAiPjxmb250IHNpemU9NT48c3Ryb25nPkp1c3QgQSBGZXcg T2YgVGhlIE92ZXIgMTgwMCsgWFhYIFJhdGVkIFNpZ2h0cyBUaGF0IA0KWW91IENhbiBFbnRl ciBDb21wbGV0ZWx5IEZyZWUhPGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PC9mb250Pjwv c3Ryb25nPjwvZm9udD48L2ZvbnQ+PC9oMT4NCjxoMSBhbGlnbj1jZW50ZXI+PGZvbnQgY29s b3I9IiM4MEZGMDAiPjxmb250IHNpemU9Nj48c3Ryb25nPlRFRU4gU0lURVM8L3N0cm9uZz48 L2ZvbnQ+PC9mb250PjwvaDE+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29sb3I9IiMwMDAw MDAiPjxmb250IHNpemU9ND4xMDAwIEhPVCBBTkQgTmFLRUQgLSBDZUxlQlJpVGlFUyBBTkQg VEVFTlMgLSAxMDAxIE51ZGUgRXJvdGljIA0KVGVlbnMgLSAyMCBBbWFUZWVucyAtIDYyMyBO dWRlIENlbGVicml0aWVzIEFuZCBUZWVucyAtIEFBQSBTTFVUWSANClRFRU5TIC0gQW1hdGV1 ciBUZWVucyAtIEFtYXRldXIgVGVlbnMgRnJvbSBBcm91bmQgVGhlIFdvcmxkIC0gQW1hdGV1 ciANClRlZW5zISAtIENvY2sgSHVuZ3J5IFRlZW5zIC0gQ3VtIERyaW5raW5nIFRlZW5zIC1G aXN0IEZ1Y2tpbmcgVGVlbnMgLSBIb3QgDQpOdWRlIFRlZW5zIC0gTm90aGluZyBCdXQgVGVl bnMgLSBUZWVucyBJbiBUcm91YmxlIC0gVHdhdCBTdWNraW5nIFRlZW5zIDwvZm9udD48L2Zv bnQ+PC9wPg0KPHAgYWxpZ249Y2VudGVyPjxmb250IGNvbG9yPSIjRkYwMDAwIj48Zm9udCBz aXplPTY+PHN0cm9uZz5BU0lBTiBTSUdIVFM8L3N0cm9uZz48L2ZvbnQ+PC9mb250PjwvcD4N CjxwIGFsaWduPWNlbnRlcj48Zm9udCBjb2xvcj0iIzAwMDAwMCI+PGZvbnQgc2l6ZT00PkFs bCBBc2lhbiBUZWVuIC0gQXNpYW4gQmFiZXMgLSBBc2lhbiBCYWJlcyBzaWFuIEJhYmVzISAt IEFzaWFuIEJlbGxlcyAtIEFzaWFuIA0KQm9uZGFnZSAtIEFzaWFuIEN1dGUgR2lybHMgLSBB c2lhbiBEZWxpdGVzIC0gQXNpYW4gRXJvcyAtIEFzaWFuIEZsb3dlcnMgLSANCkFzaWFuIEdv bGQgLSBZb3VuZyBBc2lhbiBXb21lbiAtIEFzaWFuIEhhcmRDb3JlIEFTSUFOIEhBUkRDT1JF IC0gDQpUSEUgTEVHRU5EIENPTlRJTlVFUyAtIEFTSUFOIEhvbG9zdWl0ZSAtIEFzaWFuIExv dmUgRG9sbHMgPC9mb250PjwvZm9udD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29s b3I9IiNGRjAwRkYiPjxmb250IHNpemU9Nj48c3Ryb25nPkNFTEVCUklUWSBTSUdIVFM8L3N0 cm9uZz48L2ZvbnQ+PC9mb250PjwvcD4NCjxwIGFsaWduPWNlbnRlcj48Zm9udCBjb2xvcj0i IzAwMDAwMCI+PGZvbnQgc2l6ZT00PiMxIE51ZGUgQ2VsZWJyaXR5IFBhZ2UgISAtIENlbGVi d29ybGQgQWNlJ3MgLSBYWFggQ2VsZWJyaXR5IFBpY3MgLSBDZWxlYnJpdHkgDQpDZW50cmFs IC0gQ2VsZWJyaXR5IENoaWNrIFBhdHRpZXMgLSBDZWxlYnJpdHkgQ3VtYm94IC0gQ2VsZWJy aXR5IFBpY3R1cmVzIA0KQ2VsZWJyaXR5IFBpY3R1cmVzIEdvbGQgU2l0ZSAtIENlbGVicml0 eSBTa2luIC0gQ2VsZWJyaXR5IFNraW4tQUJDIC0gRGEgUm9ja3MgDQpudWRlIGNlbGVicml0 eSBBcmNoaXZlIC0gTWFya3MgQ2VsZWJyaXR5IFBpY3R1cmVzIDIgLSBOdWRlIENlbGVicml0 eSBBcmNoaXZlIC0gDQpTdGFyTG9jYXRvciAtIENlbGVicml0eSBTdXBlclMgLSBUaGUgRGFp bHkgTnVkZSBDZWxlYnJpdHkgUGFnZSAtIFRoZSBNdXN0IA0KU2VlIE5VREUgQ2VsZWJyaXR5 IFNpdGUgLSBUaGUgVWx0aW1hdGUgQ2VsZWJyaXR5IENvbGxlY3Rpb24gPC9mb250PjwvZm9u dD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PGZv bnQgY29sb3I9IiNGRjRGQTciPjxmb250IHNpemU9NT4gPC9mb250PjwvZm9udD48Zm9udCBj b2xvcj0iIzAwODBGRiI+PGZvbnQgc2l6ZT01PjxzdHJvbmc+VU5MSU1JVEVEIEZSRUUgQUND RVNTIFRPIE9VUiBYWFggUkFURUQgDQpDSEFUIFJPT008L3N0cm9uZz48L2ZvbnQ+PC9mb250 PjwvcD4NCjxwIGFsaWduPWNlbnRlcj48Zm9udCBGQUNFPSJDb21pYyBTYW5zIE1TIj48Zm9u dCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT01PjxzdHJvbmc+R2V0IFlvdXIgQURVTFRT SUdIVFMgQUdFIFZFUklGSUNBVElPTiANCk1FTUJFUlNISVAgRk9SIE9OTFkgJDEyLjk1LiBJ dHMgR29vZCBGb3IgQW4gDQpFbnRpcmUgWWVhciBBbmQgR2V0cyBZb3UgVU5MSU1JVEVEIEZS RUUgQWNjZXNzIA0KSW50byAxODAwKyBPZiBUaGUgQmVzdCBBZHVsdCBTaXRlcyBPbiBUaGUg d2ViIEFsbCBZZWFyIA0KQWxsIEZyZWUuIE5ldmVyIFBheSBBbm90aGVyIERpbWUgVG8gR2V0 IEludG8gQSBTZXggU2l0ZSANCkFnYWluLjwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PGZvbnQg Y29sb3I9IiNGRkZGMDAiPjxmb250IHNpemU9NT48YnI+DQo8L2ZvbnQ+PC9mb250PjwvZm9u dD48Zm9udCBjb2xvcj0iI0ZGMDA4MCI+PGZvbnQgc2l6ZT03PjxzdHJvbmc+Qk9OVVMhISE8 YnI+DQo8L3N0cm9uZz48L2ZvbnQ+PGZvbnQgc2l6ZT01PjxzdHJvbmc+RlJFRSBYWFggQ2hh dCBSb29tcyEhITwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PC9wPg0KPHAgYWxpZ249Y2VudGVy PjxhIGhyZWY9Imh0dHA6Ly8yMDUuMjE2LjI0OS4xOC9jZ2ktc2hsL2RibWwuZXhlP3RlbXBs YXRlPS9hcy9zaWdudXAuZGJtJmFtcDtzaXRlPXN0MzUzMjUiPjxmb250IGNvbG9yPSIjODAw MDgwIj48Zm9udCBzaXplPTY+PHN0cm9uZz5JIFdBTlQgVEhFIFBBU1NXT1JEIE5PVzwvc3Ry b25nPjwvZm9udD48L2ZvbnQ+PC9hPjwvcD4NCjwvYm9keT4NCg0KPC9odG1sPg0K --_-=314
From: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 16 Jul 1997 08:46:46 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5qifrm$ff5$1@sps1.phys.vt.edu> References: <AFEFAFFC-20EF8@206.165.44.6> <5qeemm$45g$1@owl.slip.net> <5qeth8$gb1$1@sps1.phys.vt.edu> <1997071513252527572@accs-as31-dp01.dlls.grid.net> In article <1997071513252527572@accs-as31-dp01.dlls.grid.net>, drifterusa@macconnect.com (John Bauer) wrote: > Nathan Urban wrote: > > The NEXTSTEP development environment needs to regain its competitive edge. > You mean it's lost it? The naysayers are right? I feel like I've been > duped. Well, it's still better than pretty much anything else out there, IMHO, but it's no longer _orders of magnitude_ better than anything else out there, IMHO.
From: maury@softarc.com (Maury Markowitz) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Openstep examples OO design (was Re: Some initial OpenStep notes) Date: Wed, 16 Jul 1997 12:18:03 -0400 Organization: SoftArc Inc. Message-ID: <maury-1607971218030001@199.166.204.230> References: <5pmd4m$i21$2@news.xmission.com> <5q26ue$ndb$1@lazar.select-tech.si> <5q49m4$mt@saturn.genoa.com> <5q6p58$uh0$1@sps1.phys.vt.edu> In article <5q6p58$uh0$1@sps1.phys.vt.edu>, nurban@vt.edu wrote: > How would you implement an undo system? Personally I'd have the NeXT people at Apple design a really good one and put it in the AppKit. Maury
From: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Openstep examples OO design (was Re: Some initial OpenStep notes) Date: 16 Jul 1997 13:08:08 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5qiv5o$h0l$1@sps1.phys.vt.edu> References: <5pmd4m$i21$2@news.xmission.com> <5q49m4$mt@saturn.genoa.com> <5q6p58$uh0$1@sps1.phys.vt.edu> <maury-1607971218030001@199.166.204.230> In article <maury-1607971218030001@199.166.204.230>, maury@softarc.com (Maury Markowitz) wrote: > In article <5q6p58$uh0$1@sps1.phys.vt.edu>, nurban@vt.edu wrote: > > How would you implement an undo system? > Personally I'd have the NeXT people at Apple design a really good one > and put it in the AppKit. Me too, but supposing they don't..
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: timed wait on condition Date: Wed, 16 Jul 1997 12:11:49 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <0nnD=5i00UzxQ1oiF6@andrew.cmu.edu> References: <33CB4E8D.224A@ibp.de> In-Reply-To: <33CB4E8D.224A@ibp.de> Excerpts from netnews.comp.sys.next.programmer: 15-Jul-97 timed wait on condition by Lars Immisch@ibp.de > I am considering how to best implement a timed wait on a condition - > this is Mach terminology - #import <cthreads.h> - so I am after a > condition_wait(condition_t c, mutex_t m, time_t timeout) instead of the > plain Mach condition_wait(condition_t c, mutex_t m) > > If I use setitimer, can I have my own signal handler for any thread, or > will signal handling by synchronous only to thread 1? Unix signals (like the alarm clock) get delivered to the currently running thread; there is no per-thread signal delivery. This is because the BSD signal mechanism is really based off of various Mach exceptions which get translated to the appropriate signal. If you want to do per-thread exception handling, look at catch_exception_raise() and the "Exception Handling" example, but it's more effort than it's worth for this. > Alternatively, I consider using msg_receive/msg_send to emulate a > condition. Probably easier. Here is a function which implements a per-thread sleep in milliseconds which you might find useful: void thread_msleep(unsigned msec) { msg_header_t null_msg; port_t port; if (port_allocate(task_self(), &port) != KERN_SUCCESS) return; null_msg.msg_local_port = port; null_msg.msg_size = sizeof(null_msg); msg_receive(&null_msg,RCV_TIMEOUT,msec); // wait until timeout port_deallocate(task_self(),port); // clean up } In order to do a timed condition_wait(), I'd probably set up a variable with adequate scope to indicate whether a timeout had occurred, fork off a new thread and use thread_switch to go to it, do a thread_msleep() for however long in the new thread, set the timeout variable, and signal the condition. The original thread can then check the variable and see whether it got timed out. Of course, depending on what you're doing such as whether one or multiple threads are waiting on this condition, you might have to change this. Also, your timeout periods need to be long enough to ensure that the original thread will get CPU time when the timeout thread is waiting on msg_receive() in order to do the condition_wait(), which shouldn't be a problem if your timeout is in the human time scale and not in the tens of milliseconds. If that's not the case, then you're going to have to go ahead and implement a real message between threads via msg_send, which is somewhat more work. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: kcd@babylon5.jumpgate.com (Kenneth C. Dyke) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 16 Jul 1997 17:31:28 GMT Message-ID: <5qj0hg$8um$1@nntp2.ba.best.com> References: <5qh6rh$og7$1@news.xmission.com> <AFF1AE90-29E54@206.165.44.12> In-Reply-To: <AFF1AE90-29E54@206.165.44.12> On 07/15/97, "Lawson English" wrote: >Of course, the fact that GX fits so nicely with scripted languages and >provides features that even the most advanced version of the AppKit >apparently doesn't have (e.g. 3D perspective, multiple rendering, seperate >channel transparency/copying, etc), doesn't mean squat. QuickDraw does not do 3D of any kind. Buy a clue, or get a math book. How nicely does GX fit scripting without your XCFN thing? At least with AppKit someone could write a scripting language interface to it that would allow you to access anything accessible via Objective C without implementing custom code for every function/method call. Oh wait, oops, guess someone already did it. Darn. -Ken -- Kenneth Dyke, kcd@jumpgate.com (personal), kdyke@ea.com (work) Nuclear Strike and OPENSTEP Tools Engineer, Electronic Arts C++: The power, elegance and simplicity of a hand grenade.
Sender: Jonathan Hendry <jhendry@> From: jon@txenxe.com Subject: Re: How to save Apple Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <jpalmer-1507972144050001@jpalmer.vip.best.com> Message-ID: <17035036@NEWS.SAIC.COM> Date: Wed, 16 Jul 1997 18:57:39 GMT Joseph Palmer <jpalmer@best.com> wrote: >Music notation is another ink publishing application, Music creation is >not. Not if the user is a composer who happens to like working with musical notation while creating music. >Since there are no hardware DPS accelerators, I'm not sure DPS helps. >I really don't need DPS for doing a board with 9 Ball grid array parts >on it, since that design need never see a postscript printer. >I'm not talking about Drawing apps to print to PS printers, I'm talking >about apps that work in engineering units and keep a database that is >related to the design (Mechanical or electrical) that is created. So these apps will never display anything onscreen? That seems to be a common feature, is it not? >> >> > >PostScript can be used for a number of things, but its only >> > >mission critical to print publishing programs, and anyway >> > >it is possible to place DPS in the publishing program itself, >> > >it does not need to be in the OS. >> > >> > That would be stupid. >Then why is it DPS had been around for 8 or 10 years and has not been >picked up by IBM, Apple, Microsoft, etc.... for their OS efforts? Timing. Their OS's pretty much predated DPS. While MacOS helped spur the adoption of Postscript in printers, they weren't about to tear out Quickdraw and insert DPS. Nor could they have done so feasibly. Their OS's were also aimed at weaker machines that wouldn't have done very well with DPS. Their OS's were less capable, making them less suitable for a client-server DPS display model. Of course, there's always a flipside. Why should MacOS include Quickdraw? Why not have it be part of the applications that need to draw? Why not make app developers include Windows GDI in their applications? >> > DPS has features that make the entire >> > OpenStep experience very nice. For instance, 8-bit color on >> > OpenStep doesn't use color tables. It dithers the screen, >> > and looks gorgeous. As far as the apps care, it's a 24 bit >> > screen. On a high res screen (1152x864 and up) it's sometimes >> > hard to tell it's dithered. >Yeah, the BeOS has that, it does look cool. But, they didn't need DPS >for that. But it helps. Do BeOS app developers have to write any code to handle different bit-depths? -- "I am not a vegetarian because I love animals; I am a vegetarian because I hate plants." - A. Whitney Brown
Sender: 987576555554453@compuserve.com Control: cancel <5qj4u8$as4@news1-alterdial.uu.net> Message-ID: <cancel.5qj4u8$as4@news1-alterdial.uu.net> Subject: cmsg cancel <5qj4u8$as4@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 16 Jul 1997 18:46:35 GMT EMP article removed by jem@xpat.com. Original Headers: From: 987576555554453@compuserve.com Subject: Pa$$Word TO 1200 sex sites only $25! Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!europa.clark.net!ais.net!uunet!not-for-mail Lines: 99
From: maury@softarc.com (Maury Markowitz) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Openstep examples OO design (was Re: Some initial OpenStep notes) Date: Wed, 16 Jul 1997 15:17:41 -0400 Organization: SoftArc Inc. Message-ID: <maury-1607971517410001@199.166.204.230> References: <5pmd4m$i21$2@news.xmission.com> <5q49m4$mt@saturn.genoa.com> <5q6p58$uh0$1@sps1.phys.vt.edu> <maury-1607971218030001@199.166.204.230> <5qiv5o$h0l$1@sps1.phys.vt.edu> In article <5qiv5o$h0l$1@sps1.phys.vt.edu>, nurban@vt.edu wrote: > Me too, but supposing they don't.. Hmmm. Ok, I'd want all the base classes to be able to record "action deltas" as a standard feature, then just about anyone could do it in a few lines of code. Maury
From: John Burke <jcburke@bellatlantic.net> Newsgroups: fr.comp.lang.java,de.comp.lang.java,comp.sys.sun.misc,comp.sys.sun.apps,comp.sys.sun.admin,comp.sys.next.programmer Subject: BOOK: Java Certification Exam Guide Date: Wed, 16 Jul 1997 17:22:57 -0500 Organization: Bell Atlantic Internet Solutions Message-ID: <Forum.869088704.18892.jcburke@john> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=ISO-8859-1 There's a new book from Computing McGraw-Hill called "Java Certification Exam Guide" which preps you for Sun-certified Java. You can find information, TOC, and an excerpt on Input/Output at: http://www.smartbooks.com/bw707javacert.htm Regards, John Burke
From: mmalcolm crawford <malcolm@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 16 Jul 1997 22:35:16 GMT Organization: P&L Systems Message-ID: <5qjib4$7mr$1@ironhorse.plsys.co.uk> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <jpalmer-1507972144050001@jpalmer.vip.best.com> <17035036@NEWS.SAIC.COM> In-Reply-To: <17035036@NEWS.SAIC.COM> Joseph Palmer <jpalmer@best.com> wrote: >Yeah, the BeOS has that, it does look cool. But, they didn't need DPS >for that. > Does BeOS print yet...? Best wishes, mmalc. --
From: michael@rumah.pc.my (Michael Olan) Newsgroups: comp.sys.next.programmer Subject: distributed programming Date: 16 Jul 1997 23:13:23 GMT Organization: Personal Message-ID: <5qjkij$e6v@rumah.pc.my> Hi, I have 2 slabs connected via ethernet, and I'd like to experiment with writing code to use both cpu's. Could someone point me in the direction of tutorials, doc's, etc. on how to do this? C++ preferred, but any Next supported language would be fine. Anything simple for starters, like say matrix processing would be great. TIA, Mike
From: jpalmer@best.com (Joseph Palmer) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Wed, 16 Jul 1997 16:39:20 -0800 Organization: VideoS2 Video Storage Systems Message-ID: <jpalmer-1607971639200001@jpalmer.vip.best.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <jpalmer-1507972144050001@jpalmer.vip.best.com> <17035036@NEWS.SAIC.COM> <5qjib4$7mr$1@ironhorse.plsys.co.uk> In article <5qjib4$7mr$1@ironhorse.plsys.co.uk>, mmalcolm crawford <malcolm@plsys.co.uk> wrote: > Joseph Palmer <jpalmer@best.com> wrote: > > >Yeah, the BeOS has that, it does look cool. But, they didn't need DPS > >for that. > > > Does BeOS print yet...? YES And you can order it TODAY. <http://www.be.com> ______________________________________________________________ Joseph Palmer | Never join a titanic battle unless CEO | you get to be the iceberg. -- J. VideoS2 | Personal web site: http://www.videos2.com | http://www.best.com/~jpalmer
Sender: Dan Aaronson <defaultuser@domain.com> Control: cancel <33CD58E5.36DC0F72@domain.com> Message-ID: <cancel.33CD58E5.36DC0F72@domain.com> Subject: cmsg cancel <33CD58E5.36DC0F72@domain.com> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 16 Jul 1997 23:30:49 GMT EMP article removed by jem@xpat.com. Original Headers: From: Dan Aaronson <defaultuser@domain.com> Subject: major tech support site Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!europa.clark.net!nntprelay.mathworks.com!news.mathworks.com!cam-news-hub1.bbnplanet.com!su-news-feed4.bbnplanet.com!news.bbnplanet.com!news.psrinc.com!not-for-mail Lines: 4
From: Jonathan Hendry <jon@subsequent.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 17 Jul 1997 02:20:14 GMT Organization: Netcom Sender: Jonathan Hendry <jon@aahz.ix.netcom.com> Message-ID: <5qjvgu$hlq@sjx-ixn7.ix.netcom.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <jpalmer-1507972144050001@jpalmer.vip.best.com> <17035036@NEWS.SAIC.COM> <5qjib4$7mr$1@ironhorse.plsys.co.uk> <jpalmer-1607971639200001@jpalmer.vip.best.com> Joseph Palmer <jpalmer@best.com> wrote: > In article <5qjib4$7mr$1@ironhorse.plsys.co.uk>, mmalcolm crawford > <malcolm@plsys.co.uk> wrote: > > Joseph Palmer <jpalmer@best.com> wrote: > > > > >Yeah, the BeOS has that, it does look cool. But, they didn't need DPS > > >for that. > > > > > Does BeOS print yet...? > YES > And you can order it > TODAY. > But why would I? One just-barely-viable platform is enough for me, thanks. > http://www.videos2.com | http://www.best.com/~jpalmer
From: mike deavila <comdeavi@ihc.com> Newsgroups: comp.sys.next.programmer Subject: WebObjects generate HTML batch printing Date: Wed, 16 Jul 1997 16:37:11 -0600 Organization: ihc Message-ID: <33CD4D17.79F7@ihc.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit I have created, in HTML, a report to be accessed over the my company’s intranet. The reports are formatted in HTML Netscape 3.01 (all users will be using Netscape) to look just how they should look and print. The reports use WOConditional and WORepetition so depending on the options chosen, they will produce different content, but the format is the same. Users will be viewing some of these reports online and also I need to auto (batch) print this report nightly of about 20 to 500 reports. Does anyone know how I could do this? Remember the WOComponent generate the pages, so each report needs to be generate before it can be printed.
Sender: softshare@aol.com Control: cancel <33ccc96c.0@news.netzilla.net> Message-ID: <cancel.33ccc96c.0@news.netzilla.net> Subject: cmsg cancel <33ccc96c.0@news.netzilla.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 17 Jul 1997 02:09:30 GMT EMP article removed by jem@xpat.com. Original Headers: From: softshare@aol.com Subject: I want to buy software Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!infeed2.internetmci.com!newsfeed.internetmci.com!in2.uu.net!192.215.247.55!news1.ni.net!feed2.newsfeed.com!news.ntl.net!news-feed.inet.com!gateway.cm.org!news.cm.org!bcarh189.bnr.ca!bcarh8ac.bnr.ca!news.alt.net!news1.alt.net!news.jam.com!news1.exit109.com!wilbur.ohww.norman.ok.us!rahul.net!news.iswest.net!iswest.com!news.karlia.ru!news.netzilla.net!149.229.230.113 Lines: 15
From: arti@address.in.signature (Art Isbell) Newsgroups: comp.sys.next.programmer Subject: Re: Openstep examples OO design (was Re: Some initial OpenStep notes) Date: 17 Jul 1997 04:52:21 GMT Organization: LavaNet, Inc. Distribution: world Message-ID: <5qk8e5$8ci@mochi.lava.net> References: <5pmd4m$i21$2@news.xmission.com> <5q49m4$mt@saturn.genoa.com> <5q6p58$uh0$1@sps1.phys.vt.edu> <maury-1607971218030001@199.166.204.230> <5qiv5o$h0l$1@sps1.phys.vt.edu> nurban@sps1.phys.vt.edu (Nathan Urban) wrote: > In article <maury-1607971218030001@199.166.204.230>, maury@softarc.com (Maury Markowitz) wrote: > > In article <5q6p58$uh0$1@sps1.phys.vt.edu>, nurban@vt.edu wrote: > > > How would you implement an undo system? > > Personally I'd have the NeXT people at Apple design a really good one > > and put it in the AppKit. > Me too, but supposing they don't.. I'd check out the UndoManager miniexample that is (was?) in NeXTanswers. It's NEXTSTEP, not OPENSTEP, but it would probably be a good starting point. Or you might poke around in EOF's EOUndoManager for some ideas, APIs, etc. -- Art Isbell NeXT/MIME Mail: arti at lava dot net Trego Systems (for whom I don't speak) Voice/Fax: +1 808 394 0511 OPENSTEP/NT Voice Mail: +1 808 394 0495 managed care solutions US Mail: Honolulu, HI 96825-2638
From: ericet@CAM.ORG (Eric Tremblay) Newsgroups: comp.sys.next.programmer Subject: Over-ride the miniaturize function. Date: 17 Jul 1997 03:51:34 GMT Organization: Communications Accessibles Montreal, Quebec Canada Message-ID: <5qk4s6$5lu@tandem.CAM.ORG> Hello, A friend of mine and myself are trying to do something which I know is possible but we just can't figure it out. What we want (just for the knowledge of it) is over-ride the miniaturize function. So when you click on the miniaturize button in the upper left hand corner of a window instead of miniaturizing the window we want to hear a sound. How do we over-ride such basic functions in NEXTSTEP or OpenStep? Thanks in advance. -- Eric "E.T." Tremblay E-Mail: ericet@cam.org (NeXTMail - MIME - ASCII) Homepage: http://www.cam.org/~ericet
Sender: gary099g@erols.com Control: cancel <5qkd4e$kod@news1-alterdial.uu.net> Message-ID: <cancel.5qkd4e$kod@news1-alterdial.uu.net> Subject: cmsg cancel <5qkd4e$kod@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 17 Jul 1997 06:12:41 GMT EMP article removed by jem@xpat.com. Original Headers: From: gary099g@erols.com Subject: Get FREE SEX SITE..password is Newsgroups: comp.sys.next.programmer Path: ...!newsfeed.direct.ca!news-feed1.tiac.net!uunet!not-for-mail Lines: 99
Sender: 187734757@compuserve.com Control: cancel <5qkd4c$ln6@news1-alterdial.uu.net> Message-ID: <cancel.5qkd4c$ln6@news1-alterdial.uu.net> Subject: cmsg cancel <5qkd4c$ln6@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 17 Jul 1997 06:12:31 GMT EMP article removed by jem@xpat.com. Original Headers: From: 187734757@compuserve.com Subject: Get FREE PASSWORD TO 2000 SEX SITE..... Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!news-peer.sprintlink.net!news.sprintlink.net!Sprint!uunet!not-for-mail Lines: 15
From: Jonathan Hendry <jon@subsequent.com> Newsgroups: comp.sys.next.programmer Subject: Re: Over-ride the miniaturize function. Date: 17 Jul 1997 08:26:36 GMT Organization: Netcom Sender: Jonathan Hendry <jon@aahz.ix.netcom.com> Message-ID: <5qkkvs$oas@dfw-ixnews12.ix.netcom.com> References: <5qk4s6$5lu@tandem.CAM.ORG> Eric Tremblay <ericet@CAM.ORG> wrote: > Hello, > A friend of mine and myself are trying to do something which I know is > possible but we just can't figure it out. > What we want (just for the knowledge of it) is over-ride the > miniaturize function. So when you click on the miniaturize button in > the upper left hand corner of a window instead of miniaturizing the > window we want to hear a sound. > How do we over-ride such basic functions in NEXTSTEP or OpenStep? Create a subclass of Window which implements your override of miniaturize. In it, call [super miniaturize:sender] if you want the window to miniaturize in addition to making a sound. When your app starts up, before any Windows are created, call [YourWindowSubclass poseAs:Window]. Note, this will only apply to applications which are built to include your subclass. It won't change all running apps.
From: LTaylor7@ix.netcom.com (Daniel L. Taylor) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Thu, 17 Jul 1997 05:30:03 -0700 Organization: Netcom Message-ID: <LTaylor7-1707970530030001@vic-ca1-06.ix.netcom.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <jpalmer-1507972144050001@jpalmer.vip.best.com> <17035036@NEWS.SAIC.COM> <5qjib4$7mr$1@ironhorse.plsys.co.uk> In article <5qjib4$7mr$1@ironhorse.plsys.co.uk>, mmalcolm crawford <malcolm@plsys.co.uk> wrote: > Joseph Palmer <jpalmer@best.com> wrote: > > >Yeah, the BeOS has that, it does look cool. But, they didn't need DPS > >for that. > > > Does BeOS print yet...? > > Best wishes, Don't knock Be OS. It's quite nice. And you have to give it to Be fans, they have guts to support a new platform after seeing what the press has tried to do to a 13 year old platform owned by a company higher on the Fortune 500 than MS. -- Daniel L. Taylor LTaylor7@ix.netcom.com ================================================================= The Heritage Foundation /// Town Hall http://www.heritage.org/heritage/ /// http://www.townhall.com/ =================================================================
Sender: url123easy@aol.com (Url123easy) Control: cancel <19970717153500.LAA10836@ladder01.news.aol.com> Message-ID: <cancel.19970717153500.LAA10836@ladder01.news.aol.com> Subject: cmsg cancel <19970717153500.LAA10836@ladder01.news.aol.com> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 17 Jul 1997 15:35:37 GMT EMP article removed by jem@xpat.com. Original Headers: From: url123easy@aol.com (Url123easy) Subject: Job and Income Opportunity - 123 Easy Web Promotion Reseller Program Newsgroups: comp.sys.next.programmer Path: ...!news-peer.gsl.net!news.gsl.net!howland.erols.net!portc02.blue.aol.com!newstf02.news.aol.com!audrey01.news.aol.com!not-for-mail Lines: 53
From: url123easy@aol.com (Url123easy) Newsgroups: comp.sys.next.programmer Subject: Job and Income Opportunity - 123 Easy Web Promotion Reseller Program Date: 17 Jul 1997 15:35:38 GMT Message-ID: <19970717153500.LAA07136@ladder02.news.aol.com> Organization: AOL http://www.aol.com Job and Income Opportunity - 123 Easy Web Promotion Reseller Program Gideon Associates Inc. provides the great income opportunity for you to resell our 123 Web Promotion Service. We feel we have least restrictions and the best incentives on the net. We provide services and you get commissions. This is NOT a program where we take money from you! There is no "financial investment" required to become a reseller of our 123 easy Web promotion service. The Gideon Reseller program was created so that we can help even more companies and organizations benefit from our services while helping individuals become financially independent. This program was designed for anyone with access to the Internet. Those with sales experience, motivation, and/or good working knowledge of the Internet will find this program to be more financially rewarding. This is also a perfect opportunity for Web designers & Resellers, Internet Service Providers(ISPs) and Internet Presence Providers(IPPs) who wish to offer their clients the best service to promote their Web pages. Reseller could be either company or individual. How does the Reseller program work? The reseller's main responsibility is selling Gideon's 123 Easy Web Promotion Service. We take care of everything else. The reseller does not have to be bothered with handling technical questions. If the customer has questions they can contact us. We handle all the billing for customers. At the beginning of each month, we send out commission checks to the resellers. Reseller Compensation Gideon Associates Inc. credits monthly statement of the reseller's CONTACT account. The sales commission percentage is 25% on each service sold, not intitial set-up fees. For example, if you sell a site with 10 URLs submission, the price will be $449.50. Your income will be $112.38. For more details, please visit: http://members.aol.com/url123easy/sales.html 123 Easy Web Promotion Gideon Associates Inc. Email: URL123easy@aol.com Home Page: http://members.aol.com/url123easy/
From: alanf@izzy.net Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 17 Jul 1997 17:00:12 GMT Organization: "Comshare, Inc." Message-ID: <5qlj2s$p9h$1@inet-prime.comshare.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <jpalmer-1507972144050001@jpalmer.vip.best.com> <17035036@NEWS.SAIC.COM> Cc: Joseph Palmer <jpalmer@best.com> Joseph Palmer <jpalmer@best.com> wrote: >>Then why is it DPS had been around for 8 or 10 years and has not been >picked up by IBM, Apple, Microsoft, etc.... for their OS efforts? Actually, IBM, Sun, and HP do/did license Adobe PS/DPS, attempting to make X-windows almost useable. Regards, Alan Frabutt (alanf@izzy.net) disclaimer: my opinions, not my employers...
Sender: url123easy@aol.com (Url123easy) Control: cancel <19970717153500.LAA07136@ladder02.news.aol.com> Message-ID: <cancel.19970717153500.LAA07136@ladder02.news.aol.com> Subject: cmsg cancel <19970717153500.LAA07136@ladder02.news.aol.com> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 17 Jul 1997 15:35:44 GMT EMP article removed by jem@xpat.com. Original Headers: From: url123easy@aol.com (Url123easy) Subject: Job and Income Opportunity - 123 Easy Web Promotion Reseller Program Newsgroups: comp.sys.next.programmer Path: ...!news-xfer.netaxs.com!europa.clark.net!howland.erols.net!portc02.blue.aol.com!audrey02.news.aol.com!not-for-mail Lines: 53
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: How to save Apple Date: Thu, 17 Jul 1997 14:54:35 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <Ennadf600UhB01xnpF@andrew.cmu.edu> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> Excerpts from netnews.comp.sys.next.advocacy: 17-Jul-97 Re: How to save Apple by Daniel L. Taylor@ix.netc > Don't knock Be OS. It's quite nice. If you mean that no one should criticise BeOS unjustly, well, I don't disagree, but I also don't expect it will happen. :-) If you mean that no one should ever criticise BeOS even if they have a valid point, forget it. Any OS deserves legitimate criticism, and Be would do well to pay attention and use such criticism constructively. > And you have to give it to Be fans, they have guts to support a new > platform after seeing what the press has tried to do to a 13 year old > platform owned by a company higher on the Fortune 500 than MS. Shrug. Your comments could well be applied to people running NEXTSTEP. I have nothing against Be or it's users, and I hope they do well, but I don't think it's too likely. They strike me as trying to emulate what NeXT did, but they don't appear to have the level of innovation and compelling advantages that NEXTSTEP did when it was released. NEXTSTEP/OpenStep has an first-rate development environment and class libraries which are mature, well-tested, and very portable, whereas Be has got years to go before it reaches that point. For example, does Be have any kind of cross-platform/cross-compilation environment available? Is it likely that BeOS will run on any other CPU types besides Intel x86's? Where's something equivalent to the OpenStep API, EOF, or WebObjects? -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: bettis@inetnebr.com (Mr. Jeremy Bettis) Newsgroups: comp.sys.next.programmer Subject: Re: How can you turn on/off the Windows NT wait cursor? Date: 17 Jul 1997 14:36:44 -0500 Organization: Internet Nebraska Message-ID: <5qls8c$62v$1@falcon.inetnebr.com> References: <5q3u8h$mt@saturn.genoa.com> NNTP-Posting-User: bettis Alex Blakemore <alex@genoa.com> writes: >Does anyone know how to turn on and off the wait cursor under OS/NT4.2? >It no longer happens automatically, but I heard you could make some function >calls to some Microsoft stuff - say bracketed a long operation. #ifdef WIN32 SetCursor(LoadCursor(NULL,IDC_WAIT)); #endif And it will turn back to normal automaticaly. This works for me anyhow. Once in a while it stays as the wait cursor until you move the mouse. But only in rare cases. (I think it has to do with turning on the wait cursor during a -drawRect: call. )
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: distributed programming Date: 17 Jul 1997 19:14:01 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5qlqtp$i47$1@owl.slip.net> References: <5qjkij$e6v@rumah.pc.my> Michael Olan <michael@rumah.pc.my> wrote: > I have 2 slabs connected via ethernet, and I'd like to experiment with > writing code to use both cpu's. Could someone point me in the direction of > tutorials, doc's, etc. on how to do this? C++ preferred, but any Next > supported language would be fine. Anything simple for starters, like say > matrix processing would be great. Look in the NeXT Docs (/NextLibrary/Documentation/...) for the NSConnection and NSProxy classes. -- Steve Dekorte - OpenStep consultant - San Francisco
From: FullName <markfr@mdhost.cse.tek.com> Newsgroups: comp.sys.next.programmer Subject: Pixel size of a view Date: Thu, 17 Jul 1997 17:47:34 -0700 Organization: Tektronix, Inc. Message-ID: <33CEBD26.4A34@mdhost.cse.tek.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Does anyone know how to determine how to get the number of pixels wide and high in a view? How about if I previously modified the view using, setDrawSize: and setDrawOrigin:? - Thanks for any help, Mark mark.s.frank@tek.com
Sender: Free Cable<lartermen@mindspring.com> Control: cancel <5qme35$8m9@newsfep2.sprintmail.com> Message-ID: <cancel.5qme35$8m9@newsfep2.sprintmail.com> Subject: cmsg cancel <5qme35$8m9@newsfep2.sprintmail.com> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 18 Jul 1997 02:28:29 GMT EMP article removed by jem@xpat.com. Original Headers: From: Free Cable<lartermen@mindspring.com> Subject: CABLE BOX DESCRAMBLER BUILD YOUR OWN DESCRAMBLER FAST CHEAP AND EASY Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!europa.clark.net!ais.net!sprintmail!nntp.sprintmail.com!news@sprintmail.com Lines: 23
From: Lars Immisch <lars@ibp.de> Newsgroups: comp.sys.next.programmer Subject: Re: timed wait on condition Date: Thu, 17 Jul 1997 15:36:17 +0200 Organization: Immisch, Becker & Partner Message-ID: <33CE1FD1.7566@ibp.de> References: <33CB4E8D.224A@ibp.de> <0nnD=5i00UzxQ1oiF6@andrew.cmu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit CC: Charles William Swiger <cs4w+@andrew.cmu.edu> Charles William Swiger wrote: > Unix signals (like the alarm clock) get delivered to the currently > running thread; there is no per-thread signal delivery. This is because > the BSD signal mechanism is really based off of various Mach exceptions > which get translated to the appropriate signal. If you want to do > per-thread exception handling, look at catch_exception_raise() and the > "Exception Handling" example, but it's more effort than it's worth for > this. Thank you for clarifying. I had suspected that, but couldn't fully understand the Mach exception handling in conjunction with signals. > > Alternatively, I consider using msg_receive/msg_send to emulate a > > condition. > > Probably easier. Here is a function which implements a per-thread sleep > in milliseconds which you might find useful: > > void thread_msleep(unsigned msec) [nice code snipped] Hmmm. I always use thread_switch(THREAD_NULL, SWITCH_OPTION_WAIT, msec) for a millisecond sleep... > In order to do a timed condition_wait(), I'd probably set up a variable > with adequate scope to indicate whether a timeout had occurred, fork off > a new thread and use thread_switch to go to it, do a thread_msleep() for > however long in the new thread, set the timeout variable, and signal the > condition. The original thread can then check the variable and see > whether it got timed out. Yes. That should do fine when one thread waits. [I'm talking to myself to understand this in detail] But what will happen when multiple threads wait on the condition? Every thread will have it's own wakeup thread and needs it's own variable that indicates a timeout, so it might be on the stack of the waiting thread and the wakeup thread has a pointer to that variable that can be set to nil when the condition is signalled otherwise. Yes. Another way to do this would be to take a single thread that acts as timeout server. It has a linked list of timeouts and associated pointers to waiting threads. To do this elegantly, one would need a cancallable millisecond sleep, but this could be done with your thread_msleep(). The advantage of this method is constant thread overhead. That would be a bit of work, but I have code for it already. Hmmm. Thanks a lot for your consideration, Lars -- mailto:immisch@pobox.com http://pobox.com/~immisch Yesterdays yellow yoyo can make you yawn today
From: bradley@apple.com (Bob Bradley) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Fri, 18 Jul 1997 01:22:49 +0100 Organization: Organ...Uh huh huh Message-ID: <bradley-1807970122490001@bns.vip.best.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> In article <19970717190420809661@accs-as07-dp12.dlls.grid.net>, memphis@macconnect.com (Garry Roseman) wrote: > We Apple revolutionaries could, if we had the balls, develop a system > with a click-compile industrial-strength iconic language environment on > a fully PMT objectized game-like machine. This will be possible because > with OpenStep we will be able to quickly create entirely new application > metaphors with interactive objects, much as NextStep was able to create > compelling integrated application workspaces that home users and graphic > professionals alike have been anxiously awaiting. Kinda of off topic but, do you know of any NeXTStep/OPENSTEP applications that might demonstrate this? Thanks
From: memphis@macconnect.com (Garry Roseman) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Thu, 17 Jul 1997 19:04:20 -0500 Organization: Writer & Freelance Programmer Message-ID: <19970717190420809661@accs-as07-dp12.dlls.grid.net> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> Daniel L. Taylor <LTaylor7@ix.netcom.com> wrote concerning Lawson's modest proposal: > Internet users would be far more interested in having all of the latest > plug-ins and never having to apologize to any Wintel sites. I think I understand your orientation: it's more important to have all of the same features as Wintel and go head-to-head with them. We would never have to apologize again. Apparently, if I am understanding correctly, the advantage of Rhapsody is that it has fully PMT, PM, OO OS with DPS, and other such things that are very important to educational and home users. We Apple revolutionaries could, if we had the balls, develop a system with a click-compile industrial-strength iconic language environment on a fully PMT objectized game-like machine. This will be possible because with OpenStep we will be able to quickly create entirely new application metaphors with interactive objects, much as NextStep was able to create compelling integrated application workspaces that home users and graphic professionals alike have been anxiously awaiting. -- Garry Roseman <mailto:memphis@macconnect.com> Writer & Freelance Programmer Memphis TN USA
From: pete@ohm.york.ac.uk (-bat.) Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: 17 Jul 1997 16:56:19 GMT Organization: The University of York, UK Sender: pcf1@york.ac.uk Message-ID: <5qlirj$ojm$1@netty.york.ac.uk> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> Thomas Harte <tph1001@CL.cam.ac.uk> writes: > totally new to NEXTSTEP and even to GNU cc! > > problem with uname. i get a compile error: > > bash-2.00$ cc -g -o f f.c > f.c: In function `main': > f.c:7: storage size of `name' isn't known you need cc -posix to get the compile error to stop. If you are running under NS 3.3 then you must also link with -lposix. i.e. cc -posix -g -o f f.c -lposix this won;'t work under 4.0 and above 'cos they took away posix compliance ! -bat.
From: LTaylor7@ix.netcom.com (Daniel L. Taylor) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Fri, 18 Jul 1997 08:17:30 -0700 Organization: Netcom Message-ID: <LTaylor7-1807970817300001@vic-ca1-25.ix.netcom.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> In article <19970717190420809661@accs-as07-dp12.dlls.grid.net>, memphis@macconnect.com (Garry Roseman) wrote: > Daniel L. Taylor <LTaylor7@ix.netcom.com> wrote concerning Lawson's > modest proposal: > > > Internet users would be far more interested in having all of the latest > > plug-ins and never having to apologize to any Wintel sites. > > I think I understand your orientation: it's more important to have all > of the same features as Wintel and go head-to-head with them. We would > never have to apologize again. It's important to have all of our advantages but not give then any advantages. How many times has Apple lost a sale because "Well, I need THIS program...". That's such an easy problem to solve simply by supporting Rhapsody development. > Apparently, if I am understanding correctly, the advantage of Rhapsody > is that it has fully PMT, PM, OO OS with DPS, and other such things that > are very important to educational and home users. The PM, UI, and ease of maintenance will be most important to home users. Rhapsody should be more stable than Mac OS, yet just as easy to use, a big gain for home/education. -- Daniel L. Taylor LTaylor7@ix.netcom.com ================================================================= The Heritage Foundation /// Town Hall http://www.heritage.org/heritage/ /// http://www.townhall.com/ =================================================================
From: LTaylor7@ix.netcom.com (Daniel L. Taylor) Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: How to save Apple Date: Fri, 18 Jul 1997 08:13:37 -0700 Organization: Netcom Message-ID: <LTaylor7-1807970813380001@vic-ca1-25.ix.netcom.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <Ennadf600UhB01xnpF@andrew.cmu.edu> In article <Ennadf600UhB01xnpF@andrew.cmu.edu>, Charles William Swiger <cs4w+@andrew.cmu.edu> wrote: > Excerpts from netnews.comp.sys.next.advocacy: 17-Jul-97 Re: How to save > Apple by Daniel L. Taylor@ix.netc > > Don't knock Be OS. It's quite nice. > > If you mean that no one should criticise BeOS unjustly, well, I don't > disagree, but I also don't expect it will happen. :-) > > If you mean that no one should ever criticise BeOS even if they have a > valid point, forget it. Any OS deserves legitimate criticism, and Be > would do well to pay attention and use such criticism constructively. The former and NOT the latter. -- Daniel L. Taylor LTaylor7@ix.netcom.com ================================================================= The Heritage Foundation /// Town Hall http://www.heritage.org/heritage/ /// http://www.townhall.com/ =================================================================
Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy From: nagle@netcom.com (John Nagle) Subject: Re: How to save Apple Message-ID: <nagleEDIwKA.IKB@netcom.com> Organization: Netcom On-Line Services References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <Ennadf600UhB01xnpF@andrew.cmu.edu> <LTaylor7-1807970813380001@vic-ca1-25.ix.netcom.com> Date: Fri, 18 Jul 1997 16:31:22 GMT Sender: nagle@netcom6.netcom.com LTaylor7@ix.netcom.com (Daniel L. Taylor) writes: >In article <Ennadf600UhB01xnpF@andrew.cmu.edu>, Charles William Swiger ><cs4w+@andrew.cmu.edu> wrote: >> Excerpts from netnews.comp.sys.next.advocacy: 17-Jul-97 Re: How to save >> Apple by Daniel L. Taylor@ix.netc >> > Don't knock Be OS. It's quite nice. It is quite nice. But it's not done yet. One annoying BeOS problem, and one that shouldn't happen in a protected-mode OS, is that when an application crashes, things tend to be left dangling and you have to reboot to clean up. Spend ten minutes in their excuse for a web browser and you can experience this. Still, the BeOS is the only OS on the market that was designed for use with a modern window system. (MS Windows has too much DOS legacy, NextStep has too much UNIX legacy, and the MacOS has too much crammed-into-128K legacy). If Apple had bought Be, they might have been better off, and it sure would have been more fun. It's hard to get excited about NextStep, which was a flop as a standalone product. John Nagle
Newsgroups: comp.sys.next.programmer From: bofh@keltia.freenix.fr (Ollivier Robert) Sender: Dan Harley<netpro@op.net> Message-ID: <cancel.5q58ko$284$6732@picasso.op.net> Subject: cmsg cancel <5q58ko$284$6732@picasso.op.net> ignore Control: cancel <5q58ko$284$6732@picasso.op.net> Organization: Usenet Canal Historique Date: 11 Jul 1997 12:23:52 GMT ECP/EMP aka SPAM or pyramidal scheme (MMF) cancelled by bofh@keltia.freenix.fr. It may also be an image too small for newsbot to be activated. See report in news.admin.net-abuse.bulletins. Date: Fri Jul 18 18:07:45 1997 Original subject was: Best price on the net CD-R 74 Minute Gold $2.99 Retail package - no rebates no gimmics no surcharges
From: croehrig@cs.ubc.ca (Chris Roehrig) Newsgroups: comp.sys.next.programmer Subject: Is bcopy/memmove thread-safe? Date: 18 Jul 1997 19:20:27 GMT Organization: Computer Science, University of B.C., Vancouver, B.C., Canada Message-ID: <5qoflr$3ra$1@nnrp.cs.ubc.ca> Under NEXTSTEP 3.3. The NeXT docs say that malloc "and its related functions" are thread-safe. Does this include memmove? If not, what's my best bet for moving large chunks of memory in user task space? -- Chris Roehrig croehrig@House.ORG Neuroscience and Computer Science at University of British Columbia, Vancouver http://www.House.ORG/chris http://www.sns.cs.ubc.ca/chris
Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer From: nagle@netcom.com (John Nagle) Subject: Re: How to save Apple Message-ID: <nagleEDIw6C.HGy@netcom.com> Organization: Netcom On-Line Services References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> Date: Fri, 18 Jul 1997 16:23:00 GMT Sender: nagle@netcom6.netcom.com bradley@apple.com (Bob Bradley) writes: >In article <19970717190420809661@accs-as07-dp12.dlls.grid.net>, >memphis@macconnect.com (Garry Roseman) wrote: >> We Apple revolutionaries could, if we had the balls, develop a system >> with a click-compile industrial-strength iconic language environment on >> a fully PMT objectized game-like machine. This will be possible because >> with OpenStep we will be able to quickly create entirely new application >> metaphors with interactive objects, much as NextStep was able to create >> compelling integrated application workspaces that home users and graphic >> professionals alike have been anxiously awaiting. >Kinda of off topic but, do you know of any NeXTStep/OPENSTEP applications >that might demonstrate this? There's Prograph. But graphical programming doesn't scale well; you can write a small program that way, but big programs turn into acres of spagetti. It's worth noting that design of complex electronic systems, traditionally done in a graphical way, is moving to programming languages for hardware, like VHDL. The near future of routine programming seems to be Visual Basic, with purchased objects for the hard parts. The Mac still doesn't have an answer to that. John Nagle
From: stevehix@DeleteToReplysafemail.com (Steve Hix) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Thu, 17 Jul 1997 10:21:58 -0700 Organization: South Valley Internet Message-ID: <stevehix-1707971021580001@ip36.safemail.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <jpalmer-1507972144050001@jpalmer.vip.best.com> <17035036@NEWS.SAIC.COM> <5qjib4$7mr$1@ironhorse.plsys.co.uk> <jpalmer-1607971639200001@jpalmer.vip.best.com> <5qjvgu$hlq@sjx-ixn7.ix.netcom.com> In article <5qjvgu$hlq@sjx-ixn7.ix.netcom.com>, Jonathan Hendry <jon@subsequent.com> wrote: > Joseph Palmer <jpalmer@best.com> wrote: > > In article <5qjib4$7mr$1@ironhorse.plsys.co.uk>, mmalcolm crawford > > <malcolm@plsys.co.uk> wrote: > > > > Joseph Palmer <jpalmer@best.com> wrote: > > > > > > >Yeah, the BeOS has that, it does look cool. But, they didn't need DPS > > > >for that. > > > > > > > Does BeOS print yet...? > > > YES > > > And you can order it > > > TODAY. > > > > But why would I? > > One just-barely-viable platform is enough for me, thanks. Why order it? Be is providing the CD free (in the U.S., along with the next issue of MacUser). Check the Be webpage. Should be fun to play with, if you have one of the machines that BeOS supports.
From: LTaylor7@ix.netcom.com (Daniel L. Taylor) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Fri, 18 Jul 1997 17:59:57 -0700 Organization: Netcom Message-ID: <LTaylor7-1807971759570001@vic-ca2-10.ix.netcom.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> <nagleEDIw6C.HGy@netcom.com> In article <nagleEDIw6C.HGy@netcom.com>, nagle@netcom.com (John Nagle) wrote: > >Kinda of off topic but, do you know of any NeXTStep/OPENSTEP applications > >that might demonstrate this? > > There's Prograph. But graphical programming doesn't scale well; > you can write a small program that way, but big programs turn into > acres of spagetti. False. At least with Prograph. In fact, the larger the application, the more Prograph turns into an advantage and the better it "scales" to meet your needs. Of course Prograph is the only visual language I've seen so far that properly addresses this problem. AppWare/Microbrew does turn into a noodle problem, for instance. -- Daniel L. Taylor LTaylor7@ix.netcom.com ================================================================= The Heritage Foundation /// Town Hall http://www.heritage.org/heritage/ /// http://www.townhall.com/ =================================================================
From: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: How to save Apple Date: 18 Jul 1997 16:21:30 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5qoj8a$6tr$1@sps1.phys.vt.edu> References: <AFEFAFFC-20EF8@206.165.44.6> <Ennadf600UhB01xnpF@andrew.cmu.edu> <LTaylor7-1807970813380001@vic-ca1-25.ix.netcom.com> <nagleEDIwKA.IKB@netcom.com> In article <nagleEDIwKA.IKB@netcom.com>, nagle@netcom.com (John Nagle) wrote: > LTaylor7@ix.netcom.com (Daniel L. Taylor) writes: > Still, the BeOS is the only OS on the market that was designed > for use with a modern window system. (MS Windows has too much DOS > legacy, NextStep has too much UNIX legacy, What does that have to do with the window system? > and the MacOS has too > much crammed-into-128K legacy). If Apple had bought Be, they > might have been better off, and it sure would have been more fun. I doubt it. > It's hard to get excited about NextStep, which was a flop as a > standalone product. Well, true, but it's interesting how that "flop" is still overall a technologically superior system to anything else out there.
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: 19 Jul 1997 02:35:34 GMT Organization: WolfWare Message-ID: <5qp95m$5du$1@vader.wolfware.ipc.net> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> <5qlirj$ojm$1@netty.york.ac.uk> <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> In-Reply-To: <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> On 07/18/97, Charles William Swiger wrote: > >Speaking of which, there has been a complete lack of response to my >earlier comments about POSIX compliance for Rhapsody. Has anyone heard >anything? Hopefully, there do exist people who care whether Rhapsody is >POSIX-compliant...? At WWDC it was made pretty clear that Rhapsody would be POSIX compliant. The BSD layer is 4.4 which I believe gives them this "for free" anyways. >-Chuck > > > Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer > ----------------+---------------------+--------------------- > I know you're an optimist if you think I'm a pessimist. > > -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: Fri, 18 Jul 1997 16:31:02 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> <5qlirj$ojm$1@netty.york.ac.uk> In-Reply-To: <5qlirj$ojm$1@netty.york.ac.uk> Excerpts from netnews.comp.sys.next.programmer: 17-Jul-97 Re: uname by -bat.@ohm.york.ac.uk > you need cc -posix to get the compile error to stop. If you > are running under NS 3.3 then you must also link with -lposix. > > i.e. > > cc -posix -g -o f f.c -lposix '-posix' explicitly specifies linking with libposix.a; the -lposix is redundant (albeit harmless), as can be seen by using the '-v' option to the compiler. Check the linker line: 95-tertius% cc -posix -v -o a a.c Reading specs from /lib/m68k/specs NeXT Computer, Inc. version cc-437.2.6, gcc version 2.5.8 [ ... ] ld -arch m68k -o a -NEXTSTEP-deployment-target 3.2 -lposixcrt0.o -L/lib/m68k -L/lib/m68k /usr/tmp/cc546758-001445-a.o -lposix > this won't work under 4.0 and above 'cos they took away posix compliance ! Speaking of which, there has been a complete lack of response to my earlier comments about POSIX compliance for Rhapsody. Has anyone heard anything? Hopefully, there do exist people who care whether Rhapsody is POSIX-compliant...? -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: url123easy@aol.com (Url123easy) Newsgroups: comp.sys.next.programmer Subject: Job and Income Opportunity - 123 Easy Web Promotion Reseller Program Date: 17 Jul 1997 15:35:31 GMT Message-ID: <19970717153500.LAA10836@ladder01.news.aol.com> Organization: AOL http://www.aol.com Job and Income Opportunity - 123 Easy Web Promotion Reseller Program Gideon Associates Inc. provides the great income opportunity for you to resell our 123 Web Promotion Service. We feel we have least restrictions and the best incentives on the net. We provide services and you get commissions. This is NOT a program where we take money from you! There is no "financial investment" required to become a reseller of our 123 easy Web promotion service. The Gideon Reseller program was created so that we can help even more companies and organizations benefit from our services while helping individuals become financially independent. This program was designed for anyone with access to the Internet. Those with sales experience, motivation, and/or good working knowledge of the Internet will find this program to be more financially rewarding. This is also a perfect opportunity for Web designers & Resellers, Internet Service Providers(ISPs) and Internet Presence Providers(IPPs) who wish to offer their clients the best service to promote their Web pages. Reseller could be either company or individual. How does the Reseller program work? The reseller's main responsibility is selling Gideon's 123 Easy Web Promotion Service. We take care of everything else. The reseller does not have to be bothered with handling technical questions. If the customer has questions they can contact us. We handle all the billing for customers. At the beginning of each month, we send out commission checks to the resellers. Reseller Compensation Gideon Associates Inc. credits monthly statement of the reseller's CONTACT account. The sales commission percentage is 25% on each service sold, not intitial set-up fees. For example, if you sell a site with 10 URLs submission, the price will be $449.50. Your income will be $112.38. For more details, please visit: http://members.aol.com/url123easy/sales.html 123 Easy Web Promotion Gideon Associates Inc. Email: URL123easy@aol.com Home Page: http://members.aol.com/url123easy/
From: Jonathan Hendry <jon@subsequent.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 19 Jul 1997 04:01:44 GMT Organization: Netcom Sender: Jonathan Hendry <jon@aahz.ix.netcom.com> Message-ID: <5qpe78$sd9@sjx-ixn8.ix.netcom.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <jpalmer-1507972144050001@jpalmer.vip.best.com> <17035036@NEWS.SAIC.COM> <5qjib4$7mr$1@ironhorse.plsys.co.uk> <jpalmer-1607971639200001@jpalmer.vip.best.com> <5qjvgu$hlq@sjx-ixn7.ix.netcom.com> <stevehix-1707971021580001@ip36.safemail.com> Steve Hix <stevehix@DeleteToReplysafemail.com> wrote: > Why order it? Be is providing the CD free (in the U.S., along > with the next issue of MacUser). Check the Be webpage. > Should be fun to play with, if you have one of the machines > that BeOS supports. Does it boot from the CD? Do you have to reformat your harddrive to install it? - JH
From: daj@nwu.edu Newsgroups: comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.hardware Subject: HELP with SCSI device problem Date: 19 Jul 1997 04:45:49 GMT Organization: Flashnet Communications, http://www.flash.net Message-ID: <5qpgpt$sgs$2@excalibur.flash.net> Hello everyone, I First let me preface with I'm new to developing under OpenStep and ObjectiveC most of my work is done in Mathematica or just 'C'. I have a SCSI data acquisition device that was manufactured by Analogic, its model DASM-AD14. Technically it is a 14bit 2Mhz 8 channel acquisition device that has a 2Mb buffer and a SCSI interface. Here's my problem: Analogic provided instructions and the basic outline for acquisition code for a Sun3 or SPARC system using SunOS4.0, and I'm trying to port it over to the NeXT/Intel system. Analogic says the device can be accessed at /dev/rsdnc, where n is the scsi appropriate device number . Well I've got a properly terminated system and the acquisition device is SCSI ID target 6 LUN 0. Now the problem is analogic says it can be accessed by: char *dev_name="/dev/rsd4c"; open (dev_name, O_RDWR); but when I check /usr/adm/messages I have the errors at the bottom of this post:. Also, if I go ahead and run the program it doesn't open the device. I'm running OpenStep 4.1 on Intel hardware and I have a Adaptec 2940 SCSI Controller. I figure I probably need to write some sort of device driver for the acquisition system labeled CDA ADC 2.0 . First I haven't the faintest idea on how to write a device driver, although I've hacked quite a few C programs, I've only just begun understanding ObjectiveC. Next, in preparation I've been peaking through my system and reading everything I can get my eyes on and I found in /NextLibrary/Documentation/NextDev/ReleaseNotes/DriverDevelopment.rtf the following: OPENSTEP 4.1 cannot be used to develop Mach device drivers. Because NEXTSTEP 3.3 device drivers work on OPENSTEP 4.1 systems, we recommend that you use NEXTSTEP 3.3 Developer to create device drivers. So to make a long story shorter than what it could be, is the above note true or is there some patch I can apply to allow me to write a device driver under Openstep 4.1? Can I get NEXTSTEP3.3 Developer and use it under OPENSTEP 4.1? Last, I'm still not clear in understanding why I can't just use /dev/rsd4c as the Analogic manual suggests. The requirement as Analogic has it in there manual is that their DASM-AD14 communicates through a character-special file (ie /dev/rsd4c). Clearly the device appears unwritable (in messages below), however it should be writable. Can anyone explain why I see this problem? Oh yeah, the last note is that Analogic no longer supports the device, it appeared to be to costly to produce so there wasnt enough demand. As a result I have virttually no help from Analogic. Thank's in advance, David A. Johnson daj@nwu.edu Northwestern University ( this is just a snippet from messages ) Jul 18 14:38:56 Siqin mach: sd3: iomega jaz 1GB H.72 Jul 18 14:38:56 Siqin mach: Registering: sd3 at Target 5 LUN 0 at sc0 Jul 18 14:38:56 Siqin mach: sd3: Waiting for drive to come ready.............. Jul 18 14:38:56 Siqin mach: sd3: Disk Not Ready Jul 18 14:38:56 Siqin mach: sd4: CDA ADC 2.0 Jul 18 14:38:56 Siqin mach: sd4: Illegal request; FATAL. Jul 18 14:38:56 Siqin mach: target:6 lun:0 op:Mode Sense Jul 18 14:38:56 Siqin mach: Registering: sd4 at Target 6 LUN 0 at sc0 Jul 18 14:38:56 Siqin mach: sd4: Illegal request; FATAL. Jul 18 14:38:56 Siqin mach: target:6 lun:0 op:Mode Sense Jul 18 14:38:56 Siqin mach: sd4: WRITE PROTECTED. FATAL. Jul 18 14:38:56 Siqin mach: target:6 lun:0 op:Read block:0 blockCount:15 Jul 18 14:38:56 Siqin mach: sd4: Device Block Size: 512 bytes Jul 18 14:38:56 Siqin mach: sd4: Device Capacity: 2048 KB Jul 18 14:38:56 Siqin mach: sd4: Disk Label: UntitledDisk Jul 18 14:38:56 Siqin mach: Registering: sg0 at sc0 Jul 18 14:38:56 Siqin mach: Registering: sg1 at sc0 Jul 18 14:38:56 Siqin mach: Registering: sg2 at sc0 Jul 18 14:38:56 Siqin mach: Registering: sg3 at sc0 Jul 18 14:38:56 Siqin mach: hc0: device detected at port 0x1f0 irq 14 Jul 18 14:38:56 Siqin mach: sd4: WRITE PROTECTED. FATAL. Jul 18 14:38:56 Siqin mach: target:6 lun:0 op:Read block:0 blockCount:15 Jul 18 14:38:56 Siqin mach: hc0: Checking for ATA drive 0... Detected Jul 18 14:38:56 Siqin mach: hc0: Checking for ATA drive 1... Jul 18 14:38:56 Siqin mach: hc0: Checking for ATAPI device 1... Jul 18 14:38:56 Siqin mach: hc0: Resetting drives.. Jul 18 14:38:56 Siqin mach: Registering: hc0 Jul 18 14:38:56 Siqin mach: hd0: Conner Peripherals 850MB - CFA850A 8DT0.41 Jul 18 14:38:56 Siqin mach: hd0: 1652 cylinders, 16 heads, 63 spt (disk geometry) Jul 18 14:38:56 Siqin mach: hd0: using multisector (32) transfers. Jul 18 14:38:56 Siqin mach: Registering: hd0 Jul 18 14:38:56 Siqin mach: hd0: Device Block Size: 512 bytes Jul 18 14:38:56 Siqin mach: hd0: Device Capacity: 813 MB Jul 18 14:38:56 Siqin mach: hd0: Disk Label: Disk Jul 18 14:38:56 Siqin mach: Registering: ISASerialPort0 Jul 18 14:38:56 Siqin mach: ISASerialPort0: Base=0x03f8, IRQ=4, Type=16550AF/C/CF, FIFO=16 Jul 18 14:38:56 Siqin mach: Registering: ISASerialPort1 Jul 18 14:38:56 Siqin mach: ISASerialPort1: Base=0x02f8, IRQ=3, Type=16550AF/C/CF, FIFO=16 Jul 18 14:38:56 Siqin mach: Registering: fc0 Jul 18 14:38:56 Siqin mach: Registering: fd0 Jul 18 14:38:56 Siqin mach: Registering: PS2Controller Jul 18 14:38:56 Siqin mach: Registering: PCKeyboard0 Jul 18 14:38:56 Siqin mach: PCI bus support enabled Jul 18 14:38:56 Siqin mach: Registering: PCI0 Jul 18 14:38:56 Siqin mach: Registering: EISA0 Jul 18 14:38:56 Siqin mach: Registering: event0 Jul 18 14:38:56 Siqin mach: Registering: kmDevice0 Jul 18 14:38:56 Siqin mach: rootdev 300, howto 0 Jul 18 14:38:56 Siqin mach: Registering: PS2Mouse Jul 18 14:38:56 Siqin mach: Registering: pp0 Jul 18 14:38:56 Siqin mach: Display0: vendorID=5333 deviceID=88d0 Jul 18 14:38:56 Siqin mach: Display0: physicalAddress = 40000000 Jul 18 14:38:56 Siqin mach: Display0: 4MB RAM installed. Jul 18 14:38:56 Siqin mach: Display: Mode selected: 1280 x 1024 @ 60 Hz (RGB:256/8) Jul 18 14:38:56 Siqin mach: Registering: Display0 Jul 18 14:38:56 Siqin mach: DECchip21040: PCI Dev: 19 Func: 0 Bus: 0 Jul 18 14:38:56 Siqin mach: Registering: en0 Jul 18 14:38:56 Siqin mach: DECchip21040 based adapter at port 0xfc00 irq 9 Jul 18 14:38:56 Siqin mach: IP protocol enabled for interface en0, type "10MB Ethernet" Jul 18 14:38:56 Siqin mach: en0: Ethernet address 00:00:92:90:4a:1e Jul 18 14:38:56 Siqin mach: SB16 probed Jul 18 14:38:56 Siqin mach: SoundBlaster16 hardware version is 4.13 Jul 18 14:38:56 Siqin mach: SoundBlaster16 at dma channels 1 and 5 irq 5 Jul 18 14:38:56 Siqin mach: Registering: SoundBlaster16 Jul 18 14:38:56 Siqin mach: Registering: PDPseudo Jul 18 14:38:56 Siqin mach: Registering: ttya Jul 18 14:38:56 Siqin mach: Registering: ttyb Jul 18 14:38:56 Siqin mach: Registering: pdservd Jul 18 14:38:56 Siqin mach: Registering: NEXTIME_Sound Jul 18 14:38:56 Siqin mach: sd4: WRITE PROTECTED. FATAL. Jul 18 14:38:56 Siqin mach: target:6 lun:0 op:Read block:0 blockCount:15
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: How to save Apple Date: 19 Jul 1997 05:01:20 GMT Organization: monoChrome, Inc., NJ, USA Message-ID: <5qphn0$9t6$3@news2.digex.net> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <Ennadf600UhB01xnpF@andrew.cmu.edu> <LTaylor7-1807970813380001@vic-ca1-25.ix.netcom.com> <nagleEDIwKA.IKB@netcom.com> > Still, the BeOS is the only OS on the market that was designed > for use with a modern window system. (MS Windows has too > much DOS legacy, NextStep has too much UNIX legacy, and the > MacOS has too much crammed-into-128K legacy). If Apple had > bought Be, they might have been better off, and it sure > would have been more fun. It's hard to get excited about > NextStep, which was a flop as a standalone product. Unfortunatly BeOS was also crippled w/ C++, far more damning than unix or even DOS...IMO, YMMV. ObjC allows OS/Rhapsody to evolve beyond itself in a way that C++ wont...as will be evidenced with its coupling to Java...which I imagine will provide a relatively smooth transition into full OO environments, filesystems...somewhere into the realm of smalltalk developer dreams...maybe beyond... -- Thanks, be well, take care, later, John Kheit; Self expressed... __________________________________________________________________ monoChrome, Inc. ASCII, MIME, PGP, SUN, & NeXTmail OK NeXT/OPENSTEP Developer mailto:jkheit@cnj.digex.net Telepathy, It's coming... http://www.cnj.digex.net/~jkheit New York Law School You're dangerous because you're honest
From: markeaton@mindspring.com (Mark Eaton) Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: How to save Apple Date: Fri, 18 Jul 1997 22:21:37 -0800 Organization: MindSpring Enterprises Message-ID: <markeaton-1807972221370001@ip72.santa-clara10.ca.pub-ip.psi.net> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <Ennadf600UhB01xnpF@andrew.cmu.edu> <LTaylor7-1807970813380001@vic-ca1-25.ix.netcom.com> <nagleEDIwKA.IKB@netcom.com> <5qphn0$9t6$3@news2.digex.net> In article <5qphn0$9t6$3@news2.digex.net>, jkheit@cnj.digex.net wrote: > Unfortunatly BeOS was also crippled w/ C++, far more damning than > unix or even DOS...IMO, YMMV. gee, I dunno. DOS was the worst blight on microcomputers since, well, ever... C++ is merely misguided (suffering too much from design by commitee; what we have now is not what Stroustrop originally gave us). DOS was just plain evil in design as well as intent. > ObjC allows OS/Rhapsody to evolve > beyond itself in a way that C++ wont...as will be evidenced with > its coupling to Java...which I imagine will provide a relatively specifically the Rhapsody dynamic runtime and not Objective-C the language. ObjC is nice by itself, but its the runtime that will make Rhapsody unique among all commercial OSes. (this is just my opinion) ---> markeaton_@_mindspring_._com
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: How to save Apple Date: 19 Jul 1997 05:45:54 GMT Organization: monoChrome, Inc., NJ, USA Message-ID: <5qpkai$n62$1@news2.digex.net> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <Ennadf600UhB01xnpF@andrew.cmu.edu> <LTaylor7-1807970813380001@vic-ca1-25.ix.netcom.com> <nagleEDIwKA.IKB@netcom.com> <5qphn0$9t6$3@news2.digex.net> <markeaton-1807972221370001@ip72.santa-clara10.ca.pub-ip.psi.net> markeaton@mindspring.com (Mark Eaton) wrote: > In article <5qphn0$9t6$3@news2.digex.net>, jkheit@cnj.digex.net > wrote: > > Unfortunatly BeOS was also crippled w/ C++, far more damning > > than unix or even DOS...IMO, YMMV. > gee, I dunno. DOS was the worst blight on microcomputers since, > well, ever... C++ is merely misguided (suffering too much from > design by commitee; what we have now is not what Stroustrop > originally gave us). DOS was just plain evil in design as well > as intent. Don't get me wrong, I'm w/ you in spirit here...lord knows I don't want to ever be accused of defending dos :) But DOS was kinda ok for it's time and purpose too...remember bgates just bought it cup at the time when CPM was the cat's meow...so it wasn't all too bad then...I mean compared to unix, yea it blew...but at the time and stuff, it was reasonable for the hardware... > > ObjC allows OS/Rhapsody to evolve beyond itself in a way that > > C++ wont...as will be evidenced with its coupling to Java...which > > I imagine will provide a relatively > specifically the Rhapsody dynamic runtime and not Objective-C > the language. ObjC is nice by itself, but its the runtime that > will make Rhapsody unique among all commercial OSes. (this is > just my opinion) I agree...I didn't feel like going into all the ObjC details as to what makes it better and whatnot...so I just left it kinda vaugue, if not intellectually slothful (my MO :). -- Thanks, be well, take care, later, John Kheit; Self expressed... __________________________________________________________________ monoChrome, Inc. ASCII, MIME, PGP, SUN, & NeXTmail OK NeXT/OPENSTEP Developer mailto:jkheit@cnj.digex.net Telepathy, It's coming... http://www.cnj.digex.net/~jkheit New York Law School You're dangerous because you're honest
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: Is bcopy/memmove thread-safe? Date: Sat, 19 Jul 1997 01:41:15 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <Ano5BvS00UzxA2prRs@andrew.cmu.edu> References: <5qoflr$3ra$1@nnrp.cs.ubc.ca> In-Reply-To: <5qoflr$3ra$1@nnrp.cs.ubc.ca> Excerpts from netnews.comp.sys.next.programmer: 18-Jul-97 Is bcopy/memmove thread-safe? by Chris Roehrig@cs.ubc.ca > Under NEXTSTEP 3.3. The NeXT docs say that malloc "and its related > functions" are thread-safe. Does this include memmove? I don't think so, no. > If not, what's my best bet for moving large chunks of memory in user > task space? One obvious way is to hold a mutex over any memory moves. Define 'large'? If you're moving very large amounts of memory, it may well be to your advantage to look into using page-aligned memory and remapping pages rather than actually copying bytes around. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: bradley@apple.com (Bob Bradley) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Fri, 18 Jul 1997 23:27:35 +0100 Organization: Organ...Uh huh huh Message-ID: <bradley-1807972327360001@bns.vip.best.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> <nagleEDIw6C.HGy@netcom.com> In article <nagleEDIw6C.HGy@netcom.com>, nagle@netcom.com (John Nagle) wrote: What I meant was, is there any software for NeXTStep/OPENSTEP that demonstrate the following quote: "...NextStep was able to create compelling integrated application workspaces that home users and graphic professionals alike have been anxiously awaiting."
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 19 Jul 1997 04:08:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFF5F2E0-BD04@206.165.44.1> References: <*johnnyc*-1207971235030001@192.0.2.1> nntp://news.primenet.com/comp.sys.mac.programmer.games, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit > Yes, but not as a replacement. The current draw libraries could be > incorporated into the system with only a few hundred k of code. Put the > PC version on the net and tote it as the standard for lightweight internet > vector grphics. There is no single move that could be done to speed the > internet than reduce graphics sizes by a factor of 10. Gee, and you could name it "Apple Internet Graphics..." It is no coincidence that 2 of the 3 vector libraries recently reviewed in MacUser (MacWorld?) are based on GX. If you read what Foley and Van Dam have to say about immediate-mode graphics vs graphics databases, something like GX would be *ideal* for networked graphics, whereas something like DPS would be too slow for Internet use. Not sure where Taligent Graphics fits in. ++++++++++++++++++++++++++++++ HOW DO I SUBSCRIBE TO GX-TALK? --------------------------------- To subscribe to GX-TALK proceed as follows: Send E-mail to: <gx-talk-request@aimed.org> with anything in the subject line and the following command as the first (and only) line of the message body: SUBSCRIBE +++++++++++++++++++++++++++++++++
From: Alex Blakemore <alex@genoa.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Openstep examples OO design (was Re: Some initial OpenStep notes) Date: 19 Jul 1997 03:50:41 GMT Organization: Genoa Software Systems Message-ID: <5qpdih$c6@saturn.genoa.com> References: <5pmd4m$i21$2@news.xmission.com> <5q26ue$ndb$1@lazar.select-tech.si> <5q49m4$mt@saturn.genoa.com> <5q6p58$uh0$1@sps1.phys.vt.edu> <maury-1607971218030001@199.166.204.230> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: maury@softarc.com In article <5q6p58$uh0$1@sps1.phys.vt.edu>, nurban@vt.edu wrote: > > How would you implement an undo system? Well, the one Jeff Martin developed seems pretty elegant to me (and EOF seems to borrow some of the idea) Look at NextAnswer 1998 and 1582 for an article and the code. Even if you have limited interest in undo, its a nice example of the kind of thing you can accomplish with a dynamic language like Objective-C. -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: luomat@peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.programmer Subject: POSIX in Rhapsody (was Re: uname) Date: 19 Jul 1997 14:56:37 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Message-ID: <5qqkj5$2ei$3@ha2.rdc1.nj.home.com> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> <5qlirj$ojm$1@netty.york.ac.uk> <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: cs4w+@andrew.cmu.edu In <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> Charles William Swiger wrote: > > this won't work under 4.0 and above 'cos they took away posix compliance ! > > Speaking of which, there has been a complete lack of response to my > earlier comments about POSIX compliance for Rhapsody. Has anyone heard > anything? Hopefully, there do exist people who care whether Rhapsody is > POSIX-compliant...? I do! I do! As someone who frequently travels around ftp sites looking for cool programs, I am consistently disappointed when they fail to compile on NS. POSIX would be a big step in improving this condition. The fact that they removed it all for 4.x has always scared me a little for Rhapsody, but perhaps they will add it (please G'd) or plan to add it (ugh... won't ever happen if it doesn't exist early). TjL ps -- Hebrew started as of 2 July, and ends on 22 August, so please understand if responses are slow. I am taking a 2 semester class in 8 weeks for 6 credits -- TjL <luomat@peak.org> f ` cn rd ths ` my hv lrnd hbrw
From: dcoyle@ctp.com (David A. Coyle) Newsgroups: comp.sys.next.programmer Subject: Making an NSTextField scroll Date: 19 Jul 1997 17:06:58 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5qqs7i$9bd$1@concorde.ctp.com> Hi all: Does anyone know how I can make an NSTextField scroll to the right, programmatically? I have the situation where I always want my text field to show the rightmost text, even if the text is too wide for the cell. I tried [[myTextField currentEditor] scrollRangeToVisible:rangeOfKknownLastCharacter options: NSBackwardsSearch]; but that didn't seem to work... (FWIW, the field is right-aligned and scrollable) cheers.. Dave
From: fozztexx@nvc.cc.ca.us (Chris Osborn) Newsgroups: comp.sys.next.programmer Subject: How do I draw text in OpenStep? Date: 19 Jul 1997 18:16:34 GMT Organization: Napa Valley College Message-ID: <5qr0a2$ec9@wiley.napanet.net> How do I draw text in OpenStep? At the moment I've had to resort to using PSshow, which is bad because I had to use C strings. NSString can do drawAtPoint and drawInRect, *except* it seems to shift the coordinates by a *lot*. It also returns sizes much too large when you ask it for the size of the text. Is there a way to use NSText or NSTextView to draw text within my drawRect: method? -- Chris Osborn, Network Administrator Napa Valley College 707 253 3130 - Voice 2277 Napa-Vallejo Hwy. 707 253 3063 - Fax Napa, CA 94558 <fozztexx@nvc.cc.ca.us> <http://www.nvc.cc.ca.us/~fozztexx>
From: h1ouu88d77@compuserve.com Newsgroups: comp.sys.next.programmer Subject: Why pay for sex sites just.... Date: 19 Jul 1997 18:35:12 GMT Organization: fHdffdjj Ifnc. Message-ID: <5qr1d0$d0s@news1-alterdial.uu.net> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary=_-=195 This is a multi-part message in MIME format. --_-=195 Content-type: text/html; name="Sex~1.htm" Content-transfer-encoding: base64 PCFET0NUWVBFIEhUTUwgUFVCTElDICItLy9JRVRGLy9EVEQgSFRNTC8vRU4iPg0KDQo8aHRt bD4NCg0KPGhlYWQ+DQo8dGl0bGU+U0VDUkVUIFBBU1NXT1JEIEZPUiBBRFVMVCBTSUdIVFM8 L3RpdGxlPg0KPG1ldGEgbmFtZT0iR0VORVJBVE9SIiBjb250ZW50PSJNaWNyb3NvZnQgRnJv bnRQYWdlIDEuMSI+DQo8L2hlYWQ+DQoNCjxib2R5IGJnY29sb3I9IiNDMEMwQzAiIHRleHQ9 IiM4MEZGRkYiIGxpbms9IiMwMDAwRkYiIHZsaW5rPSIjMDAwMEZGIiBhbGluaz0iI0ZGMDAw MCI+DQo8ZGl2IGFsaWduPWNlbnRlcj48Y2VudGVyPg0KPHRhYmxlIGJvcmRlcj01IGNlbGxw YWRkaW5nPTI+DQo8dHI+PHRkIGFsaWduPWNlbnRlciB3aWR0aD0xMDAlPjxmb250IGNvbG9y PSIjRkZGRjAwIj48Zm9udCBzaXplPTc+PGI+RlJFRSEgRlJFRSEgRlJFRSE8YnI+DQo8L2I+ PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT01PjxiPlRI QVQnUyBSSUdIVC4uLiBDT01QTEVURUxZIDwvYj48L2ZvbnQ+PGZvbnQgc2l6ZT02PjxiPkZS RUUhITwvYj48L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjMDAwMEZGIj48Zm9udCBzaXpl PTU+PGI+PGk+PGJyPg0KRU5URVIgVEhST1VHSCBBRFVMVFNJR0hUUyAmIzE2OTsgPC9pPjwv Yj48L2ZvbnQ+PGZvbnQgc2l6ZT02Pjxicj4NCjwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9 IiNGRkZGMDAiPjxmb250IHNpemU9Nz48Yj5JVCdTIEZSRUUhISEhITwvYj48L2ZvbnQ+PC9m b250PjwvdGQ+PC90cj4NCjwvdGFibGU+DQo8L2NlbnRlcj48L2Rpdj4NCjxwIGFsaWduPWNl bnRlcj48Zm9udCBGQUNFPSJDb21pYyBTYW5zIE1TIj48Zm9udCBjb2xvcj0iIzIyMTY1QSI+ PGZvbnQgc2l6ZT03PiA8L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjRkYwMDAwIj48Zm9u dCBzaXplPTc+T05FIFBBU1NXT1JEIDxmb250IEZBQ0U9IkNvbWljIFNhbnMgTVMiPjxicj4N CjwvZm9udD48L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjMjIxNjVBIj48Zm9udCBzaXpl PTU+IDwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9IiNGRkZGMDAiPjxmb250IHNpemU9NT48 c3Ryb25nPkdFVFMgWU9VIFVOTElNSVRFRCBGUkVFIEFDQ0VTUyBUTyBPVkVSIA0KMTgwMCsg T0YgVEhFIEhPVFRFU1QgQURVTFQgU0lURVMgPC9zdHJvbmc+PC9mb250PjwvZm9udD48L3A+ DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29sb3I9IiMwMDAwMDAiPjxmb250IHNpemU9NT48 c3Ryb25nPklGIFlPVSBBUkUgVElSRUQgT0YgUEFZSU5HIFRPIEdFVCBJTlRPIFNFWCANClNJ VEVTLi4uPC9zdHJvbmc+PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGRkZGRiI+PGZv bnQgc2l6ZT01PjxzdHJvbmc+IDwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9 IiMwMEZGRkYiPjxmb250IHNpemU9NT5PbmUgUGFzc3dvcmQgaXMgYWxsIHlvdSBuZWVkISEh PGJyPg0KPC9mb250PjwvZm9udD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgRkFDRT0i Q29taWMgU2FucyBNUyI+PGZvbnQgY29sb3I9IiMyMjE2NUEiPjxmb250IHNpemU9Nj4gPC9m b250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT02PjxzdHJvbmc+ TE9PSyBXSEFUIFlPVSBHRVQhPC9zdHJvbmc+PC9mb250PjwvZm9udD48L3A+DQo8cCBhbGln bj1jZW50ZXI+PGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PGZvbnQgY29sb3I9IiMyMjE2 NUEiPjxmb250IHNpemU9NT4gPC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGRkYwMCI+ PGZvbnQgc2l6ZT01PjxzdHJvbmc+VU5MSU1JVEVEIEZSRUUgQUNDRVNTIFRPIE9WRVIgMTgw MCsgT0YgDQpUSEUgSE9UVEVTVCBBRFVMVCBTSVRFUzwvc3Ryb25nPiA8Zm9udCBGQUNFPSJD b21pYyBTYW5zIE1TIj48L2ZvbnQ+PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iIzIyMTY1 QSI+PGZvbnQgc2l6ZT01PiA8L2ZvbnQ+PC9mb250PjwvcD4NCjxoMSBhbGlnbj1jZW50ZXI+ PGZvbnQgY29sb3I9IiNGRjAwMDAiPjxmb250IHNpemU9NT48c3Ryb25nPkp1c3QgQSBGZXcg T2YgVGhlIE92ZXIgMTgwMCsgWFhYIFJhdGVkIFNpZ2h0cyBUaGF0IA0KWW91IENhbiBFbnRl ciBDb21wbGV0ZWx5IEZyZWUhPGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PC9mb250Pjwv c3Ryb25nPjwvZm9udD48L2ZvbnQ+PC9oMT4NCjxoMSBhbGlnbj1jZW50ZXI+PGZvbnQgY29s b3I9IiM4MEZGMDAiPjxmb250IHNpemU9Nj48c3Ryb25nPlRFRU4gU0lURVM8L3N0cm9uZz48 L2ZvbnQ+PC9mb250PjwvaDE+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29sb3I9IiMwMDAw MDAiPjxmb250IHNpemU9ND4xMDAwIEhPVCBBTkQgTmFLRUQgLSBDZUxlQlJpVGlFUyBBTkQg VEVFTlMgLSAxMDAxIE51ZGUgRXJvdGljIA0KVGVlbnMgLSAyMCBBbWFUZWVucyAtIDYyMyBO dWRlIENlbGVicml0aWVzIEFuZCBUZWVucyAtIEFBQSBTTFVUWSANClRFRU5TIC0gQW1hdGV1 ciBUZWVucyAtIEFtYXRldXIgVGVlbnMgRnJvbSBBcm91bmQgVGhlIFdvcmxkIC0gQW1hdGV1 ciANClRlZW5zISAtIENvY2sgSHVuZ3J5IFRlZW5zIC0gQ3VtIERyaW5raW5nIFRlZW5zIC1G aXN0IEZ1Y2tpbmcgVGVlbnMgLSBIb3QgDQpOdWRlIFRlZW5zIC0gTm90aGluZyBCdXQgVGVl bnMgLSBUZWVucyBJbiBUcm91YmxlIC0gVHdhdCBTdWNraW5nIFRlZW5zIDwvZm9udD48L2Zv bnQ+PC9wPg0KPHAgYWxpZ249Y2VudGVyPjxmb250IGNvbG9yPSIjRkYwMDAwIj48Zm9udCBz aXplPTY+PHN0cm9uZz5BU0lBTiBTSUdIVFM8L3N0cm9uZz48L2ZvbnQ+PC9mb250PjwvcD4N CjxwIGFsaWduPWNlbnRlcj48Zm9udCBjb2xvcj0iIzAwMDAwMCI+PGZvbnQgc2l6ZT00PkFs bCBBc2lhbiBUZWVuIC0gQXNpYW4gQmFiZXMgLSBBc2lhbiBCYWJlcyBzaWFuIEJhYmVzISAt IEFzaWFuIEJlbGxlcyAtIEFzaWFuIA0KQm9uZGFnZSAtIEFzaWFuIEN1dGUgR2lybHMgLSBB c2lhbiBEZWxpdGVzIC0gQXNpYW4gRXJvcyAtIEFzaWFuIEZsb3dlcnMgLSANCkFzaWFuIEdv bGQgLSBZb3VuZyBBc2lhbiBXb21lbiAtIEFzaWFuIEhhcmRDb3JlIEFTSUFOIEhBUkRDT1JF IC0gDQpUSEUgTEVHRU5EIENPTlRJTlVFUyAtIEFTSUFOIEhvbG9zdWl0ZSAtIEFzaWFuIExv dmUgRG9sbHMgPC9mb250PjwvZm9udD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29s b3I9IiNGRjAwRkYiPjxmb250IHNpemU9Nj48c3Ryb25nPkNFTEVCUklUWSBTSUdIVFM8L3N0 cm9uZz48L2ZvbnQ+PC9mb250PjwvcD4NCjxwIGFsaWduPWNlbnRlcj48Zm9udCBjb2xvcj0i IzAwMDAwMCI+PGZvbnQgc2l6ZT00PiMxIE51ZGUgQ2VsZWJyaXR5IFBhZ2UgISAtIENlbGVi d29ybGQgQWNlJ3MgLSBYWFggQ2VsZWJyaXR5IFBpY3MgLSBDZWxlYnJpdHkgDQpDZW50cmFs IC0gQ2VsZWJyaXR5IENoaWNrIFBhdHRpZXMgLSBDZWxlYnJpdHkgQ3VtYm94IC0gQ2VsZWJy aXR5IFBpY3R1cmVzIA0KQ2VsZWJyaXR5IFBpY3R1cmVzIEdvbGQgU2l0ZSAtIENlbGVicml0 eSBTa2luIC0gQ2VsZWJyaXR5IFNraW4tQUJDIC0gRGEgUm9ja3MgDQpudWRlIGNlbGVicml0 eSBBcmNoaXZlIC0gTWFya3MgQ2VsZWJyaXR5IFBpY3R1cmVzIDIgLSBOdWRlIENlbGVicml0 eSBBcmNoaXZlIC0gDQpTdGFyTG9jYXRvciAtIENlbGVicml0eSBTdXBlclMgLSBUaGUgRGFp bHkgTnVkZSBDZWxlYnJpdHkgUGFnZSAtIFRoZSBNdXN0IA0KU2VlIE5VREUgQ2VsZWJyaXR5 IFNpdGUgLSBUaGUgVWx0aW1hdGUgQ2VsZWJyaXR5IENvbGxlY3Rpb24gPC9mb250PjwvZm9u dD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PGZv bnQgY29sb3I9IiNGRjRGQTciPjxmb250IHNpemU9NT4gPC9mb250PjwvZm9udD48Zm9udCBj b2xvcj0iIzAwODBGRiI+PGZvbnQgc2l6ZT01PjxzdHJvbmc+VU5MSU1JVEVEIEZSRUUgQUND RVNTIFRPIE9VUiBYWFggUkFURUQgDQpDSEFUIFJPT008L3N0cm9uZz48L2ZvbnQ+PC9mb250 PjwvcD4NCjxwIGFsaWduPWNlbnRlcj48Zm9udCBGQUNFPSJDb21pYyBTYW5zIE1TIj48Zm9u dCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT01PjxzdHJvbmc+R2V0IFlvdXIgQURVTFRT SUdIVFMgQUdFIFZFUklGSUNBVElPTiANCk1FTUJFUlNISVAgRk9SIE9OTFkgJDEyLjk1LiBJ dHMgR29vZCBGb3IgQW4gDQpFbnRpcmUgWWVhciBBbmQgR2V0cyBZb3UgVU5MSU1JVEVEIEZS RUUgQWNjZXNzIA0KSW50byAxODAwKyBPZiBUaGUgQmVzdCBBZHVsdCBTaXRlcyBPbiBUaGUg d2ViIEFsbCBZZWFyIA0KQWxsIEZyZWUuIE5ldmVyIFBheSBBbm90aGVyIERpbWUgVG8gR2V0 IEludG8gQSBTZXggU2l0ZSANCkFnYWluLjwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PGZvbnQg Y29sb3I9IiNGRkZGMDAiPjxmb250IHNpemU9NT48YnI+DQo8L2ZvbnQ+PC9mb250PjwvZm9u dD48Zm9udCBjb2xvcj0iI0ZGMDA4MCI+PGZvbnQgc2l6ZT03PjxzdHJvbmc+Qk9OVVMhISE8 YnI+DQo8L3N0cm9uZz48L2ZvbnQ+PGZvbnQgc2l6ZT01PjxzdHJvbmc+RlJFRSBYWFggQ2hh dCBSb29tcyEhITwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PC9wPg0KPHAgYWxpZ249Y2VudGVy PjxhIGhyZWY9Imh0dHA6Ly8yMDUuMjE2LjI0OS4xOC9jZ2ktc2hsL2RibWwuZXhlP3RlbXBs YXRlPS9hcy9zaWdudXAuZGJtJmFtcDtzaXRlPXN0MzUzMjUiPjxmb250IGNvbG9yPSIjODAw MDgwIj48Zm9udCBzaXplPTY+PHN0cm9uZz5JIFdBTlQgVEhFIFBBU1NXT1JEIE5PVzwvc3Ry b25nPjwvZm9udD48L2ZvbnQ+PC9hPjwvcD4NCjwvYm9keT4NCg0KPC9odG1sPg0K --_-=195
Sender: h1ouu88d77@compuserve.com Control: cancel <5qr1d0$d0s@news1-alterdial.uu.net> Message-ID: <cancel.5qr1d0$d0s@news1-alterdial.uu.net> Subject: cmsg cancel <5qr1d0$d0s@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 19 Jul 1997 18:35:15 GMT EMP article removed by jem@xpat.com. Original Headers: From: h1ouu88d77@compuserve.com Subject: Why pay for sex sites just.... Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!news-peer.sprintlink.net!news.sprintlink.net!Sprint!howland.erols.net!infeed2.internetmci.com!newsfeed.internetmci.com!news-feed1.tiac.net!uunet!not-for-mail Lines: 99
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: Sat, 19 Jul 1997 15:15:14 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <AnoF92a00UhBE1ymgp@andrew.cmu.edu> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> <5qlirj$ojm$1@netty.york.ac.uk> <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> <5qp95m$5du$1@vader.wolfware.ipc.net> In-Reply-To: <5qp95m$5du$1@vader.wolfware.ipc.net> Excerpts from netnews.comp.sys.next.programmer: 19-Jul-97 Re: uname by Christopher Wolf@wolfwar >>Speaking of which, there has been a complete lack of response to my >> earlier comments about POSIX compliance for Rhapsody. Has anyone heard >> anything? Hopefully, there do exist people who care whether Rhapsody is >> POSIX-compliant...? > > At WWDC it was made pretty clear that Rhapsody would be POSIX compliant. That's excellent news; thanks. > The BSD layer is 4.4 which I believe gives them this "for free" anyways. Yes and no. The problem was that you couldn't combine NEXTSTEP Obj-C classes (like stuff from the AppKit, etc) with POSIX, presumably because various Obj-C classes had dependancies on the BSD system call semantics and broke when working under a POSIX syscall environment. They should have fixed this moving to OpenStep, but, if they haven't, they need to do so. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: luomat@peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: 19 Jul 1997 20:27:01 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Message-ID: <5qr7ul$kab$1@ha2.rdc1.nj.home.com> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> <5qlirj$ojm$1@netty.york.ac.uk> <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> <5qp95m$5du$1@vader.wolfware.ipc.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: cwolf@wolfware.com In <5qp95m$5du$1@vader.wolfware.ipc.net> Christopher Wolf wrote: > On 07/18/97, Charles William Swiger wrote: > > > >Speaking of which, there has been a complete lack of response to my > >earlier comments about POSIX compliance for Rhapsody. Has anyone heard > >anything? Hopefully, there do exist people who care whether Rhapsody is > >POSIX-compliant...? > > At WWDC it was made pretty clear that Rhapsody would be POSIX compliant. The > BSD layer is 4.4 which I believe gives them this "for free" anyways. That's very good news..... 4.4 also gets rid of the 2gig limit for disks, correct? (the new limit is 4gig?) TjL ps -- Hebrew started as of 2 July, and ends on 22 August, so please understand if responses are slow. I am taking a 2 semester class in 8 weeks for 6 credits -- TjL <luomat@peak.org> f ` cn rd ths ` my hv lrnd hbrw
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer Subject: autorelease/release question Date: 19 Jul 97 17:02:53 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFF6A3C3-575AE@141.214.134.235> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit If I send a release or autorelease message to a dictionary, array, set, or other store object, will the object remove all the objects it has stored in it (thereby sending release messages), or do I have to first tell the store object to remove everything? (this would be to avoid memory leaks) thanks, rob -- <mailto: "Robert A. Decker" comrade@umich.edu> Listen to my Realaudio playlist:<http://hmrl.cancer.med.umich.edu/Rob/index.ssi> Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #10: "Wow. I'm a genius too. I think. BEEP." -Chris Elliott
From: mpaque.spa-am@nospam.wco.com (Mike Paquette) Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: 19 Jul 1997 16:50:33 -0700 Organization: Electronics Service Unit No. 16 Sender: mpaque@mpaque Distribution: world Message-ID: <5qrjs9$ha@mpaque.mpaque> References: <5qr7ul$kab$1@ha2.rdc1.nj.home.com> In article <5qr7ul$kab$1@ha2.rdc1.nj.home.com> luomat@peak.org (Timothy J. Luoma) writes: > That's very good news..... 4.4 also gets rid of the 2gig limit for disks, > correct? (the new limit is 4gig?) The disk partition limit under BSD4.4 is up in the terabyte range, so this shouldn't be a limiting factor in the personal computing market for at least a year... -- Mike Paquette (mpaque AT wco.com ; Yank that .spa-am and nospam to reply.) Well, if there *were* anything to say, it would be with the understanding that the PR/Marketing people want to make the announcements on products, so anything I have to say wouldn't actually exist until after then, so what I might have to say now doesn't exist, and what I may say in future can't be said, so theoretically what exists, doesn't, for the immediate future. (With apologies to Joe Straczynski)
Sender: Abdul Safar<ASE1000@1stfamily.com> Control: cancel <7137cd$1740.16@NEWS> Message-ID: <cancel.7137cd$1740.16@NEWS> Subject: cmsg cancel <7137cd$1740.16@NEWS> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 20 Jul 1997 03:53:05 GMT EMP article removed by jem@xpat.com. Original Headers: From: Abdul Safar<ASE1000@1stfamily.com> Subject: Free Internet Access World Wide Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!NEWS!not-for-mail Lines: 50
From: sanguish@digifix.com Newsgroups: comp.sys.next.advocacy,comp.sys.next.announce,comp.sys.next.hardware,comp.sys.next.software,comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.bugs,comp.sys.next.programmer Subject: NEXTSTEP/OpenStep Resources on the Net Supersedes: <1452867556825@digifix.com> Date: 20 Jul 1997 03:54:17 GMT Organization: Digital Fix Development Message-ID: <11913869371228@digifix.com> Topics include: Major OpenStep/NEXTSTEP World Wide Web Sites OpenStep/NEXTSTEP/Rhapsody Related Usenet Newsgroups Major OpenStep/NEXTSTEP FTP sites NeXTanswers Major OpenStep/NEXTSTEP World Wide Web Sites ============================================ The following sites are a sample of the OpenStep related WWW sites available. A comprehensive list is available on Stepwise. Stepwise OpenStep/Rhapsody Information Server http://www.stepwise.com Stepwise has been serving the OpenStep/NEXTSTEP community since March 1993. Some of the many resources on the site include: OpenStep Third Party Software guide, Developer Directory, Mailing List information, extensive listing of FTP and WWW sites related to OpenStep and NEXTSTEP, OpenStep related Frequently Asked Questions. NeXT Software Archives @ Peak.org http://www.peak.org/next http://www.peak.org/openstep PEAK is the premier NeXTStep/OpenStep FTP site in North America. NeXT Software Archives @ Peak.org http://www.peak.org/next http://www.peak.org/openstep PEAK is the premier NeXTStep/OpenStep FTP site in North America. This is the World Wide Web interace to the FTP site. Apple Enterprise Software Group (formerly NeXT Computer, Inc.) http://www.next.com Here is where you'll find the NeXTanswers archive, with information on OpenStep installation, drivers and software patches. Apple Computer's 'Prelude to Rhapsody' Self Support Site http://devworld.apple.com/dev/prelude.html This site has been constructed to help you help yourself to learn as much as possible about the foundation for Rhapsody, today's OPENSTEP. The site provides an informal collection of pointers, references, and starting points for developers who are using the Prelude to Rhapsody bundle, distributed at this year's Worldwide Developer Conference. OpenStep/NEXTSTEP/Rhapsody Related Usenet Newsgroups ==================================================== COMP.SYS.NEXT.ADVOCACY This is the "why NEXTSTEP is better (or worse) than anything else in the known universe" forum. It was created specifically to divert lengthy flame wars from .misc. COMP.SYS.NEXT.ANNOUNCE Announcements of general interest to the NeXT community (new products, FTP submissions, user group meetings, commercial announcements etc.) This is a moderated newsgroup, meaning that you can't post to it directly. Submissions should be e-mailed to next-announce@digifix.com where the moderator (Scott Anguish) will screen them for suitability. Messages posted to announce should NOT be posted or crossposted to any other comp.sys.next groups. COMP.SYS.NEXT.BUGS A place to report verifiable bugs in NeXT-supplied software. Material e-mailed to Bug_NeXT@NeXT.COM is not published, so this is a place for the net community find out about problems when they're discovered. This newsgroup has a very poor signal/noise ratio--all too often bozos post stuff here that really belongs someplace else. It rarely makes sense to crosspost between this and other c.s.n.* newsgroups, but individual reports may be germane to certain non-NeXT-specific groups as well. COMP.SYS.NEXT.HARDWARE Discussions about NeXT-label hardware and compatible peripherals, and non-NeXT-produced hardware (e.g. Intel) that is compatible with NEXTSTEP. In most cases, questions about Intel hardware are better asked in comp.sys.ibm.pc.hardware. Questions about SCSI devices belong in comp.periphs.scsi. This isn't the place to buy or sell used NeXTs--that's what .marketplace is for. COMP.SYS.NEXT.MARKETPLACE NeXT stuff for sale/wanted. Material posted here must not be crossposted to any other c.s.n.* newsgroup, but may be crossposted to misc.forsale.computers.workstation or appropriate regional newsgroups. COMP.SYS.NEXT.MISC For stuff that doesn't fit anywhere else. Anything you post here by definition doesn't belong anywhere else in c.s.n.*--i.e. no crossposting!!! COMP.SYS.NEXT.PROGRAMMER Questions and discussions of interest to NEXTSTEP programmers. This is primarily a forum for advanced technical material. Generic UNIX questions belong elsewhere (comp.unix.questions), although specific questions about NeXT's implementation or porting issues are appropriate here. Note that there are several other more "horizontal" newsgroups (comp.lang.objective-c, comp.lang.postscript, comp.os.mach, comp.protocols.tcp-ip, etc.) that may also be of interest. COMP.SYS.NEXT.SOFTWARE This is a place to talk about [third party] software products that run on NEXTSTEP systems. COMP.SYS.NEXT.SYSADMIN Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. ** RELATED NEWSGROUPS ** COMP.SOFT-SYS.NEXTSTEP Like comp.sys.next.software and comp.sys.next.misc combined. Exists because NeXT is a software-only company now, and comp.soft-sys is for discussion of software systems with scope similar to NEXTSTEP. COMP.LANG.OBJECTIVE-C Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. COMP.OBJECT Technical talk about OOP in general. Lots of C++ discussion, but NeXT and Objective-C get quite a bit of attention. At times gets almost philosophical about objects, but then again OOP allows one to be a programmer/philosopher. (The original comp.sys.next no longer exists--do not attempt to post to it.) Exception to the crossposting restrictions: announcements of usenet RFDs or CFVs, when made by the news.announce.newgroups moderator, may be simultaneously crossposted to all c.s.n.* newsgroups. Getting the Newsgroups without getting News =========================================== Thanks to Michael Ross at antigone.com, the main NEXTSTEP groups are now available as a mailing list digest as well. next-nextstep next-advocacy next-announce next-bugs next-hardware next-marketplace next-misc next-programmer next-software next-sysadmin object lang-objective-c (For a full description, send mail to listserv@antigone.com). The subscription syntax is essentially the same as Majordomo's. To subscribe, send a message to *-request@lists.best.com saying: subscribe where * is the name of the list e.g. next-programmer-request@lists.best.com Major OpenStep/NEXTSTEP FTP sites ================================= ftp://ftp.next.peak.org The main site for North American submissions formerly ftp.cs.orst.edu ftp://ftp.peanuts.org: (Peanuts) Located in Germany. Comprehensive archive site. Very well maintained. ftp://ftp.nluug.nl/pub/comp/next NeGeN/NiNe (NEXTSTEP Gebruikers Nederland/NeXTSTEP in the Netherlands) ftp://cube.sm.dsi.unimi.it (Italian NEXTSTEP User Group) ftp://ftp.nmr.embl-heidelberg.de/pub/next eduStep ftp://ftp.next.com: See below ftp.next.com and NextAnswers@next.com ===================================== [from the document ftp://ftp.next.com/pub/NeXTanswers/1000_Help] Welcome to the NeXTanswers information retrieval system! This system allows you to request online technical documents, drivers, and other software, which are then sent to you automatically. You can request documents by fax or Internet electronic mail, read them on the world-wide web, transfer them by anonymous ftp, or download them from the BBS. NeXTanswers is an automated retrieval system. Requests sent to it are answered electronically, and are not read or handled by a human being. NeXTanswers does not answer your questions or forward your requests. USING NEXTANSWERS BY E-MAIL To use NeXTanswers by Internet e-mail, send requests to nextanswers@next.com. Files are sent as NeXTmail attachments by default; you can request they be sent as ASCII text files instead. To request a file, include that file's ID number in the Subject line or the body of the message. You can request several files in a single message. You can also include commands in the Subject line or the body of the message. These commands affect the way that files you request are sent: ASCII causes the requested files to be sent as ASCII text SPLIT splits large files into 95KB chunks, using the MIME Message/Partial specification REPLY-TO address sets the e-mail address NeXTanswers uses These commands return information about the NeXTanswers system: HELP returns this help file INDEX returns the list of all available files INDEX BY DATE returns the list of files, sorted newest to oldest SEARCH keywords lists all files that contain all the keywords you list (ignoring capitalization) For example, a message with the following Subject line requests three files: Subject: 2101 2234 1109 A message with this body requests the same three files be sent as ASCII text files: 2101 2234 1109 ascii This message requests two lists of files, one for each search: Subject: SEARCH Dell SCSI SEARCH NetInfo domain NeXTanswers will reply to the address in your From: line. To use a different address either set your Reply-To: line, or use the NeXTanswers command REPLY-TO If you have any problem with the system or suggestions for improvement, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY FAX To use NeXTanswers by fax, call (415) 780-3990 from a touch-tone phone and follow the instructions. You'll be asked for your fax number, a number to identify your fax (like your phone extension or office number), and the ID numbers of the files you want. You can also request a list of available files. When you finish entering the file numbers, end the call and the files will be faxed to you. If you have problems using this fax system, please call Technical Support at 1-800-848-6398. You cannot use the fax system outside the U.S & Canada. USING NEXTANSWERS VIA THE WORLD-WIDE WEB To use NeXTanswers via the Internet World-Wide Web connect to NeXT's web server at URL http://www.next.com. USING NEXTANSWERS BY ANONYMOUS FTP To use NeXTanswers by Internet anonymous FTP, connect to FTP.NEXT.COM and read the help file pub/NeXTanswers/README. If you have problems using this, please send mail to nextanswers-request@next.com. USING NEXTANSWERS BY MODEM To use NeXTanswers via modem call the NeXTanswers BBS at (415) 780-2965. Log in as the user "guest", and enter the Files section. From there you can download NeXTanswers documents. FOR MORE HELP... If you need technical support for NEXTSTEP beyond the information available from NeXTanswers, call the Support Hotline at 1-800-955-NeXT (outside the U.S. call +1-415-424-8500) to speak to a NEXTSTEP Technical Support Technician. If your site has a NeXT support contract, your site's support contact must make this call to the hotline. Otherwise, hotline support is on a pay-per-call basis. Thanks for using NeXTanswers! _________________________________________________________________ Written by: Eric P. Scott ( eps@toaster.SFSU.EDU ) and Scott Anguish ( sanguish@digifix.com ) Additions from: Greg Anderson ( Greg_Anderson@afs.com ) Michael Pizolato ( alf@epix.net ) Dan Grillo ( dan_grillo@next.com )
From: Landivar2 <landivar2@geocities.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.mac.programmer.games,comp.sys.mac.programmer.misc,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Amelio & Hancock Leave Apple Date: 20 Jul 1997 04:03:17 GMT Organization: gte.net Message-ID: <5qs2m5$q5b$2@gte1.gte.net> References: <*johnnyc*-1207971235030001@192.0.2.1> <AFF5F2E0-BD04@206.165.44.1> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lawson English wrote: > > > Yes, but not as a replacement. The current draw libraries could be > > incorporated into the system with only a few hundred k of code. Put the > > PC version on the net and tote it as the standard for lightweight > internet > > vector grphics. There is no single move that could be done to speed the > > internet than reduce graphics sizes by a factor of 10. > > Gee, and you could name it "Apple Internet Graphics..." > > It is no coincidence that 2 of the 3 vector libraries recently reviewed in > MacUser (MacWorld?) are based on GX. > > If you read what Foley and Van Dam have to say about immediate-mode > graphics vs graphics databases, something like GX would be *ideal* for > networked graphics, whereas something like DPS would be too slow for > Internet use. I think you both nailed it. GX graphics lib. for the internet and DPS for the desktop. Don't forget to include GX with Java. > > Not sure where Taligent Graphics fits in. >
Sender: 0987263268@compuserve.com Control: cancel <5qsuov$kpi@news1-alterdial.uu.net> Message-ID: <cancel.5qsuov$kpi@news1-alterdial.uu.net> Subject: cmsg cancel <5qsuov$kpi@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 20 Jul 1997 12:02:43 GMT EMP article removed by jem@xpat.com. Original Headers: From: 0987263268@compuserve.com Subject: FREE 2000 $ex-WebSites FREE! @? Newsgroups: comp.sys.next.programmer Path: ...!newsfeed.direct.ca!news-feed1.tiac.net!uunet!not-for-mail Lines: 99
From: "Frank Alviani" <alviani@ix.netcom.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Openstep examples OO design (was Re: Some initial OpenStep notes) Date: 20 Jul 97 08:47:51 -0500 Organization: Netcom Message-ID: <AFF7813B-2FDF4@205.186.163.34> References: <5qpdih$c6@saturn.genoa.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit nntp://nntp.netcruiser/comp.sys.next.programmer Just a quick question - is there another source for NextAnswer 1582? The server claims it doesn't exist.. Thanks in advance Frank Alviani
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer Subject: Re: autorelease/release question Date: 20 Jul 97 12:27:35 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFF7B4BF-3916EE@141.214.134.235> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: "Andrew Abernathy" <andrew@omnigroup.com>, "Thomas Engel" <tsengel@slivovic.nl.ctp.com> If you read the documentation on the array class cluster it says that "in general, objects that you add to an (array) aren't copied; rather, each object receives a retain message before its id is added to the array. When an object is removed from an array, it's sent a release message". That's where my confusion comes from. If I send a release message to an array (or other store object), I'll lose that object, but do all the objects that were in the array still have an extra retain? I guess from what's written above, it could be saying that I have to explicitely give the object the retain message before I add it to the array? That would seem to give me a lot of work as I would have to send the release message myself as well. I could just forget the retain messages alltogethor if the only place I keep the object is in the array... Do most people just ignore what's written above? rob -- <mailto: "Robert A. Decker" comrade@umich.edu> Listen to my Realaudio playlist:<http://hmrl.cancer.med.umich.edu/Rob/index.ssi> Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #10: "Wow. I'm a genius too. I think. BEEP." -Chris Elliott
From: luomat@peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: 20 Jul 1997 16:22:51 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Distribution: world Message-ID: <5qte0r$kab$3@ha2.rdc1.nj.home.com> References: <5qr7ul$kab$1@ha2.rdc1.nj.home.com> <5qrjs9$ha@mpaque.mpaque> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: mpaque.spa-am@nospam.wco.com In <5qrjs9$ha@mpaque.mpaque> Mike Paquette wrote: > > The disk partition limit under BSD4.4 is up in the terabyte range, so this > shouldn't be a limiting factor in the personal computing market for at > least a year... I think Micro$oft Office 2000 is scheduled to be .3 terabyte, but that won't change my life at all ;-) Thanks for the info.... TjL ps -- Hebrew started as of 2 July, and ends on 22 August, so please understand if responses are slow. I am taking a 2 semester class in 8 weeks for 6 credits -- TjL <luomat@peak.org> f ` cn rd ths ` my hv lrnd hbrw
From: kcd@babylon5.jumpgate.com (Kenneth C. Dyke) Newsgroups: comp.sys.next.programmer Subject: Re: autorelease/release question Date: 20 Jul 1997 18:09:52 GMT Message-ID: <5qtk9g$atq$1@nntp2.ba.best.com> References: <AFF7B4BF-3916EE@141.214.134.235> In-Reply-To: <AFF7B4BF-3916EE@141.214.134.235> On 07/20/97, "Robert A. Decker" wrote: > > If you read the documentation on the array class cluster it says that "in >general, objects that you add to an (array) aren't copied; rather, each >object receives a retain message before its id is added to the array. When >an object is removed from an array, it's sent a release message". Right, the documentation means that the retain and release messages are sent by the array object itself, not by the user of the array. > That's where my confusion comes from. If I send a release message to an >array (or other store object), I'll lose that object, but do all the >objects that were in the array still have an extra retain? That depends. To be precise, releasing an array will decrement it's retain count. If the retain count goes to 0, then the array will be deallocated. At that point, it will send a release to every object it contains. Any objects that were contained in the array but not retained by something else will be deallocated. > I guess from what's written above, it could be saying that I have to >explicitely give the object the retain message before I add it to the >array? That would seem to give me a lot of work as I would have to send the >release message myself as well. I could just forget the retain messages >alltogethor if the only place I keep the object is in the array... Do most >people just ignore what's written above? Nah, you just have to read it a bit differently. ;) -Ken -- Kenneth Dyke, kcd@jumpgate.com (personal), kdyke@ea.com (work) Nuclear Strike and OPENSTEP Tools Engineer, Electronic Arts C++: The power, elegance and simplicity of a hand grenade.
From: stevehix@DeleteToReplysafemail.com (Steve Hix) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Fri, 18 Jul 1997 21:56:32 -0700 Organization: South Valley Internet Message-ID: <stevehix-1807972156320001@ip22.safemail.com> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <jpalmer-1407971726250001@jpalmer.vip.best.com> <16819017@NEWS.SAIC.COM> <5qh6rh$og7$1@news.xmission.com> <jpalmer-1507972144050001@jpalmer.vip.best.com> <17035036@NEWS.SAIC.COM> <5qjib4$7mr$1@ironhorse.plsys.co.uk> <jpalmer-1607971639200001@jpalmer.vip.best.com> <5qjvgu$hlq@sjx-ixn7.ix.netcom.com> <stevehix-1707971021580001@ip36.safemail.com> <5qpe78$sd9@sjx-ixn8.ix.netcom.com> In article <5qpe78$sd9@sjx-ixn8.ix.netcom.com>, Jonathan Hendry <jon@subsequent.com> wrote: > Steve Hix <stevehix@DeleteToReplysafemail.com> wrote: > > > Why order it? Be is providing the CD free (in the U.S., along > > with the next issue of MacUser). Check the Be webpage. > > > Should be fun to play with, if you have one of the machines > > that BeOS supports. > > Does it boot from the CD? Do you have to reformat your > harddrive to install it? IIRC, you install it on your hard disk, or at least a separate partition of (one) of your disk(s). They did specifically mention dual-boot volumes.
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: autorelease/release question Date: 20 Jul 1997 18:13:57 GMT Organization: WolfWare Message-ID: <5qtkh5$6uo$1@vader.wolfware.ipc.net> References: <AFF7B4BF-3916EE@141.214.134.235> In-Reply-To: <AFF7B4BF-3916EE@141.214.134.235> On 07/20/97, "Robert A. Decker" wrote: > > If you read the documentation on the array class cluster it says that "in >general, objects that you add to an (array) aren't copied; rather, each >object receives a retain message before its id is added to the array. When >an object is removed from an array, it's sent a release message". > > That's where my confusion comes from. If I send a release message to an >array (or other store object), I'll lose that object, but do all the >objects that were in the array still have an extra retain? > When you add an object to an array it's retain count get's incremented by one. When you remove an object from an array it's retain count is decremented by one. When the array is deallocated (this is different from "when I send a release message to an array") all objects in the array receive a release message. > I guess from what's written above, it could be saying that I have to >explicitely give the object the retain message before I add it to the >array? Only if you want the object to persist after it is removed from the array. In many (most?) cases you want the objects in the array to have the same lifetime as the array and so there is no need to explictly send an additional retain to the objects in the array. >That would seem to give me a lot of work as I would have to send the >release message myself as well. I could just forget the retain messages >alltogethor if the only place I keep the object is in the array... As long as the object is part of the array the array will have a retain on the object and it won't go away. The only time you would need to explicitly retain an object in the array is if you are planning on deallocing the array but still need to maintain references to an object that was in it. It's too early in the morning - hope this helped and didn't confuse the issue more. -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: Peter Kulik <pkulik@klci.com> Newsgroups: comp.sys.next.programmer Subject: New White Paper Date: Sun, 20 Jul 1997 18:00:08 -0400 Organization: Kulik & Lazarus Consulting, Inc. Message-ID: <33D28A68.5D93@klci.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit New White Paper - A Hybrid Process for the New Reality of Software Development. If you are involved with software projects including Java, Internet/Intranets, Object Oriented methods, Client-Server, etc., you should find this paper interesting. Get it in Kulik & Lazarus Consulting's InfoBase at: http://www.klci.com/accelerate Please let us know what you think! Peter Kulik pkulik@klci.com
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: Re: autorelease/release question Date: 20 Jul 1997 22:06:50 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5qu25q$a5m$1@concorde.ctp.com> References: <AFF7B4BF-3916EE@141.214.134.235> (While I already replied in a mail I still want to add some points publicly) "Robert A. Decker" <comrade@umich.edu> wrote: > > If you read the documentation on the array class cluster it says that "in >general, objects that you add to an (array) aren't copied; rather, each >object receives a retain message before its id is added to the array. When >an object is removed from an array, it's sent a release message". > I get the feeling that most C and C++ programmer who get the the Openstep autorelease/retain world stumble accorss the same problems because they try to read too much into the docu. Seems like they expect evert methdo to have its custom behavior or special conventions. Being on a project for the last 4 weeks it really seems like most novice develoeprs don't accept the quite simple policy: "You dan't have to release what you did not retain/alloc/copy....and you must release that you did retain/alloc/copy". Now there are some "nifty" problems which go along with Openstep's autorelease/retain policy...bbut those are not the ones which people stumble acrross all the time. The souces of these guys are filled with APIs comments like ("The receiver has to release that..." or "The receiver should never release that object" etc.pp) From these comments it is easy to spot that they simply did not get the base idea. > That's where my confusion comes from. If I send a release message to an >array (or other store object), I'll lose that object, but do all the >objects that were in the array still have an extra retain? > No...unless you did something wrong before release the array. e.g. aString = [[NSString alloc] init]; [myArray add String:aString]; this is "wrong" since you forgot to do: [aString release]; since you actually don't care about the string...you care about the array which holds he string from now one...so you code no longer needs the string and therefore _must_ release it, since it did create it (alloc/init). > I guess from what's written above, it could be saying that I have to >explicitely give the object the retain message before I add it to the >array? No...you are only interested in keeping the array around....and the array is interested in keeping its objects around. This reflects the way in which retain/release should be handled (and is by the Array..and should be by your code) >That would seem to give me a lot of work as I would have to send the >release message myself as well. even if you had (attention: you _must_ not release them by hand !!) there is the beauty of ObjC [myArrar makeObjectPerfor:@selector(release)]; and you are done (or [myArrar releaseObjects] is a similar shortcut in this situation.) >I could just forget the retain messages >alltogethor if the only place I keep the object is in the array... Do most >people just ignore what's written above? Nope..they don't ignore it (written above means the quote form the NSArray doc ?) They jsut do the right thing which is to pass in an object which is release after adding it to the array (see example above). Nowever...if you mehtod receive the object as an arument..you usually just add it to the array..that's it. "You did not create it..you don't have to release it" E.g.: - (void)myMethdo:aString { [myArray addObject:aString]; } thats it... no retain..no release no nothing needed Ok..we know that there are some gotches and strange side effects and setups where we should act differently...but thats another story. e.g. - (void)myRemoveMethdo:aString { [myArray removeObject:aString]; NSLog( @"Its no longer there %@", aString ); } would most likely crash. Becasu eyou don't have a reference to the string..and propably the "removeObejct" would send the final release thereby causing the object to get deallocated which causes teh NSLog to fail. In order to ensure that aString is still around when you get to NSLog you must "claim" interest in the string to be still availble for your method. Which means: - (void)myRemoveMethdo:aString { [aString retain]; [myArray removeObject:aString]; NSLog( @"Its no longer there %@", aString ); [aString release]; // Now it can really go away... } How these are the "tricky" sideeffects which can catch you every now and then (and there are more tricky issues with circular references and such) but before you worry about them..try to get to the essence behine the autorelase/retain policy. Hope this helped someone. Aloha Tomi
Newsgroups: comp.sys.mac.system,comp.sys.mac.wanted,comp.sys.newton.misc,comp.sys.newton.programmer,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.palmtops,comp.sys.pen,comp.sys.pen,comp.sys.powerpc.advocacy,comp.sys.psion.apps,comp.sys.psion.marketplace,comp.sys.psion.misc,comp.sys.psion.programmer,comp.sys.sgi.admin,comp.sys.sgi.graphics,comp.sys.sgi.hardware,comp.sys.sgi.misc,comp.sys.sinclair,comp.sys.sun.admin,ct.jobs,cz.soc.mensa,dc.forsale.computers,dc.forsale.misc,dc.general,dc.jobs,control From: news@news.msfc.nasa.gov Message-ID: <cancel.33D2948F.82D5A445@sprintmail.com> Control: cancel <33D2948F.82D5A445@sprintmail.com> Subject: cmsg cancel <33D2948F.82D5A445@sprintmail.com> no reply ignore Organization: Semi-Automatic Lupine Remover Date: Sun, 20 Jul 1997 23:53:40 GMT Sender: "William S. Taylor" <wtaylor001@sprintmail.com> ignore Make Money Fast post canceled by J. Porter Clark.
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: Peter Kulik <pkulik@klci.com> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33D28A68.5D93@klci.com> Control: cancel <33D28A68.5D93@klci.com> Date: 20 Jul 1997 18:39:23 GMT Message-ID: <cancel.33D28A68.5D93@klci.com> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: Timothy Luoma <luomat@peak.org> Newsgroups: comp.sys.next.sysadmin,comp.sys.next.programmer Subject: Re: NextPrinter Followup-To: comp.sys.next.programmer Date: Sun, 20 Jul 1997 10:17:10 -0700 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.SUN.3.96.970720100051.18512A-100000@kira> References: <5qlre3$b2a@lyra.csx.cam.ac.uk> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII To: teb@eng.cam.ac.uk In-Reply-To: <5qlre3$b2a@lyra.csx.cam.ac.uk> Followups-set to comp.sys.next.programmer, where hopefully someone will have a workaround. On 17 Jul 1997 teb@eng.cam.ac.uk wrote: > has anybody had this message before when printing (from > Openwrite)? Any recommendations? Thanx. > > next syslog: /usr/lib/NextPrinter/npd-lpr: possible buffer overrun > attack by uid 20 If you are using 4.2 it has the lpr buffer overrun patch on it. If you are using >4.2 then you probably are running into the lpr-wrapper patch. OpenWrite (and WordPerfect) both seem to do this, but I do not know why. I had to remove the patch on my system for the time being. Eventually I'll have some time to muck around with the patch and figure out what they are trying to send. The patch can be found in the 'new_arrivals' folder or ftp://next-ftp.peak.org/pub/next/apps/utils/unix//lpr-wrapper.1.0.1.NIHS.bs.tar.gz the original source code is at ftp://ftp.auscert.org.au/pub/auscert/tools/overflow_wrapper/overflow_wrapper.c and it was compiled this way: cc -arch m68k -arch i386 -arch hppa -arch sparc \ -DREAL_PROG='"/usr/ucb/lpr.orig"' -DMAXARGLEN=32 -DSYSLOG -o lpr lpr.c in case someone else wants to try and debug it. Increasing -DMAXARGLEN=32 has got to be the solution, but I don't know what the correct value is. Perhaps if the program could say what the "ARGLEN" was that OpenWrite tried to send. Could this be OpenWrite/WP sending the file name which would would easily be longer than 32 characters? TjL
From: luomat@peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.programmer Subject: Re: Openstep examples OO design (was Re: Some initial OpenStep Date: 21 Jul 1997 02:51:58 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Message-ID: <5quise$3n5$2@ha2.rdc1.nj.home.com> References: <5qpdih$c6@saturn.genoa.com> <AFF7813B-2FDF4@205.186.163.34> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: alviani@ix.netcom.com In <AFF7813B-2FDF4@205.186.163.34> "Frank Alviani" wrote: > Just a quick question - is there another source for NextAnswer 1582? The > server claims it doesn't exist.. I was able to get it from http://www.next.com/NeXTanswers/HTMLFiles/1582.htmld/1582.html There's also ftp://ftp.next.com/pub/NeXTanswers/ByNumber/1582.compressed TjL ps -- Hebrew started as of 2 July, and ends on 22 August, so please understand if responses are slow. I am taking a 2 semester class in 8 weeks for 6 credits -- TjL <luomat@peak.org> f ` cn rd ths ` my hv lrnd hbrw
From: freeman@cornell-iowa.edu Newsgroups: comp.sys.next.programmer Subject: Posix: 3.3: tempnam Date: 21 Jul 1997 03:27:14 GMT Organization: McleodUSA - http://www.mcleodusa.net Message-ID: <5qukui$ffm1@odie.mcleod.net> I am trying to compile the php module for apache. Unfortunately it uses some POSIX compliant functions. I have work arounds for putenv, getcwd. I do not have anything that will work for tempnam. Does anyone have some code to define tempnam in terms of mkstemp that will work on 3.3. If you do, would you be willing to share it. Thanks. Jim Freeman --- Dr. James H.Freeman freeman@Cornell-Iowa.edu (Nextmail OK) Department of Mathematics (Mime OK) Cornell College 319-895-4393(office) 600 First Street West 319-895-6866(home) Mount Vernon, Iowa 52314-1098
From: Garance A Drosehn <gad@eclipse.its.rpi.edu> Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: 21 Jul 1997 03:41:07 GMT Organization: Rensselaer Polytechnic Institute, Troy NY, USA Message-ID: <5quloj$cq8@usenet.rpi.edu> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> <5qlirj$ojm$1@netty.york.ac.uk> <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> <5qp95m$5du$1@vader.wolfware.ipc.net> <AnoF92a00UhBE1ymgp@andrew.cmu.edu> Charles William Swiger <cs4w+@andrew.cmu.edu> wrote: > Excerpts from netnews.comp.sys.next.programmer: 19-Jul-97 > Re: uname by Christopher Wolf@wolfwar > > The BSD layer is 4.4 which I believe gives them this "for free" > > anyways. > > Yes and no. The problem was that you couldn't combine NEXTSTEP > Obj-C classes (like stuff from the AppKit, etc) with POSIX, > presumably because various Obj-C classes had dependancies on the > BSD system call semantics and broke when working under a POSIX > syscall environment. Well, there was also a problem with just plain posix, in unix programs that weren't doing anything with Objective-C. I do agree with you that POSIX should work with Objective-C, I won't say that was the only problem with POSIX under earlier versions of NeXTSTEP. --- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu Senior Systems Programmer (MIME & NeXTmail capable) Rensselaer Polytechnic Institute; Troy NY USA
From: leigh@cs.uwa.edu.au (Leigh Smith) Newsgroups: comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.hardware Subject: Re: HELP with SCSI device problem Date: 21 Jul 1997 04:20:42 GMT Organization: The University of Western Australia Message-ID: <5quo2q$epf$1@enyo.uwa.edu.au> References: <5qpgpt$sgs$2@excalibur.flash.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: daj@nwu.edu In <5qpgpt$sgs$2@excalibur.flash.net> daj@nwu.edu wrote: > I have a SCSI data acquisition device that was manufactured by Analogic, its model DASM-AD14. Technically it is a 14bit 2Mhz 8 channel acquisition device that has a 2Mb buffer and a SCSI interface. > [snip] > So to make a long story shorter than what it could be, is the above note true or is there some patch I can apply to allow me to write a device driver under Openstep 4.1? Can I get NEXTSTEP3.3 Developer and use it under OPENSTEP 4.1? Last, I'm still not clear in understanding why I can't just use /dev/rsd4c as the Analogic manual suggests. The requirement as Analogic has it in there manual is that their DASM-AD14 communicates through a character-special file (ie /dev/rsd4c). Clearly the device appears unwritable (in messages below), however it should be writable. Can anyone explain why I see this problem? > Ok, you have a big task ahead of you. The DriverKit is being overhauled so yes you will need 3.3 to develop on. You may be able to get the 3.3 libraries to work on 4.X, others have managed to get the 3.3 AppKit working, but I just settle for a machine I leave on 3.3 for device driver development. You will need to write an indirect device driver that uses the SCSI Driver to communicate with your device. Look at the SCSITape driver example which was on 3.3 but seems to have disappeared off 4.2. Using the indirect device driver you will be able to set up a character device that will appear in /dev. The device naming will only be specific to the Sun (I doubt the naming would be passed down to the DASM device) > Oh yeah, the last note is that Analogic no longer supports the device, it appeared to be to costly to produce so there wasnt enough demand. As a result I have virttually no help from Analogic. > I'd reconsider doing the port then. Device driver is not for the weak of heart and if you don't have the company supporting you, it could literally be impossible. Good luck -- Leigh Computer Science, University of Western Australia Smith +61-9-9380-3778 leigh@cs.uwa.edu.au (NeXTMail/MIME) "Home pages are the pet rock of the 90s. We all have them, we all think they're very cute. But in a few years we're going to look back and be pretty embarrassed." -- Tony Shepps <toad@pond.com> "Why wait?" -- Peter Langston
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: autorelease/release question Date: 21 Jul 1997 08:10:33 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5qv5hp$4r8$1@brachio.zrz.TU-Berlin.DE> References: <AFF7B4BF-3916EE@141.214.134.235> In article <AFF7B4BF-3916EE@141.214.134.235> "Robert A. Decker" <comrade@umich.edu> writes: [ Does an NSArray manage its contents ? ] Yes! Like just about any object, NSArrays properly manage their contents, that is, all objects referenced (contained) by the array are properly retained by it. This also means that the NSArray will release all the objects it contains when it goes away. No! It does not manage its contents' _memory_, just one point of reference counting. Therefore, deallocating an array will only deallocate those objects that don't have outstanding references. Marcel
Date: Mon, 21 Jul 1997 06:04:59 -0600 From: ajmas@bigfoot.com Subject: [Q] WDEFs & Rhapsody? Newsgroups: comp.sys.mac.programmer.help,comp.sys.next.programmer Message-ID: <869482436.23064@dejanews.com> Organization: Deja News Posting Service I have just taken a look at the WDEF definition in Inside Mac and thought how can a programmer under Rhapsody create their own custom window? If anyone knows how this will be done or is currently achieved under OpenStep then I would appreciate this information very much. Also, if there are any examples of doing this, then I would like to see them. Thanks - Please CC me your reply. Andre -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: steph&lise@usa.net (Stephanie&Lise) Newsgroups: comp.sys.next.marketplace,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.northstar,comp.sys.novell,comp.sys.nsc.32k,comp.sys.oric,comp.sys.os.linux.admin Subject: cmsg cancel <33d286d9.17126164@ntnews.entech.com> Control: cancel <33d286d9.17126164@ntnews.entech.com> Date: 21 Jul 1997 07:45:42 GMT Message-ID: <cancel.33d286d9.17126164@ntnews.entech.com> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: ians@cam-ani.co.uk (Ian Stephenson) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] WDEFs & Rhapsody? Date: Mon, 21 Jul 1997 13:21:13 GMT Organization: Cambridge Animation Systems Ltd Sender: news@cam-ani.co.uk Message-ID: <EDo7rE.9EC@cam-ani.co.uk> References: <869482436.23064@dejanews.com> In article <869482436.23064@dejanews.com> ajmas@bigfoot.com writes: > I have just taken a look at the WDEF definition in Inside Mac and > thought how can a programmer under Rhapsody create their own custom > window? erm? You might be better reading the OpenStep docs? > If anyone knows how this will be done or is currently achieved > under OpenStep then I would appreciate this information very much. Also, > if there are any examples of doing this, then I would like to see them. By and large you don't. Windows are containers for things, so as a rule you don't create custom window classes. Windows hold views, and these are generally subclasses to provide the UI. The window generall stays the same in all apps. You can modify the windows behaviour via a delegate (OK! this is 3.3) - Immediatly before and after doing (just about) anything signifigant, a window sends a message to it's delegate warning it, and allowing it to reject or modify the behaviour (say pop up a save/cancel box). This is slightly different under OpenStep which uses a noticfication system, but the effect is the same. If you really did want to radically change a window, you subclass it. This is generally not a good idea, and interface builder stops being helpfull, but it allows you complete control over the windows behaviour. $an
From: bchin@NOSPAM.richmond.freedomnet.com (Bill Chin) Newsgroups: comp.sys.next.programmer Subject: Re: LISP on Next Date: 21 Jul 1997 14:49:04 GMT Organization: FreedomNet Message-ID: <5qvst0$29r$1@news.freedomnet.com> References: <33CBC2F1.D1EEA930@cris.com> krispos@cris.com wrote: >I remember reading somewhere that the NeXT cubes had a built in LISP >compiler. However apparently the color slabs do not, and I was wondering >if anyone knew of any LISP compilers/interpreters for the NeXT color >system (Mach 3.2). I don't currently own a NeXT, but am thinking about >getting one. However if I can't use it to program than its usefullness >would decrease dramatically. Thanks in advance for the help. Once upon a time, academic NeXT machines came with Mathematica, Allegro Common Lisp, and WriteNow among other things. These three have subsequently been dropped from the standard distribution. Mathematica is still for sale for NEXTSTEP, WriteNow died, but I do not believe Allegro Common Lisp is still available. Of course, Rhapsody will probably change things but not in the near future. There are, however, many varieties of Lisp that are freely available and can be compiled for NEXTSTEP/OPENSTEP. I personally used XLISP-Plus. Many distributions can be embedded and called through a C language API, so they can be integrated into an Objective-C NEXTSTEP/OPENSTEP application. -- Bill Chin - bchin@richmond.freedomnet.com - NeXTmail/MIME welcomed
From: j_hendry@ix.netcom.com (Jonathan Hendry) Newsgroups: comp.sys.mac.programmer.help,comp.sys.next.programmer Subject: Re: [Q] WDEFs & Rhapsody? Date: 21 Jul 1997 17:13:05 GMT Organization: Netcom Message-ID: <5r05b1$p3q@sjx-ixn5.ix.netcom.com> References: <869482436.23064@dejanews.com> ajmas@bigfoot.com wrote: >I have just taken a look at the WDEF definition in Inside Mac and >thought how can a programmer under Rhapsody create their own custom >window? If anyone knows how this will be done or is currently achieved >under OpenStep then I would appreciate this information very much. Also, >if there are any examples of doing this, then I would like to see them. > >Thanks - Please CC me your reply. (Trying to remember Mac programming...<thinking> <thinking> I'm not entirely sure what sort of customization you wish to do. There are several ways to create customized windows. If you mean, setting up a window 'resource' that contains various subviews (text fields, etc.), this is done using InterfaceBuilder. You create a window object (which in InterfaceBuilder is a real Window instance), and drop UI elements onto it. You can set various characteristics of the window (closable, resizeable, etc) and connect it to other objects (set the window's Delegate object, etc.) The window is then 'freeze-dried' into a 'nib' file which is later dynamically loaded by your application. Or, by someone else's application. If you mean, creating windows that are substantially different from the standard NSWindow class, there are two ways. One is to create a subclass of NSWindow. In the subclass you can override whatever you need in order to create a new Window. One real-life example: Jason Beaver wrote an app whose windows are very different from the normal windows; the window's frame and title bar are colored instead of gray, the title bar is half the height of the standard bar, the title is double-click editable, and dropping a color swatch on the title bar changes the window's color. Another way to create different kinds of windows is to use Postscript windows instead of Appkit windows. Check the Adobe Purple book for more information. If I'm not mistaken, going down to the Postscript level lets you do things like creating a window shaped like a character from a Font (or, by extension, any postscript path). (Is this correct, anyone? I may be mis-remembering the nifty thing Mike P. posted.) - Jonathan
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: Mon, 21 Jul 1997 13:14:46 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <AnotY6m00WB=0rlZc0@andrew.cmu.edu> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> <5qlirj$ojm$1@netty.york.ac.uk> <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> <5qp95m$5du$1@vader.wolfware.ipc.net> <AnoF92a00UhBE1ymgp@andrew.cmu.edu> <5qtkp8$71m$1@vader.wolfware.ipc.net> In-Reply-To: <5qtkp8$71m$1@vader.wolfware.ipc.net> Excerpts from netnews.comp.sys.next.programmer: 20-Jul-97 Re: uname by Christopher Wolf@wolfwar >> Yes and no. The problem was that you couldn't combine NEXTSTEP Obj-C >> classes (like stuff from the AppKit, etc) with POSIX, presumably because >> various Obj-C classes had dependancies on the BSD system call semantics >> and broke when working under a POSIX syscall environment. > > Ah that's interesting and something I hadn't realized. Yup. Unfortunate, because it made the POSIX environment a seriously second-class citizen in terms of support and testing, compared to the effort NeXT spent making sure the non-POSIX (Obj-C, AppKit) environment worked fine. > The main reason I've had for wishing for POSIX compliance is to help with > compiling generic off-the-shelf Unix source code which doesn't use th > NeXT classes at all anyways. Sure. I could live with that too, but there is no good reason for there to be a special POSIX environment under Rhapsody. They should simply offer one, POSIX-compliant, bug-free environment which works with OpenStep-specific code and with generic Unix code. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: nospam@all.pls (Timothy J. Luoma) Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: 21 Jul 1997 21:57:32 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Message-ID: <5r0m0c$h72$1@ha2.rdc1.nj.home.com> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> <5qlirj$ojm$1@netty.york.ac.uk> <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> <5qp95m$5du$1@vader.wolfware.ipc.net> <AnoF92a00UhBE1ymgp@andrew.cmu.edu> <5qtkp8$71m$1@vader.wolfware.ipc.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: cwolf@wolfware.com In <5qtkp8$71m$1@vader.wolfware.ipc.net> Christopher Wolf wrote: > On 07/19/97, Charles William Swiger wrote: > >Excerpts from netnews.comp.sys.next.programmer: 19-Jul-97 Re: uname by > >Christopher Wolf@wolfware.com > > >Yes and no. The problem was that you couldn't combine NEXTSTEP Obj-C > >classes (like stuff from the AppKit, etc) with POSIX, presumably because > >various Obj-C classes had dependancies on the BSD system call semantics > >and broke when working under a POSIX syscall environment. > > Ah that's interesting and something I hadn't realized. The main reason I've had > for wishing for POSIX compliance is to help with compiling generic off-the-shelf > Unix source code which doesn't use the NeXT classes at all anyways. So are you saying that Rhapsody with POSIX will be easier to compile off-the-shelf generic UNIX source code, or that there really won't be much difference? TjL ps -- Hebrew started as of 2 July, and ends on 22 August, so please understand if responses are slow. I am taking a 2 semester class in 8 weeks for 6 credits -- TjL <luomat@peak.org> f ` cn rd ths ` my hv lrnd hbrw
Sender: 0987263268@compuserve.com Control: cancel <5r0nf1$3vb@news1-alterdial.uu.net> Message-ID: <cancel.5r0nf1$3vb@news1-alterdial.uu.net> Subject: cmsg cancel <5r0nf1$3vb@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 21 Jul 1997 22:22:34 GMT EMP article removed by jem@xpat.com. Original Headers: From: 0987263268@compuserve.com Subject: Why Pay For Sex Site Entry? Newsgroups: comp.sys.next.programmer Path: ...!fnews.nuri.net!uunet!not-for-mail Lines: 99
Sender: 18q34757@compuserve.com Control: cancel <5r1hbc$1sf@news1-alterdial.uu.net> Message-ID: <cancel.5r1hbc$1sf@news1-alterdial.uu.net> Subject: cmsg cancel <5r1hbc$1sf@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 22 Jul 1997 05:44:17 GMT EMP article removed by jem@xpat.com. Original Headers: From: 18q34757@compuserve.com Subject: Get FREE SEX SITE..password is Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!uunet!not-for-mail Lines: 99
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: uname Date: 22 Jul 1997 02:35:28 GMT Organization: WolfWare Message-ID: <5r169g$2o1$1@vader.wolfware.ipc.net> References: <5qfvng$ki0@lyra.csx.cam.ac.uk> <5qlirj$ojm$1@netty.york.ac.uk> <0nnx_6q00UhWE1zf9e@andrew.cmu.edu> <5qp95m$5du$1@vader.wolfware.ipc.net> <AnoF92a00UhBE1ymgp@andrew.cmu.edu> <5qtkp8$71m$1@vader.wolfware.ipc.net> <5r0m0c$h72$1@ha2.rdc1.nj.home.com> In-Reply-To: <5r0m0c$h72$1@ha2.rdc1.nj.home.com> On 07/21/97, Timothy J. Luoma wrote: >In <5qtkp8$71m$1@vader.wolfware.ipc.net> Christopher Wolf wrote: >> I hadn't realized. The main reason I've had for wishing for POSIX >> compliance is to help with compiling generic off-the-shelf >> Unix source code which doesn't use the NeXT classes at all anyways. > >So are you saying that Rhapsody with POSIX will be easier to compile >off-the-shelf generic UNIX source code, Yes, exactly. -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
Newsgroups: comp.sys.next.programmer From: fgalot@x-lan.alienor.fr Subject: Using a serial port. Message-ID: <EDpuM5.8vA@x-lan.alienor.fr> Sender: news@x-lan.alienor.fr Organization: x&lan Date: Tue, 22 Jul 1997 10:32:29 GMT Hello. I'm using a serial port but I have some troubles. I'm receiving packets on the serial port. These packets are formated like that : <STX>aaa,b,c,dddd,eee,f,g,h,iii,j,k,l<ETX><LF><CR> <STX>mmm,n,o,p,q,r,s,t,uuuuuuuu,v,www<ETX><LF><CR> <STX>xxxx,y<ETX><LF><CR> example : 030,0,0,0769,000,0,0,0,000,0,0,0 001,0,0,0,1,2,7,0,00000000,1,000 1234,1 But I'm not receiving the good length each time I call the handler that is looking at the serial port. I can have 030,0,0,0769,001,0,0,0,000,0,0, for example. If you've got an idea thanks for help. -- --------------------------------------- ® ® | ® O_O ® ® | O_O -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- Fred Galot fgalot@x-lan.alienor.fr
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: CVS and NIBs - again Date: Tue, 22 Jul 1997 05:24:34 -0600 Organization: ErgoTech Message-ID: <33D49872.73EC@ergotech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Is anyone having any success with CVS 1.9 and NIBs? Using the script files that were posted here I get errors trying to check in a change. The error comes from rcs's "ci" program as it tries to diff the binary files. Looking at the code 1.9 appears to feed everything through ci whether wrapped or not and there seems to be no way to stop ci from diff'ing the files. The options in the cvswrappers files appear to make no difference to this. As an incidental side effect it also trashes the .nib directory, replacing it with the tar'ed file. If anyone has this working I'd appreciate it if they could post a "How To". For a while I thought I had a solution by uuencoding, but if it does happen to find common areas it croaks reading the file back out again. Can someone explain to me how this is supposed to work? Are the binary archives just a single snapshot of the wrapped file or is it a real rcs file with multiple copies of the file? Jim
Sender: Success@bellatlantic.net Control: cancel <5r27q1$p4v@world1.bawave.com> Message-ID: <cancel.5r27q1$p4v@world1.bawave.com> Subject: cmsg cancel <5r27q1$p4v@world1.bawave.com> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 22 Jul 1997 12:05:12 GMT EMP article removed by jem@xpat.com. Original Headers: From: Success@bellatlantic.net Subject: ========= SUCCESS ========================= Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!europa.clark.net!news-peer.sprintlink.net!news-pull.sprintlink.net!news-in-east.sprintlink.net!news.sprintlink.net!Sprint!199.45.32.37!world1.bellatlantic.net!news Lines: 37
From: "Robert Fisher" <rfisher@onr.com> Newsgroups: comp.sys.next.programmer Subject: Re: [Q] WDEFs & Rhapsody? Date: 22 Jul 1997 16:20:11 GMT Organization: Renegade Software Message-ID: <01bc9615$0da44400$0100000a@ivanova> References: <869482436.23064@dejanews.com> <EDo7rE.9EC@cam-ani.co.uk> Ian Stephenson <ians@cam-ani.co.uk> wrote in article <EDo7rE.9EC@cam-ani.co.uk>... > If you really did want to radically change a window, you subclass it. This > is generally not a good idea, and interface builder stops being helpfull, > but it allows you complete control over the windows behaviour. So, if I want to, for instance, create a circular window like the one in SimEarth (or the Windows NT clock), I'd just create a window (NSWindow?) subclass? Would it be possible to do something like just overriding the method that draws the title bar? (Something like MS Office 95 did to provide a gradient filled title bar with the MS logo in it.) Or would you have to override a lot of methods to get do this sort of thing? One thing I always liked about the Mac OS is the fact that the standard WDEF and CDEF code resources (which draw and define some basic behaviors of windows and controls) can be replaced (a la Aaron & Kaleidoscope) to change the look of the whole system. Is this sort of thing possible under OPENSTEP for Mach? -- Robert Fisher Renegade Software rfisher@onr.com
From: bettis@inetnebr.com (Mr. Jeremy Bettis) Newsgroups: comp.sys.next.programmer Subject: NT modal window problem Date: 22 Jul 1997 12:28:39 -0500 Organization: Internet Nebraska Message-ID: <5r2qk7$5dn$1@falcon.inetnebr.com> NNTP-Posting-User: bettis When a modal panel is up in Windows NT, it does not show up in the task bar, nor the alt-tab list. So it is possible for an OpenStep app to bring up a modal panel, and when the user clicks on the task bar or alt-tabs, another window obscures it. How can I find the NSWindow which is currently modal, so that I can force it to the front when my app becomes active?
From: tfu@bigfoot.com (Thomas F. Unke) Newsgroups: comp.sys.next.programmer Subject: Re: LISP on Next Date: Mon, 21 Jul 1997 20:21:51 GMT Organization: Disorganization Sender: thomas@gamelan.shnet.org (thomas) Message-ID: <1997Jul21.202151.599@gamelan.shnet.org> References: <33CBC2F1.D1EEA930@cris.com> <5qvst0$29r$1@news.freedomnet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Cc: bchin@NOSPAM.richmond.freedomnet.com In <5qvst0$29r$1@news.freedomnet.com> Bill Chin wrote: > I do not believe Allegro Common Lisp is still available. Of course, A Linux version of the latest Allegro CL is available for *free* . Have a look at their homepage. They even send you a CD - nice service.
From: Dave Taylor <ddt@crack.com> Newsgroups: comp.sys.next.programmer Subject: Abuse Source Code Released Date: Tue, 22 Jul 1997 13:32:52 -0500 Organization: Crack dot Com, http://crack.com Message-ID: <33D4FCD4.5D59@crack.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Crack released the Abuse source code to the public domain recently. Abuse was a shareware and retail game released for DOS, MacOS, Linux, Irix, and AIX platforms. The source is available at http://games.3dreview.com/abuse/files/abuse_pd.tgz and http://games.3dreview.com/abuse/files/abuse_pd.zip If you don't know the 1st thing about Abuse, http://crack.com/games/abuse and http://games.3dreview.com/abuse Lastly, if you want to discuss the source (this is a just-in-case thing- it may very well not get used), we put a small newsgroup up at news://addicted.to.crack.com/crack.technical That is also where we'll prolly host a newsgroup about Golgotha DLL's, mods, editting, movies and stuff like that later on. =-ddt->
From: Steve Watt <steve@watt.com.nospam> Organization: USENET spam abatement Sender: mcconent@cyberbundle.net Date: 12 Jul 97 08:39:47 GMT Message-ID: <cancel.5q5t31$r7d$5@its.hooked.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5q5t31$r7d$5@its.hooked.net> Control: cancel <5q5t31$r7d$5@its.hooked.net> I have cancelled this article which had a BI of more than 20. The original subject was: }Subject: ...Free Cash Grants
From: stefan@ping.at (Stefan Schneider) Newsgroups: comp.sys.next.programmer Subject: Help: nroff vs. /usr/lib/term/tab* files Date: 22 Jul 1997 19:49:28 GMT Organization: Customer of EUnet/PING Austria Message-ID: <5r32s8$q3o$1@news.Austria.EU.net> Does any of you happen to know the meaning/format of the files in /usr/lib/term/tab* ? These get invoked by nroff via options like -Tepson or -Txerox. I'd like to implement an nroff-2-html filter (for man pages etc.), and if I knew how to put up a special, own-purpose -T* file, I could probably preserve more attribs than 'bold' and 'italic' (which I've found out via trial & error using the -Tepson option), like carrying over accented chars and the like to HTML. I'm thankful for any pointer in the right direction. - Stefan -- Stefan Schneider Software Dipl.Ing. Stefan Schneider Lerchenfelder St. 85/6 A-1070 Vienna, Austria, Europe voice/fax: +43-1-523-5834 e-mail: stefan@ping.at (NeXTMail preferred, MIME welcome) web: http://members.ping.at/stefan/
From: opengraph_gamma@vvi.com Newsgroups: comp.sys.next.programmer Subject: OpenGraph/Prelude To Rhapsody Gamma Number 4 CD Ships Date: 22 Jul 1997 20:28:58 GMT Organization: VVI Data Control Specialists Message-ID: <5r356a$3s5$1@news2.digex.net> Originator: gsupport@ OpenGraph/Prelude To Rhapsody Gamma Number 4 CD Ships VVI Data Control Specialists (VVI-DCS) 311 Adams Ave.; State College, PA 16803 USA 888-DCS-OPEN ; 814-234-9613 info@vvi.com http://www.vvi.com State College, PA USA, 11 July 1997: Showing its continued commitment to the OpenStep/Rhapsody marketplace, VVI-DCS ships fourth gamma version of OpenGraph to major OpenStep customers worldwide, including OpenStep customers in financial service, manufacturing, pharmaceutical and biotech industries with combined assets of billions of dollars. Join those premier OpenStep/Rhapsody customers in the OpenGraph Gamma Program and receive: 1. Proven OpenGraph technology used now by the largest OpenStep customers to monitor billions of dollars worth of products. 2. No-fee technical support via e-mail, no-fee use of the OpenGraph gamma version during the gamma program, and no entrance fee (its free). 3. Free commercial copies of the GraphBuilder** application for ALL computers in participant's company, and one copy of OpenGraph-Developer** and OpenGraph-User*** sent to participants at the end of the gamma program. To take advantage of this offer contact VVI-DCS at opengraph_gamma@vvi.com. See http://www.vvi.com for additional information. _________________________________ PREMIER PRESS RELEASE: State College, PA, 20 January 1997: VVI-DCS announced an expansion of its OpenGraph on OpenStep/Rhapsody gamma program. If your business is interested in gamma testing OpenGraph please contact VVI-DCS at opengraph_gamma@vvi.com. "We've been working on the OpenGraph/OpenStep port for about a year now." said John Brilhart, Chief Technical Officer at VVI-DCS. "Our customers are reporting complex data sets from global real-time data feeds up to 500 events per second. That type of data reporting requires reliable and optimized report software. The gamma program is an important final part of our total quality control of the OpenGraph port." John adds, "The improvements and new features accompanying that port ensures our commanding lead in the high-end data reporting markets. With OpenStep and OpenGraph we provide compelling and unique solutions which have market advantages for our customers. For that reason we've always been fully committed to OpenStep on all platforms and have been working with NeXT and Sun for quite a while. We expect to apply the same level of commitment to Rhapsody (the Apple version of OpenStep) when it becomes available." About VVI-DCS: VVI-DCS, founded in 1989, builds custom OpenStep based data report and acquisition systems for the financial service, manufacturing, pharmaceutical and biotech industries and is the leading supplier of high-end data report software for the OpenStep marketplace. About OpenGraph: OpenGraph is a framework of Objective-C and C++ objects for reporting data in graph and textual formats and consists of a graph building application and pre-built objects. OpenGraph accepts real-time feeds from any source and serves as a graphing front-end for real-time financial analysis, transaction, production and inventory analysis, database systems, and instrumentation. OpenGraph is fully object-oriented and is well suited to systems which require reliability, exacting specifications and performance. _________________________________ A non-disclosure agreement is required for participation in the gamma program. ** no-license-fee commercial use license. ***no-license-fee and royalty-free commercial use license. (C) Copyright 1997 VVimaging, Inc. (VVI-DCS); All rights reserved. OpenGraph, GraphBuilder, VVI Data Control Specialists, VVI-DCS, and VVimaging are trademarks of VVimaging, Inc. (VVI-DCS). NeXT, NEXTSTEP and OpenStep are trademarks of NeXT Software, Inc. All other trademarks and service marks belong to their respective owners.
From: mmalcolm crawford <malcolm@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 22 Jul 1997 21:19:09 GMT Organization: P&L Systems Message-ID: <5r384d$5nm$1@ironhorse.plsys.co.uk> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> <nagleEDIw6C.HGy@netcom.com> <bradley-1807972327360001@bns.vip.best.com> In-Reply-To: <bradley-1807972327360001@bns.vip.best.com> On 07/19/97, Bob Bradley wrote: >What I meant was, is there any software for NeXTStep/OPENSTEP that >demonstrate the following quote: > >"...NextStep was able to create compelling integrated application >workspaces that home users and graphic professionals alike have been >anxiously awaiting." > Umm, the Web? For graphics professionals, Cambridge Animation's Animo: http://www.animo.com/ or Tailor (graphical PostScript editing tool -- OneVision's DigiScript also does this, but I think Tailor got there first?) Best wishes, mmalc. --
From: jalon@flute.ens.fr (Julien Jalon) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] WDEFs & Rhapsody? Date: 22 Jul 1997 22:49:59 GMT Organization: Ecole Normale Superieure, Paris Message-ID: <5r3den$h6b$1@nef.ens.fr> References: <869482436.23064@dejanews.com> <EDo7rE.9EC@cam-ani.co.uk> <01bc9615$0da44400$0100000a@ivanova> In article <01bc9615$0da44400$0100000a@ivanova>, Robert Fisher <rfisher@onr.com> wrote: >Would it be possible to do something like just overriding the method that >draws the title bar? (Something like MS Office 95 did to provide a gradient >filled title bar with the MS logo in it.) Or would you have to override a >lot of methods to get do this sort of thing? The object which draws the window is not the window object but a "FrameView" object which is not public and with no interface (private class from NeXT). You can hack this class (with a poseAS:) but first you need the "FrameView" interface. Fortunately some smart debugging can help you and you can find the instance variables and the method names (and types). I was looking for the same problem some times ago and I found : @interface FrameView: View { id titleCell; id closeButton; id iconifyButton; struct _frFlags { unsigned int style:4; unsigned int buttonMask:3; unsigned int defeatTitleWrap:1; unsigned int :8; } frFlags; } - _display:(NXRect *)rect :(int)something; // I don't have the other methods with my archives but you can find // them easily with debugging. @end You can also debug the DPS code produced by the application (with an option when you launch the application [I don't remember which one]) to see the code for the FrameView object. the FrameView object in the window is _borderView (you can look at /usr/include/appkit/Window.h It's a very bad hack. I didn't try to look at the NSWindow with Openstep but I think you can found something like. --Julien -- Julien Jalon | Ecole normale superieure jalon@clipper.ens.fr | 45 rue d'Ulm http://www.eleves.ens.fr:8080/home/jalon/ | 75230 Paris Cedex 05 | FRANCE
From: gtupar@ctp.com (Georg Tuparev) Newsgroups: comp.sys.next.programmer Subject: Re: CVS and NIBs - again Date: 23 Jul 1997 01:15:21 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5r3lv9$7b4$1@concorde.ctp.com> References: <33D49872.73EC@ergotech.com> In article <33D49872.73EC@ergotech.com> Jim Redman <jim@ergotech.com> writes: > Is anyone having any success with CVS 1.9 and NIBs? Using the script > files that were posted here I get errors trying to check in a change. > The error comes from rcs's "ci" program as it tries to diff the binary > files. 1. Put /usr/gnu/bin on the first place in your path 2. After source-ing the new .cshrc (1) compile and install rcs 5.7 3. Use the tricks we ware discussing here two weeks ago have fun -- grorg -- -- ------- /\/\ Georg Tuparev <georg_tuparev@ctp.com> | Currently in Dublin / /_ \ Cambridge Technology Partners 118/119 Lower Baggot Street \ / / Apollo House, Apollolaan 15 Dublin 2, Republic of Ireland \/\/ 1077 AB Amsterdam, The Netherlands Tel: +353(1)607-9083 Tel: +31(20)575-0492 Fax: +31(20)575-0500 WWW: http://www.ctp.com Hiroshima 45, Tschernobyl 86, Windows 95, Office 97
Sender: 188333477@compuserve.com Control: cancel <5r45sh$su3@news1-alterdial.uu.net> Message-ID: <cancel.5r45sh$su3@news1-alterdial.uu.net> Subject: cmsg cancel <5r45sh$su3@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 23 Jul 1997 05:47:01 GMT EMP article removed by jem@xpat.com. Original Headers: From: 188333477@compuserve.com Subject: YOU GET FREE 2000 $ex-WebSites FREE! @? Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!uunet!not-for-mail Lines: 99
From: jcr@idiom.com (John C. Randolph) Newsgroups: comp.sys.next.programmer Subject: Malformed PS code from Microsquish. Date: 23 Jul 1997 00:59:55 -0700 Organization: Idiom Communications Message-ID: <5r4dlr$hkj@idiom.com> Well, the geniuses in Redmond have done it again, and managed to barf out a postscript doc that Preview will render, but which manages to confuse the printpackage. The symptom is that only the last page of the doc prints, and I get a bunch of stack underflow errors from DPS on the console. Anyone know offhand, what needs to be done to a microsquish-generated postscript file to make it printable under OpenStep 4.2? -jcr PS: %!PS-Adobe-3.0 %%Title: Microsoft Word - tr-aop.doc %%Creator: Windows NT 4.0 %%CreationDate: 15:25 3/7/1997 %%Pages: (atend) %%BoundingBox: 1 1 612 792 %%LanguageLevel: 2 %%DocumentNeededFonts: (atend) %%DocumentSuppliedFonts: (atend) %%EndComments %%BeginProlog
From: znek@object-factory.com (Marcus Mueller) Newsgroups: comp.sys.next.programmer Subject: Re: CVS and NIBs - again Date: 23 Jul 1997 09:19:06 GMT Organization: Object Factory GmbH (Germany) Message-ID: <5r4iaa$3ou$1@leonie.object-factory.com> References: <33D49872.73EC@ergotech.com> Hi, Jim Redman <jim@ergotech.com> wrote: >Is anyone having any success with CVS 1.9 and NIBs? Using the script >files that were posted here I get errors trying to check in a change. >The error comes from rcs's "ci" program as it tries to diff the binary >files. Looking at the code 1.9 appears to feed everything through ci >whether wrapped or not and there seems to be no way to stop ci from >diff'ing the files. The options in the cvswrappers files appear to make >no difference to this. As an incidental side effect it also trashes the >.nib directory, replacing it with the tar'ed file. you have to use GNU rcs 5.7 in order to have this function properly. However, if you're working in an NFS environment occasionaly these .nfs* files are left for some time after removing a file placed in an NFS mounted directory under NextStep. These files cause the .nib wrapper (and every other wrapper) to fail, as the directory it tries to remove isn't empty as expected. Tom Hageman <tom@basil.icce.rug.nl> has released a patch which fixes this and other minor problems. According to his mail, the cvs maintainers accepted his patch and will incorporate it in the next official release. We're using this patch here and have had no problems since. >If anyone has this working I'd appreciate it if they could post a "How >To". For a while I thought I had a solution by uuencoding, but if it >does happen to find common areas it croaks reading the file back out >again. > >Can someone explain to me how this is supposed to work? Are the binary >archives just a single snapshot of the wrapped file or is it a real rcs >file with multiple copies of the file? Ok, here you go: Take a .nib file for example. If you have a bla.nib file then it gets versioned with rcs thus giving bla.nib,v. The contents of this file do in no way differ from a normal text file. However, where you expect text in this bla.nib,v file there's actually binary data - which is clear, because the underlying versioning framework is GNU rcs which is capable of handling binary data. This binary data is in fact the representation of a tar archive. Because I think that this patch will be of major interest to anyone using cvs under NeXTSTEP/OPENSTEP I'll include it in this mail (uuencoded). Hope that solves your problems, Marcus -- Marcus Mueller _________________________________________________________ OBJECT FACTORY Gesellschaft fuer Informatik und Datenverarbeitung mbH Otto-Hahn-Stra{e 18, 44227 Dortmund, Germany Telephon +49 (0) 231 - 97 51 37 -0 Telefax +49 (0) 231 - 97 51 37 -99 e-mail marcus@object-factory.com http://www.object-factory.com begin 666 TRH-cvs-1.9.diff.gz M'XL("`?)U3,"`U122"UC=G,M,2XY+F1I9F8`U5GY3]O8%OX9_HH3JHZR.,'. M8I+T4=1APBL2VY!TAJ=Y563L:V*1^%J^3@!-YW]_YR[VM;,,::$=O:@JP;[+ M6;_SG4.U6@5WP>I6H[?/8G?_K3LA[GT0-MR=WXD'EVX"T`2SVS=;_8X-5J]G M[];K]<(>O64X#^&"+L`RP<(MS;[9E5NJQ0\<RRU03IXB8H`?A#XU('89_D<6 M!B3.G0$T2OA&Z)A&YP#XOEW(/H&/FRF*<>4DDPJ^V/ESMP3YCTNCI[$?3$EN M(5X5.C-2>8=K=])CYN$T"._%TK$7Q%`6XM3?\P<5^!>8XO@=OK0$Y#%@"0E= M,B9Q3'$Q_@AII<(/%">JQY8!XHT!>ZX3AC1!Q69T0>`MVU,*XPWSZ53)`U*> M#4+G!!)+_Q).X(;I`7ZK/VL8<?A^%7YVW/MY!,F$@#N/8Q(FL"`Q"V@(U!>/ M^24-J.[_O86$U-_)-D4/H8SXN[))SA`K"U&[8[2>4.(A=J((XU?L2FA!7U2` MN`F-Q<H0[BA?\$#C>Z[TU[KA18%M-PV[(P.;N]-N&W8W[\Z_\.=N#91R1T=' M\,?H^N-G\%!>@N<YT^D3E]^9$,<#LD!MT!$A)8_$A8`!(\F15(I_V$.0N!,H M7Q\/QVXJII#LXM/9F0$SPIAS1Y2@9J6B]OWY,B6[/:/73K,WM>^C.YE1;\F? M5C$M']U9M-;C<'BHXDZ*QP.P&"7\^QIW[:`5>20DP0Q#Q&'HYIC'@@&,PI0X M&(%!`LZ4AH2;C?NDUS0LLY4Z90OQ<]$HG75R>G,^Z,/#Q$FT=XZ6$FR-KEG< MP5'N>W^##6I2N^,TG?BB7*3?8MH#9E@L<S0([W2>UU3^;@&#-97'SZ0[7[;S M53BH7?0]'/FB^.T=8`1TTP"6,#9E1)N^M&J_(D`NH9^Y%?HI+=E3Z`("-N;N M1*OLQW0F?L-<EA@GH6L',WXL5AR"E=44`2-U^9T#XYAOQX`:1S%U,>E7L17O MKV?!OI_I#54X]3GT/#@AE]KQZC1$".);T':S.089ERHB\2Q@O*@PF)`837Q+ M?!H3D5&6:1N6E>'<!H-N+:?,N)UGJU1IN]`MY4+WGW'4ZUO]1>%OM5J&U>KJ M*F6U\4''6H9$85<);@(`-"QME]%KX(+.IQYPLXMJ!2CQ6X;7@L]KM>_@7F_/ M$)L5]5EE535M6$4FDGF,FEOB5J%@@?UZ@>^OI;Z]]=17K3^)`\%[H0O-)E^_ MB??R]<(2/*2A[&(%)['G),HB%<EWK:;1L>P,<K9A8%N&]K>!\O,4#!>]"OT2 MK%:IGP+$#U=?Z_3C%/^F0+$MT["MC%RMR[057ZY4*MTK9$#YU6U,>KJN-IS. M*NE2/_YPZ?3>@FPK22^`='X;J\0?D@B:';#L?KO9;W?7)WYN3S'YFU:_U=F0 M_(1$8RF@#.)(=&C\5=ON&>V#MD99\:#;S'<#+(GGK@PCC*:J%Z5:N1,GYE:Y MG?M_7'T8?1R??[CY_$XU#D&8"'QDB9/,V?)3)TG(+$JP"C7?R6;C#;K`W-6M M9#S3HD()22?\])-**LY!!Q>7H\'YU>@_\.4+Z*>#F]/AJ%(1!Y(0@UFW,:,) M=B;XSPFSVS$//.(3),@H)7J]$?KLAG]@XH0>KW)J.Y>8(<'SD=S%=!YZ</L$ M%^1F-!P-KAA<G`SQ3$X`6:*HK>`\$Q%*96D"U#6ODU9*28/&?R\>J.*QO:YI MNZ3BN(R71'@;C;@%]'VXAS=<E:^)CXYY@)C8R3)=%#1W2AD1"_E-*6*)`I?7 M4#15BEHH'_R.5G1N)42IC.%(I1%*'6/J[C%]PF_)4AQ%,CI-S>76BZ6#:I-X M*D9*^:XW'X@O%7DI"M6+-"4VX4(PBVB<K-*!=M\TUZ-"MF,)$^P^XN!:3)A' M".IDC.1+`5[6A2]DH5D(-J8&82VLC"T[WTSKSO!0<GOQ8(DQ+S*NO,/K"A$` M<@BRZ7Q,9I&ZJHC#Q2'2QE'0,D*+N%""YCCB:PI:WM0;R_TO5F'928[G85XN M<LX19#GU$(F9RM(>'\BU-(K+!^T\BB/RPI2Z]\0S5``:$`@PR<,Y5#.94L@6 MCV_%^"Z/\EE:<\U2_OVM"MC-GF&W#@HP(U.R!%D54B"Z/$:JOX_$D"SG:`ET MZ"-Q$?I'S]($;4>/Z'"_C9W0U2_$F.KD[,._A^-?/YT.1NES^((.9))&C&?4 M$TW4$>C5YY>_C$[/!WB;6:EDE3]0>`V9J25':2%'L9L%`$L5E@!4@MPT!_.Y MCFE=[]F?91UCE##1H@@>AXU)@0'B^P"9J&C^W$"/IW%C0BFPF1,G8E.0,*`/ MG!-2#PNCAPCEHY)\F6Y^1%-).,(X\9,^3)@6HTJ2S;N`3P%C$E$6"%CD/*L! M,`SN)HU&0V\;*G(J!R=0=J;)A,[O)DA79[<8$)33ZP?G211GRL<I*/4TN.=W M@6Q[#'T:EN%,-$!PY@,F?CR-@[L@=*922CI/TBDS/YDKR@>MN%B?5-!2ZL6M M$TT=%Z-$VH:A9OP&QT](_.#$'FODQIS_;P&Z5)SX+(3'6P[D2HHQ:$HC.NNM MQ_G<AX;!OXFQH)SYL0AMF/AE"2G&WELF2//Q;\.KZ\')Z8VN![4-/9C<J4># M6TX&M^+NZFQQ>:%Q4)B5?X]VX%R13<3`0.9+PL>M:@S*YJY+B$<\3`"E_*OT M;'C.^K\3:,-)+*FE4_QG8U-J)P>HZV*OMB'V:J\-CM]<05I80=J=]2WIR>G9 M8/SQ\NR7P76>;>0G@%O5Y]R\R-RZ`RSE"Z5TA/X+AZ@%_.\O!_;Z?O4YT7-I MN5:%17&*_IJIHJ-M)5E4;B\SN^]EX)2];[+R"L-66,7_<HSI*BAV!YIFWSSH MMS8TWGI+D6.WD)4?K.78:HQI6!FO>8-5>3KW"*JT8(W)'@_XDN)=ZOPQ=MN\ MQ!S"WG_#8QJF@/";?,U@^(1NFT$9\;("*-R>HF+(&K"7A>.ST\'%:#S\='5U >>3V2@U,A@HJN[R&"H"AOK,]_+\K_`).\HVOR'P`` ` end
From: frank@this.NO_SPAM.net (Frank M. Siegert) Newsgroups: comp.sys.next.programmer Subject: Re: HELP with SCSI device problem Date: 23 Jul 1997 10:47:17 GMT Organization: Frank's Area 51 Message-ID: <5r4nfl$ctk$1@orista.ipc.uni-tuebingen.de> References: <5qpgpt$sgs$2@excalibur.flash.net> Cc: daj@nwu.edu In <5qpgpt$sgs$2@excalibur.flash.net> daj@nwu.edu wrote: > I have a SCSI data acquisition device that was manufactured by Analogic, its model DASM-AD14. Technically it is a 14bit 2Mhz 8 channel acquisition device that has a 2Mb buffer and a SCSI interface. > Here's my problem: > Analogic provided instructions and the basic outline for acquisition code for a Sun3 or SPARC system using SunOS4.0, and I'm trying to port it over to the NeXT/Intel system. > > Analogic says the device can be accessed at /dev/rsdnc, where n is the scsi appropriate device number . Well I've got a properly terminated system and the acquisition device is SCSI ID target 6 LUN 0. Now the problem is analogic says it can be accessed by: > char *dev_name="/dev/rsd4c"; > open (dev_name, O_RDWR); > but when I check /usr/adm/messages I have the errors at the bottom of this post:. Also, if I go ahead and run the program it doesn't open the device. This won't work. You have to use the SCSI device driver already provided in NS/OS either in the sd or sg incarnation. See my web page (download area) for the source for cdwrite-2.0 for NeXTSTEP which provides a software layer for access to the sg devices. For example to write to a SCSI device use (no error handling is done here): result = scsigetdev("SCSIDevice_name"); // opens the generic layer for named device // device should respond to an inquiry scsisettimeout(10); // 10 sec timeout result = scsiwrite6(data_buf, // ptr to data buffer olen, // data length C6OP_WRITE, // SCSI opcode 0, // SCSI device lun 0, // lba numblocks, // len 0); // ctrl scsiclose(0); Reading from a device is similar. The common SCSI opcodes are defined in /usr/include/bsd/dev/scsireg.h. There is usually no need for a special device driver, all can be handled in user space. -- * Frank M. Siegert [frank@this.net] - Home http://www.this.net * NeXTSTEP, Linux, BeOS & PostScript Guy
Sender: ILCJS<pentaf@ilink.nis.za> Control: cancel <5r0ld2$rb5@news1.saix.net> Message-ID: <cancel.5r0ld2$rb5@news1.saix.net> Subject: cmsg cancel <5r0ld2$rb5@news1.saix.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 23 Jul 1997 12:06:54 GMT EMP article removed by jem@xpat.com. Original Headers: From: ILCJS<pentaf@ilink.nis.za> Subject: For the best golfing experience of your life ! Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!news-peer.sprintlink.net!news.sprintlink.net!Sprint!howland.erols.net!infeed2.internetmci.com!newsfeed.internetmci.com!in1.uu.net!196.25.1.18!news1.saix.net!usenet Lines: 19
From: Steve Watt <steve@watt.com.nospam> Organization: USENET spam abatement Sender: Kris Jacobs<jtsnake@serv01.net-link.net> Date: 10 Jul 97 07:49:22 GMT Message-ID: <cancel.33c1fb15.2@nntp.kalnet.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33c1fb15.2@nntp.kalnet.net> Control: cancel <33c1fb15.2@nntp.kalnet.net> I have cancelled this article which had a BI of more than 20. The original subject was: }Subject: Try Me!
From: tom@basil.icce.rug.dev.null.nl (Tom Hageman -- remove .dev.null to reply) Newsgroups: comp.sys.next.programmer Subject: Re: CVS and NIBs - again Date: 23 Jul 1997 17:42:16 GMT Organization: Warty Wolfs Sender: news@basil.icce.rug.nl (NEWS pusher) Message-ID: <EDqKwz.Dtr@basil.icce.rug.nl> References: <33D49872.73EC@ergotech.com> Jim Redman <jim@ergotech.com> wrote: > Is anyone having any success with CVS 1.9 and NIBs? Using the script > files that were posted here I get errors trying to check in a change. > The error comes from rcs's "ci" program as it tries to diff the binary > files. Looking at the code 1.9 appears to feed everything through ci > whether wrapped or not and there seems to be no way to stop ci from > diff'ing the files. The options in the cvswrappers files appear to make > no difference to this. As an incidental side effect it also trashes the > .nib directory, replacing it with the tar'ed file. > > If anyone has this working I'd appreciate it if they could post a "How > To". For a while I thought I had a solution by uuencoding, but if it > does happen to find common areas it croaks reading the file back out > again. > > Can someone explain to me how this is supposed to work? Are the binary > archives just a single snapshot of the wrapped file or is it a real rcs > file with multiple copies of the file? Sounds like you are experiencing some of the wrapper-related bugs in the base 1.9 version, or you don't have the latest versons of GNU rcs and/or diff installed. Anyway, here is a semblance to a tutorial I whipped up after the previous wave of cvs questions (but hadn't got around to distributing it yet. Here goes): begin 644 cvswrappers-HOWTO.1.1.rtfd.tar.gz M'XL(`(00U3,"`^P]VVX<1W8C[SK8&2"@*",(D`1@:<@-*8;3/1=R*(XM43)% M6T36'%DB:0<:F>KIKI[IL*=KU)<9TK+!Y,U^TNONOO#="2!@7P)D`^@#$L,_ ML%[]@?4#09ASJKI[>BZ4*)&Z6*AN%+NGNNK4N=6IT\6JTWK'Z[I:NTU=+W>] M^ME&52G`Z?JFH:9.ZR#S^<6%!9(BA!3FY_F5D'QX#7^0<JF\4,PO8@%2*"V4 MRRFRD'H%1^#YFDM(RF>MIY;K-BFU4V_=H1\M_XV;UW/P.%=0EA3#,DW%AS\O MTD8AGR^'<A\A_T*AE"_&\B_.ER%KL5A83)&\E/]+/];69E._.O>V4+.Y??IG M/_S#4SY28ZG,2X?_KH3_?/`_W?]T7_P5=[W[9/Y1]V\&_H.8#>8.X_IF\7^8 MP\,T/?U\W?IY7+P'->U-TY_C<W^X#[S^_CL:UW]].+J_OFGZ_^G^O0<]G1BV M1D=)Y4VRGT??/]LBO9GZ?S15@S)X3?BS&_M7V0V6O$;Y/[?Q_>KV%79X>&,; MKU>W>]?#P][O-[G_'E?O"]5>S=>-?V$Y7\5T>"BNO=^C<'Q3[>?15N?PD"P3 M]>>A_T=1</2O)&6O!_^S$\?SS\Y.8$G\.UQ#OK^\B?#A]VF?"?AP9%["^6[? M),&[+^'\V1]_F4J=^47J#-SAK-N9Q/T[</]+N'YW+I7Z19B/Z2_"^W<@C9_Y M);__%:2_"O.QSE^'^2FH^S=GWN7W?X#[OPWO_Q/N_RX!<R8!\V8B/\7AB?-? M_C>3.CN=BJ_R>%GSOY]OX,W)VWCZ_"]9+)5+`_._"_G%DIS_?17'_1J(.%_3 M',^Z7S.9X_MUNP89IM>U/(]<IW:'^I:NO5\S"S6SQ0SJ.J3:M$WO_:\SM;8& MBM-=*I;SX7VSL%#*PX^6YC;L0C&\<_'N?DUG-G,!_OLUEQKY6L.EU,G7ZG9` M\P*8:]3\7=`"^%O(E_FEG,=+L<`SBV6>62KRS-(BSYPO\LSYBSQS`1XB_O5\ MS<K7`MMA#@5BBO,U4\_7=#,/)+F4:)#\)B6>[EIMWR-K)/`@AQ'J:'5;/$MT M#6)2S0^@#C/)RM8MA6PTJ2>@U#6/&H0YA.YJK;;=`]EV6<<RX%G7\IM$@P*: M:UO4)1T`:$$%`8O,%)1%:-]O6L[.A4HMDP%T+Q*XWN=XWZ^MT\\W/@9$FI8. M[""($ZEU+<-O+L'/6I-:C:8/MYFO,__^]6ERL3#$1<C4"\#%`L<B(Q0EP5N2 MN3^`;YD$SIN!L<!C&.<$Q_D#LC;=`=$:*#J?BQG4`O4!A'6S6MV`1RW+L3S? MU7P+2AJ62W6?N7L*X;HU[9$FZPI!6CD@I&9;*+?_('J3ZCN$!7X$2JEE()>U M]SA\K<XZ/?6Q'&AVCP5NW&Z7N3N6TT@TB-4!45[;T5K40YT2.$=@`$B&=]TD M*S@B@(<-5`PS)*2#6!X!0FFDDD`-DI+OJ:8@K$_@I:6%/NBO3^P<MY$4<WKR MG"0@9T`/D`\#4NMXG,E<BX42(=\-:@)S2+>I^<AD_M1R1C&[9T9.E]?S%\M] M-NJ5L'J498U8+4A*H!11$U(RDKTC&88,'V%=N)[3KKT7=D_3LD'G9X9!C.[I M<R,:.\HJS)$V\SRK#FT];_\!6V\\KQY<X'UY-"]TUFI9_I&&*ZDKM7IHPCZY M4;VY<75]`W+RI$+63#0EI*FA9<.A"A!'4^)2H-)"6P*\!476P0'0+,<+V=(& M'BN.59\CW",DU-<58C(;G`!O#B%.V[8`"CT`ZL+(YO=L9@0A'.\\4M^#PIPW M%G85E&8/`<CLZPZ]T;DW.!MHB"T/2LVL^5A((^"16*;5:P4MX);EZ)1<HY]H MKL[`'(]D*[2LZ3LCI"<:50BTX+L6Z->]P/(I8.X:B$/;!?)<(#G!Q(XE"$3B M?*T!O*F#D;<XBD"TPTB+@H=%0/8FF&Y%49XUT"\M%GM(OOZQ\RG\&AH^P37B M3E%2(N$RDIY^>:R%FNBP/>1@V"]R+K4U'V19#QHH9#X0F]8NY/#R\>"&?/:" M.O0*7PS2I*WYT"F]:*C&;M."AK`QP.)97E6A!&@/K'9Y_4P?Q.@8O+^*Z@GV MH@V&L:]+8+<)F61PKU,',U`'[K(`>*GYE2%0Z>%>8_KMBJK"7\6AN[[2IMJ. MPMR&V@[J*N:H($5/#1QK5P6\%<1[?>WZ+:6NM'<:"KQF*8TOCZ5`'%=]1VN` MJ#TO0,<&I8K*`2Y%;$^`KH_7-XFKBYZ']\BG.=0"H`WTS-=L9`08&S7P7-5F MNF:KH%DWAKF4\)W0K=<ZFF7S5X'39PX@K"PHB\=CSLF:XNNVBL=MJT=E.(ZL M5S=6PR$DZ6F`W0/#;S!NR1WF<]O-_5/QLN/2%@.3.<,-I:OJ\-KC^!<0A#*L MLB`\NZOM@=%$*P^#$'=Z6=<AKN7M)"K@W3TWU!'6(M=!/UJ@QAR+#^#]_0J8 M'<M6+%VGBALT%,>^+#`CA:6E136_J!:+T(/EI,_QYG]BHW\*\S]/6_^Y,%_B M\S^E^85\$8PJSO\LE.7ZOU=R3)Y7ZY:C>LW,))["^(9^EZ_M4/2QUE<_W[BU ML7HC&J=[;Z#":VQJX.90ZI"VK>G4`"@PJ%!7C#,PXKO,!CL.@!HN"]IH9RW' ML#J6$6BV\.'GN`$//4AN[`%(HA4&E4%,TV"NL6#C2ZL=C1`<ZVCNQ>(F',"W MP?_JU>]"ZQ\QAJZL,%!82+P\P-`0/@$PNA@N#,W7%-V&H0<PZ_`6>19ZPIW0 M;?4"VP=88`9AI-$XL+`ZP`DA0F$Q]P.EQ.!"F&LU+`?([CF-_&4"IY``Y8%2 M?4Q`-`2#Q=Q`^#(4\4$A,P,<HL:%4*2A2PV6NHGN%A<:_([EZK6I#EZT+KQR M5(`F!9R%&G`871=]+0>]V3[OFGS0,<`(P]V5.@L4KVDK,+!>AAJQ8PY5DM;Z M"$--BJ7<.NODEA:@O<R-JQO7+R6&;%#1BOB#>?%-H-?Q!XB_S>#-`VME,FV7 M-2[=10\4IT3(5/YN)@-Z<9M,39(<O0>&YPYY'[GG9-)4;S*2W?0`MTJ63&%5 MDKW--4-1[F2AP"YX\?F,:64R)G/)VOJ-S8WMC]9^LXJLRDY=R0(H@V4R:=Y" MSH"\7IELHJ5T+_O27=%L7\FO"$XC3H.K.G/["_5.[8*JSDY5:H7*]%TR"9Y@ M#.'#JRO_N'E#`.H#H=2U'2CUR=7/M[=6;]Y:JZY?RBMY0"W=ZI"<.=!>]"L! M#LA-Z\;H)_!(('U[)517BRN#8%F.)&#?R9++1/5;;75J2K%9H^4UL#JR+V+< MK.!:.IV.,+WKV@BG3J9X&<&-G`,,4;]H4LVHD%E5;4_?Q3K`:A"X"XB&M8&: MRTA1@G3`035H1W4"T.-8!OW<B:K'#SA/.0*8!3)/IPWPRN'"L9N*RB304[\P MJ.@FT)<K4^J<^D7N*<>4>A]!#]8R1.;3:_)";?CS]32Y/(K#0D`W7*93SX-^ MGDU*A61;VF[\1F!Y69)D%U;_\.JMU>VDEO:Z4"\7!=#:`:-$%$6=&JB"4";) M;=#6.R22IPENM!?:Y7#>/7!BVV"UL.-"7U-"#0GQ026Y.\3T&I=40C,B(RJ4 M`\6"9B`'+^\YYV[4,R/&U/R0>B`]07;:;9&<:XZB1YWE!49K;AJ,7^X>5(V` MD5ASL,,EN,?KSG7NCFP"RW,EPQOH?5"&WS6!1R1WDP3_T"5#?.YA#0H04H69 M#2=`9^96V]--[)0#]8!!?.3,Y1R6X[CE<G5\H\KIT&7[0>D620-Y@N4)(L., MUFA#P(?!(1I[,HST+P)W)Q-*><CD)"I]U>MLL<"G(PC3-=[IP&KF+H'5K!5F M:]C)$D6AXT!_@U?'"K^KNYJ#4P7B5W\](]/.?'TW&Q'I]_>Q)(>X"."]$S5@ MITYR]B#[C%$T97JRCOI)/)WD[;7JS(8Q6$SDST"7T9R]"U`4GO"\2W>?8H;Z MJPM+=/N+])W(X*15Z"_IVHPR6[M0(>(ZI>8<&&%J1568%>S<8AS#L2UJ%D8Q MR(\L:$AT_)0K=9)T;C:/[E.94,N1!Q$#@W[FB>&JEP5N5,("S75"^.:@"4Q3 MVZ.1'>PS?<+W`3>MYTU!GS88]9QI7TSDP>`'F`OI\$&_D)'O?^'[WPMO_'KA M_5_EA5)9OO^]C?N_5C\_/'QR^-/A3T]^>O+GQS\^^?'QGQY__\/_//ZO'_[[ MAS\^^N.C?WOTW:/O'G[W\/<'OSOX]N"W!]\^^.;![GX7TWZ#F?LF,]G6]B9; MW=[<[NX/0_^_0W/[O)J=R*I9M8AIN81I.;HFSF0^)%$>SHGSZCE(XQ-X]L,_ M/-QBX_`TJYY3L\N34&M2+2U/+HLTZHR>3?+RO!XD;&&<MS"_W`^_A_LDQRN" ML\+3?"*5XE2*:9CLIP'Q'TO"_^D)?ZIV#QH/?OMP]Z#[H+%O[F^Q:]OSU6O5 MH])J=9-M`>\;^]\<?/OP=X]^_ZA[\,U!EM/0#__/C[,<[WX^GN><'!\[.W$& M4FKL[!@FO!^/$C]%>:P7U1_$_\<GDYPC675S'T^*Z<%6E!Z(9,(IKHE\N(KR MXD3Y82L#\!\7.4^S$T8,G4(R]P4T\T$P<`U36"ZLL[_)(AKZX?_I,4"OSB^? M5S=%Z0C"`>!\(*Z=\#[ZS:^"EI@.Q+\X`O_O?T#H*]7QB21TQ#4X$*F32%%> MCY8>#>=#.0[R9WX9]6%\8FM?\/FXJSQ#FF)9G9O`?E%<'L;_6G5C&_$7/$^- M??<P"0=M@SB3>:FQSD%2%H#_!.]]ZB#_5P'Z*L`W.?;FP;'Q/^B$\A!R&#\" M?X!>W=P>'XOXCL^/L\HVDD<DA[,3PC8-\A]WI&X!_ENAKAP7?N.@TZ=+9\9* M5>30(/X4X%-VAN./-01\7%<<I5&_$/^.T*E0!F<GA%T:Y#_:[BT6Z4.$?P_B M(&QQ3>)OAOBOC,1_,X%_,!+_7AM)^,D^(?@O;.P@_P7^7'_Z\.^'.LR?(?Y/ M8$\=A?_6MLE>1'_Z^3\N\!_B/W+'9'WZ_]SZ@_I_;N(H_F]!"[S_/J#/81^2 MN"/_SZNC].?'QV(L&E<C^\-Q`MP:O?2P"ZDA4IP?VY\'L?T9B?\&UY]S$V`) MN=T/0CGT[&9\]MO2@7%`X#_*_B"'A/W?&K+_0J>XK3D8'`<&[?]H^X/6F6Y' MX]?18U>_W>_''>MFU=)H^U^]!O9SQ/C[())'_W4K@?=6W_C+?:+A\6OYVO)& M5?@`23\L]A]"?>$^!/<G0M^AYS^$O@/WA4:/O]53\7^6T8\;Q!]M-M*`M80N M;?$>AV='7/%WF,_M!HY(D*Y58>Q;QEY5Y)P9Z?\\YGXG:-9\V(+0)YXX1'$5 MOQ&Z*(-T"'L3^J+\/#?D'W[_@QCWT0?JT<!MZK:`SL>',,6X5Y.XEQ+^W_@0 M_U_"_I%G'L6EX\`IS;SX6TWIF2U,+IWLO>GI-)0NG/2]K+0DWDVRZJAKZ838 MCY\K+I66GW)>//O>B>"_!S"X?A]U/?/>R:"O)-[#1E_'WWG^*"6IL>>.1C(V M.C))WY'0C,WMY/O4LZX#^[N.A?_S]-47P?]9?$]>CX>_/.3^+WD\<_X7\T\Z M]?N"\[\+"T4Y_ROG?^7\KYS_E?._<OY7SO_*^5\Y_ROG?^7\KYS_E?._<OY7 MSO_*^5\Y_ROG?^7\KYS_E<=+7/\;/3K9-/`+?/^A4)J7\[]OX_SORSWD]Q]D M?%/Y_0?Y_0?Y_0?Y_0?Y_0?Y_0?Y_0?Y_0?Y_0?Y_0?Y_8=7B[_\_H/\_H/\ M_H.<_Y7SOV_'_._+7/]+2A@(FI2+\\7Y0KFPN(CQ'PHR_M^K.>*8?SQTCPA+ M5:<>B6-]8C0YAL&;2-UR-'<OC)OG,PSO!Y4_BPN&GU30F:/3MD^Z/.!3%''% M$L'I165>E,=H]5V*$7\!#@\0B&&K;!&:CW]C`2/=40=#?6-L4,N+(ZT';2@= MA2/DX?#BRB!.`.<%>E,$ND.<4,2:JS<Q7G\8^39LN:_=!(4`PG+XMQZ008QX MS.X(^MHNJ]NT1:B)L9P!HKTWA[&>,<B@9GN,.!0#<&DB;B%&!^0?E&``I^W2 M#G5\C%C;HFX#"'59B^"M",N'P6=%\#DH3(:#@&'N1\QM:3Q28$0^E]S,5/B9 M`#6Z)@.V,I<HB=\B+B"&+S1TS372MQF/1$8ZFAW0._V_,#:7*,WE&3X#43`' M(S)S1,UTFE."89<!&Y^ZZ32O7<&8PTWD@\CFI?UT&L.%'ZML*YT.VAB\";CD M-YG!;-;8B\I_LGKSXU6D;:5ZXY]X\9UT>H?N=9EK8(@X_*@)<Z+2<0:&)A1A M([ER\\<BF+A0@]R]`*/7&M2V6A:J""_!N9#)S/(@D@3;(M/U:0Q#-#W$^O"3 M%[_VX+D_XGGX%`L`0!Y:_30!&J[6/56`GU!/.UT,+:WA:JWB*8+T-6^GI7F@ M.*<(%&U9<)H`':X[IPBP&YRR-K9/&R#^ZS@&*(,O_WS\/R'F4VCC&?&?R2*/ M_UPHSY>+Q7*>QW\NY.7WOUZ-_W>>A`&@28X/BNFP=^=@P/1=3?=#IZX%I6`L MC.+RS^BA_\0C'&,-[E18!O@WZ>R5F<D+H%EJ&)?9JTRM:RT8@\E496K-J(2? M%YGK$%`V#,Y>5O-EM7B1Y/.58J%2S(,7T"*KNVTRE<WP6,*A[\5='>[L8>1C M'L(R_KK6+GB*/()@&'PP.U7(1DY/IM7AO_LS>:!JE'X$F@<*%B$R[;ZB&-$V M#B=;O/SWA8R(A]CX4@=?C(?]313^BH2!.'.Y.+QC+N>U-1?\N]PN&M5<1@0K M?&;!/H0Q#/$DCZX?\B,.%#V*YA.__R6^<G.B!4`OL/ZGN"#W?\KU/W+]CYQ_ MENM_Y/H?N?Y'KO^1ZW_D^A^Y_D>N_Y'K?^3Z'[G^1Z[_D>\O<OV/7/\CU__( MX^7^_^<4(@`^W_PO7_\#-W+^]VV<_Y7Q_V3\/QG_3\;_D_'_^KUL&?]/QO^3 M\?]D_#\9_T_&_Y/Q_V3\O^>C0<;_D_'_$H>,_R?C_\GY7WF<;/YWX^;U'#S. M%90EQ3AI_+_RD?._Q4*Q.,_G?^'A8F&A@/._Q46Y_O^5'"M-S6G0W[!&)9/Y MR+7(.NN0P@(IE"KY<B5?XHOS"=E@+7)=:]"6YA#R`;#J2EWS+%NQ=)TJ;M!0 M'/MR)I.>)7S-N.4H.IE9$;<7*N1#3=\)VO%B==R[&*N=P1?>S^%.O$S:I9[/ M7(J[*2V'Z)J'.PR)PPBJ'W6IHU./&`'?`A"V-(?[_YA#J.LR5^$X8&%$`*_; M"+WM,AW06'-(V[4<'_=:\N*DA?LT&]2;X]M+_2;-I'7FXH95CA7?7V!"N6[3 MXKM):5B/Z7H`Q0S1'M^Q&=1=WB:E[6VQ0'_;L%QLU>QM")TC+6V'$H_JS,&M M%)Z724-)-]XER[>,ANO[02]M2M8_ND5`1`;?%)K&YJQ6F[D^-B9V1FZ[NL?) MA,966*N-FVO_&;1:<(@:.18(<OA6":!PD/66"=S'/:R.YU/-0)9'HH(F9S3# M`(HZ`/TSYNX0S64!(*];Q&MIKN\`916Q06,`,(JP;6LZ,#4!D=2IB2*.Q8=; M,".YQX4TTZ=N5W.1P[.SLR0T0ZKGZNJO8QU+?P9-54%8I$CR%T%;*PO_W]ZU M/J>-)/'/]E\QD,LNV,)!/`V)[4HY^-95MN,U9-=7N10KA#"J"(EBA!^UV?_] MNGMF-",>,8X?MU>'/N0AYM$]T_WKQTBM&@GL9J%02/717=K34,AXD=G0I=0L M[HHN6^F+2?%EN?AN#(LT\,-!9.&;N_"'=VVQV+FR\&U8HJ]:M*IUAOTV67+! MNN;B",@X=^)A'G[8^',SP\S+C<9W7?'Z;M(0ID*YR[^%MAMJF&D8^.%7:HIB MQ7)$3F&?MIV]8T4:?@.;9IAWZ\-&@JYTA;3FX*\PRN=Q0!I1WK8M1K]8+.LZ M81C%2O)>\ZQD&&:8!H&DAPEZEA!M$$1-_Z)-P(5IT)O,]RX,#?YF2Z$%O76$ M6A;&#%2$7MP%.5*OZNRPK3??7R&B^IG6)KU#$T(*N2;&0LPU!.X.8?6(B92J M@-:;_!IP,/1"=A71>^^H?<#T0[?A48)=*UFUJA!LW,Y:Q:KMFMOY%_R]N<TD M<P<'!^PSN`U?6!_H]6`\)PCND'YGB,B";[_C;H61=^NY^,XU]^(#P11>'!`* M@#9W<=CNNHI,HNSLT\F)I?!:$EK,YV6_/Q_'Y&[#:E24]JKUO76'HZ@_LY]V M6BUOW=%XX8ZSO3TI=X(\%,"TE."_%VS7!JPB%2SP\=4VAS-ICRS&(Q9XSC49 M1R?`E^]AV7!/&B7+!D,M-V4%\@UI%)MU='QYVFJ"E7-BO3L',PJV@-=$[MB! M\>_FDC78%MP=*G4BZZHEO0=JS\@8DHZBD4[T?%OJ[PHPN"WU^!YUQV8;#\)! MO47/L9&/DM]&'21@5PFP@#%\NS!9^LS\^J4!<@;]BBNAG^22WX4NUN.@%S$3 MEJD>!/X/=%E@G("N#=#X+K788W9B4PA&"N+?"(Q=[`X"U477#91^'EMA_D(B M[&\2OMD6.EPWN.189P-`IQ"%@2Q:`@`R!2&C(A[>9.1S-"J<85T+2_HEI%%V ML6;9=H)S2Q9T93J%QFW<:Z4RJXENQA#=_\Y&/?VJ/TK\[7+9LLN[VDK9%;@! M<>0,)-*Z"G`C`-"PM)I&+X"+:!KT&2X[62OT>%]SF!:#AJ]LX$#??M:BSM+U MF?>JMO7"2F<BGDZ`<YMFG?=^172SP/5M+'9]97L5V[%=5BIA^V5^;RIJ8CD7 M++@W@3A#KDA>^+MVR:K:M01R5O'`5A3M'P/E^UTP:/0D[A=YM9)]!1`OSK[F MZ>48_R%!J=E%JV8GSM4B39O;RSE+I6.%!"@?',:HT;6U07=64J?V\<6ITWU3 MM,TIO4XQD.*WO3$K59E=:U9*S<KN8L4W^J25OV0WR]4ERI].7[#<F"(T_*E2 M:UB5>D6C+-W8+9G1`(\G4U>($4C35G^LN'*'S@17I3<=?#Y_W_FE>_K^\LM; M&3CX84SXR&,GGO+9NTX<>Z-Q#%:H]%8$&Z]@"XJ;.I2<C#2I+`-.)_OI)ZE4 MZ(.VSCYV6J?GG7^Q;]^8OMNZ/&YW\GD:T`M!F'480_70L"!4F,P.>M#W!AXX MR"(MLQ,.^"5>*CECR>Y(,0<';Q"K3$GOCIUYEYUVIW7.*9T#@X`#R&/IVI+/ M,R11RHDE`%Y-GC13DAI8_'VZ(8W'ZKRJ<$G*<0XF&<-LT1A70,\'?3#@RC]$ M/JK%.F!B-=%T,FAN$'&/&N),"K'(P)D<4E`E70NY![]C-351.,TL>*$12@Y3 MU-&CNH.S)"H.)%G5DO;E%I.EA6H9>5)&,F;4:PKB8TF>D4+Y@U*)9;B@<H&S M[D"E62PN1H6DQPPFU)J`@PLQ82;+R'))%'XM#,TU>6,R$58&RUBNF<&TC@SW MA&]/-V8\YNO$5]Y0N5Y4>1%TWL:CL9PJC</I)-+25-`L0I-<2$(-'_$I"<TM MBXU%_T>S,+M),DEK;`XYRVJ'J.8?[4\#$W)EC>+B1L5$<4!>%D28.K:D`%K, M)S`QX9QM)30IR*;;/4K?F2B?J#5RIOSO'V6@5FI8M7(]!3-")3,LL4(21&?3 M2(7],27)C(T60`=[1!/!_NA<&KGML"-:W'L3)W3U#Y2F.CIY_\]V]]=/QZV. MNL^^P09RX49T1U&?@J@#IEN??OS0.3YMP6S%?#ZQ_+[$:Y8LM?!1RN"CU$HI M`%,,"P#*,".;`_I<`+4N-&I?A!WCD<<I1"$_CDXS#`^08\U#5P1_KJ_3TU1J M,Q(I?NKDQYQ%-^@31GV9LQ\`DSPYBL'@AX)*#Q'&F=SIP>1I@'0VKWS,`DZ\ M<<1]@D7TLW88:_M70RPVF71K2^=4GLCDG"`>1M.K(;BKHQX(1(3N]8U#9R;] MB`Z-P/+Z7^GD080]EAX-S'!"&IY7R(J;^LR!J,2S$IEEQI&144RT0F,]4HI+ MP1>N#IUUF.<9.$-RD,%WC#3G_YJ`SA@GS(6@O!D@EY$>@W9I*+)>.9V/>VA9 MR:&;R/EQ.JX;Y`2D6-G7G)SFP]_:YQ>MH^-+;0^VE\1@HJ=.#:Z8&5S)=Y=C MT^2IP$%BEOD[K`/ZBGQ("0.A+U2<5J9!^=1U/:_O]4$!)/-/$K/!.(O/"?3" M"2S95EG\>V53<"<2J(MD;WN)[&T_-3C^L`4I@P6I5!>'I$?')ZWN+Q]//K0N M3&_#S`"N9)^-?%%QY0@P8QI*L1'ZA(-L`9Z_U&N+X]7[2#?4<B$+U^DL^E.J MBI:V.661NCWKV3W7`BOO?=DJSWG8$JOPY!C4E5SL*BL5F\5ZL[PD\-9=TCYV M&;SR^D(?6Z8Q+3OQ:UZ!50ZF?0]8NN8[PRP*?$;Z77+\+D3;:&+V6/;?X2$5 MJB5`^$W\S%G[#K9MQ'*`EWD&Q&6E*P9>`\2R[/#DN'76Z;8_G9]_O.B(Q"F1 M(*7K.4@@%^65_>7[I*R?ROI;//_U1-5?[ZW_:MMUVZC_BL]_5:KK^H\O<^GZ MKU3\599^/13E7IECN!=S5?95+=B=!Q1^_6[95WNF[*N(7]R)/R;7W@G`)>>, M$LU]?8HU[PAQ+Q;U]D7)>M/'[!FE;,FN@5_>`Y_?9R-P]Z^@HSF\@S7[95%Z M,9:H6C_AE`><Q#2*YT[QN'H'C!02C-\'<'K!G?+V>O@M@8%_"VWQH;9KGL<C M8$3.,84;,)_RA,W%'E"=?[0CFQ^.+\[>G[;V_N@Y7/B+6.?U#RIA?Q;=T%Y( M%]LX3X>A;SR(_&ZPSC],@0=RR*`#YC%P\,,%Q!N,H1[+P`'`-\/'T2#&(UYA ML4806.TP]2&%:(S;G92M5;5JW0%(3?8?DM(LEL"E,KJ%,"H0Q85"#X(B:,CV MH5TINP;YOSW^F_5_G['^`V-U`_\K^"V8]?._+X?_])F5K+G769FPX0@&?<0^ M/AUCUAA#+!\AR0F2&-Z-PG@2(63B-USZ(S^$$28"7D2""4!O]@L!EO@\B\,I MCQ+@YSW49T\V!4$"8P'6`<E@%H<%:([H(QU)DD`\E4O@UO/4TZX$J?A,`PR4 M9(_X.`K[Z*@FSSC(9-4L81H_=R0A$$HX?M@'4)08G1`2IA\BQG6BC]0820T1 MCL!(:?H$FT?80J!\DVSOYW?`40S$.JZW_^6=XG/?O/TN-2?\-^P7HD$!B=JG M41&PT1KB'0[S7@&G9,M^?O7S)I4%%V5=UOB[OM;7^EI?ZVM]K:_UM;[6U__? *]1_MT.D%`/````&G ` end Hope this helps, Tom. -- __/__/__/__/ Tom Hageman <tom@basil.icce.rug.nl> [NeXTmail/Mime OK] __/ __/_/ IC Group <tom@icgned.nl> (work) __/__/__/ __/ _/_/ Confused? You won't be after the NeXT episode.
Sender: 98761237835@compuserve.com Control: cancel <5r5r4t$5k1@news1-alterdial.uu.net> Message-ID: <cancel.5r5r4t$5k1@news1-alterdial.uu.net> Subject: cmsg cancel <5r5r4t$5k1@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 23 Jul 1997 20:56:01 GMT EMP article removed by jem@xpat.com. Original Headers: From: 98761237835@compuserve.com Subject: FREE...Don't Pay For $ex $site Pa$$words, Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!news-peer.sprintlink.net!news.sprintlink.net!Sprint!uunet!not-for-mail Lines: 99
From: memphis@macconnect.com (Garry Roseman) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Mon, 21 Jul 1997 23:35:44 -0500 Organization: Writer & Freelance Programmer Message-ID: <199707212335441409135@accs-as08-dp15.dlls.grid.net> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> Bob Bradley <bradley@apple.com> wrote: > In article <19970717190420809661@accs-as07-dp12.dlls.grid.net>, > memphis@macconnect.com (Garry Roseman) wrote: > > > We Apple revolutionaries could, if we had the balls, develop a system > > with a click-compile industrial-strength iconic language environment on > > a fully PMT objectized game-like machine. This will be possible because > > with OpenStep we will be able to quickly create entirely new application > > metaphors with interactive objects, much as NextStep was able to create > > compelling integrated application workspaces that home users and graphic > > professionals alike have been anxiously awaiting. > > Kinda of off topic but, do you know of any NeXTStep/OPENSTEP applications > that might demonstrate this? > > Thanks I guess I have to learn better how to be REALLY sarcastic. Go back and re-read everything I wrote with _sarcasm_. No, I don't know of any OpenStep applications that home users or graphic professionals have been anxiously awaiting, and that's my point. And I have not the darndest idea what a click-compile industrial-strength iconic language environment on a fully PMT objectized game-like machine IS. But some folks, on hearing that b.s., want it; and that is also my point. :-) <---<belated smiley> -- Garry Roseman <mailto:memphis@macconnect.com> Writer & Freelance Programmer Memphis TN USA
From: kamran@tybrin3.hsv.tybrin.com (Kamran Talai) Newsgroups: comp.sys.next.programmer Subject: C Question Date: 23 Jul 1997 19:42:14 GMT Organization: HiWAAY Information Services Message-ID: <5r5mqm$pdn$1@parlor.HiWAAY.net> Is there a way to declare an unsized array with elements of arbitrary size? Specifically, I am trying to declare a pointer to an array of structures to be malloc'ed during program execution. Is there a way to have the size of the elements made known in the declaration so that an element can be pointed to solely by its index number (without having to multiply the index number by the sizeof the structure)? I apologize if this question is too broad or basic for this forum. I think Microsoft specific compilers do allow declaration of unsized arrays and so I didn't want to be flamed in one of those forums. Thanks in advance for any and all replies. -- Kamran Talai TYBRIN Corporation (205) 837-2027 FAX 837-3472 kamran@hsv.tybrin.com
From: bettis@inetnebr.com (Mr. Jeremy Bettis) Newsgroups: comp.sys.next.programmer Subject: __try-__except in OpenStep Date: 23 Jul 1997 13:38:00 -0500 Organization: Internet Nebraska Message-ID: <5r5j28$lrn$1@falcon.inetnebr.com> NNTP-Posting-User: bettis Does anyone know if __try{} __except() {} blocks work in OpenStep gcc? I am linking against the Oracle (not EOF) and it raises an exception sometimes that gives gdb fits. I was going to try to catch and ignore it. Don't tell me about NS_DURING NS_HANDLER, I am talking about Win32 exceptions.
From: 98761237835@compuserve.com Newsgroups: comp.sys.next.programmer Subject: FREE...Don't Pay For $ex $site Pa$$words, Date: 23 Jul 1997 20:55:57 GMT Organization: Gming Inc. Message-ID: <5r5r4t$5k1@news1-alterdial.uu.net> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary=_-=496 This is a multi-part message in MIME format. --_-=496 Content-type: text/html; name="Sex~1.htm" Content-transfer-encoding: base64 PCFET0NUWVBFIEhUTUwgUFVCTElDICItLy9JRVRGLy9EVEQgSFRNTC8vRU4iPg0KDQo8aHRt bD4NCg0KPGhlYWQ+DQo8dGl0bGU+U0VDUkVUIFBBU1NXT1JEIEZPUiBBRFVMVCBTSUdIVFM8 L3RpdGxlPg0KPG1ldGEgbmFtZT0iR0VORVJBVE9SIiBjb250ZW50PSJNaWNyb3NvZnQgRnJv bnRQYWdlIDEuMSI+DQo8L2hlYWQ+DQoNCjxib2R5IGJnY29sb3I9IiNDMEMwQzAiIHRleHQ9 IiM4MEZGRkYiIGxpbms9IiMwMDAwRkYiIHZsaW5rPSIjMDAwMEZGIiBhbGluaz0iI0ZGMDAw MCI+DQo8ZGl2IGFsaWduPWNlbnRlcj48Y2VudGVyPg0KPHRhYmxlIGJvcmRlcj01IGNlbGxw YWRkaW5nPTI+DQo8dHI+PHRkIGFsaWduPWNlbnRlciB3aWR0aD0xMDAlPjxmb250IGNvbG9y PSIjRkZGRjAwIj48Zm9udCBzaXplPTc+PGI+RlJFRSEgRlJFRSEgRlJFRSE8YnI+DQo8L2I+ PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT01PjxiPlRI QVQnUyBSSUdIVC4uLiBDT01QTEVURUxZIDwvYj48L2ZvbnQ+PGZvbnQgc2l6ZT02PjxiPkZS RUUhITwvYj48L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjMDAwMEZGIj48Zm9udCBzaXpl PTU+PGI+PGk+PGJyPg0KRU5URVIgVEhST1VHSCBBRFVMVFNJR0hUUyAmIzE2OTsgPC9pPjwv Yj48L2ZvbnQ+PGZvbnQgc2l6ZT02Pjxicj4NCjwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9 IiNGRkZGMDAiPjxmb250IHNpemU9Nz48Yj5JVCdTIEZSRUUhISEhITwvYj48L2ZvbnQ+PC9m b250PjwvdGQ+PC90cj4NCjwvdGFibGU+DQo8L2NlbnRlcj48L2Rpdj4NCjxwIGFsaWduPWNl bnRlcj48Zm9udCBGQUNFPSJDb21pYyBTYW5zIE1TIj48Zm9udCBjb2xvcj0iIzIyMTY1QSI+ PGZvbnQgc2l6ZT03PiA8L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjRkYwMDAwIj48Zm9u dCBzaXplPTc+T05FIFBBU1NXT1JEIDxmb250IEZBQ0U9IkNvbWljIFNhbnMgTVMiPjxicj4N CjwvZm9udD48L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjMjIxNjVBIj48Zm9udCBzaXpl PTU+IDwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9IiNGRkZGMDAiPjxmb250IHNpemU9NT48 c3Ryb25nPkdFVFMgWU9VIFVOTElNSVRFRCBGUkVFIEFDQ0VTUyBUTyBPVkVSIA0KMTgwMCsg T0YgVEhFIEhPVFRFU1QgQURVTFQgU0lURVMgPC9zdHJvbmc+PC9mb250PjwvZm9udD48L3A+ DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29sb3I9IiMwMDAwMDAiPjxmb250IHNpemU9NT48 c3Ryb25nPklGIFlPVSBBUkUgVElSRUQgT0YgUEFZSU5HIFRPIEdFVCBJTlRPIFNFWCANClNJ VEVTLi4uPC9zdHJvbmc+PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGRkZGRiI+PGZv bnQgc2l6ZT01PjxzdHJvbmc+IDwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9 IiMwMEZGRkYiPjxmb250IHNpemU9NT5PbmUgUGFzc3dvcmQgaXMgYWxsIHlvdSBuZWVkISEh PGJyPg0KPC9mb250PjwvZm9udD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgRkFDRT0i Q29taWMgU2FucyBNUyI+PGZvbnQgY29sb3I9IiMyMjE2NUEiPjxmb250IHNpemU9Nj4gPC9m b250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT02PjxzdHJvbmc+ TE9PSyBXSEFUIFlPVSBHRVQhPC9zdHJvbmc+PC9mb250PjwvZm9udD48L3A+DQo8cCBhbGln bj1jZW50ZXI+PGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PGZvbnQgY29sb3I9IiMyMjE2 NUEiPjxmb250IHNpemU9NT4gPC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGRkYwMCI+ PGZvbnQgc2l6ZT01PjxzdHJvbmc+VU5MSU1JVEVEIEZSRUUgQUNDRVNTIFRPIE9WRVIgMTgw MCsgT0YgDQpUSEUgSE9UVEVTVCBBRFVMVCBTSVRFUzwvc3Ryb25nPiA8Zm9udCBGQUNFPSJD b21pYyBTYW5zIE1TIj48L2ZvbnQ+PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iIzIyMTY1 QSI+PGZvbnQgc2l6ZT01PiA8L2ZvbnQ+PC9mb250PjwvcD4NCjxoMSBhbGlnbj1jZW50ZXI+ PGZvbnQgY29sb3I9IiNGRjAwMDAiPjxmb250IHNpemU9NT48c3Ryb25nPkp1c3QgQSBGZXcg T2YgVGhlIE92ZXIgMTgwMCsgWFhYIFJhdGVkIFNpZ2h0cyBUaGF0IA0KWW91IENhbiBFbnRl ciBDb21wbGV0ZWx5IEZyZWUhPGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PC9mb250Pjwv c3Ryb25nPjwvZm9udD48L2ZvbnQ+PC9oMT4NCjxoMSBhbGlnbj1jZW50ZXI+PGZvbnQgY29s b3I9IiM4MEZGMDAiPjxmb250IHNpemU9Nj48c3Ryb25nPlRFRU4gU0lURVM8L3N0cm9uZz48 L2ZvbnQ+PC9mb250PjwvaDE+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29sb3I9IiMwMDAw MDAiPjxmb250IHNpemU9ND4xMDAwIEhPVCBBTkQgTmFLRUQgLSBDZUxlQlJpVGlFUyBBTkQg VEVFTlMgLSAxMDAxIE51ZGUgRXJvdGljIA0KVGVlbnMgLSAyMCBBbWFUZWVucyAtIDYyMyBO dWRlIENlbGVicml0aWVzIEFuZCBUZWVucyAtIEFBQSBTTFVUWSANClRFRU5TIC0gQW1hdGV1 ciBUZWVucyAtIEFtYXRldXIgVGVlbnMgRnJvbSBBcm91bmQgVGhlIFdvcmxkIC0gQW1hdGV1 ciANClRlZW5zISAtIENvY2sgSHVuZ3J5IFRlZW5zIC0gQ3VtIERyaW5raW5nIFRlZW5zIC1G aXN0IEZ1Y2tpbmcgVGVlbnMgLSBIb3QgDQpOdWRlIFRlZW5zIC0gTm90aGluZyBCdXQgVGVl bnMgLSBUZWVucyBJbiBUcm91YmxlIC0gVHdhdCBTdWNraW5nIFRlZW5zIDwvZm9udD48L2Zv bnQ+PC9wPg0KPHAgYWxpZ249Y2VudGVyPjxmb250IGNvbG9yPSIjRkYwMDAwIj48Zm9udCBz aXplPTY+PHN0cm9uZz5BU0lBTiBTSUdIVFM8L3N0cm9uZz48L2ZvbnQ+PC9mb250PjwvcD4N CjxwIGFsaWduPWNlbnRlcj48Zm9udCBjb2xvcj0iIzAwMDAwMCI+PGZvbnQgc2l6ZT00PkFs bCBBc2lhbiBUZWVuIC0gQXNpYW4gQmFiZXMgLSBBc2lhbiBCYWJlcyBzaWFuIEJhYmVzISAt IEFzaWFuIEJlbGxlcyAtIEFzaWFuIA0KQm9uZGFnZSAtIEFzaWFuIEN1dGUgR2lybHMgLSBB c2lhbiBEZWxpdGVzIC0gQXNpYW4gRXJvcyAtIEFzaWFuIEZsb3dlcnMgLSANCkFzaWFuIEdv bGQgLSBZb3VuZyBBc2lhbiBXb21lbiAtIEFzaWFuIEhhcmRDb3JlIEFTSUFOIEhBUkRDT1JF IC0gDQpUSEUgTEVHRU5EIENPTlRJTlVFUyAtIEFTSUFOIEhvbG9zdWl0ZSAtIEFzaWFuIExv dmUgRG9sbHMgPC9mb250PjwvZm9udD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29s b3I9IiNGRjAwRkYiPjxmb250IHNpemU9Nj48c3Ryb25nPkNFTEVCUklUWSBTSUdIVFM8L3N0 cm9uZz48L2ZvbnQ+PC9mb250PjwvcD4NCjxwIGFsaWduPWNlbnRlcj48Zm9udCBjb2xvcj0i IzAwMDAwMCI+PGZvbnQgc2l6ZT00PiMxIE51ZGUgQ2VsZWJyaXR5IFBhZ2UgISAtIENlbGVi d29ybGQgQWNlJ3MgLSBYWFggQ2VsZWJyaXR5IFBpY3MgLSBDZWxlYnJpdHkgDQpDZW50cmFs IC0gQ2VsZWJyaXR5IENoaWNrIFBhdHRpZXMgLSBDZWxlYnJpdHkgQ3VtYm94IC0gQ2VsZWJy aXR5IFBpY3R1cmVzIA0KQ2VsZWJyaXR5IFBpY3R1cmVzIEdvbGQgU2l0ZSAtIENlbGVicml0 eSBTa2luIC0gQ2VsZWJyaXR5IFNraW4tQUJDIC0gRGEgUm9ja3MgDQpudWRlIGNlbGVicml0 eSBBcmNoaXZlIC0gTWFya3MgQ2VsZWJyaXR5IFBpY3R1cmVzIDIgLSBOdWRlIENlbGVicml0 eSBBcmNoaXZlIC0gDQpTdGFyTG9jYXRvciAtIENlbGVicml0eSBTdXBlclMgLSBUaGUgRGFp bHkgTnVkZSBDZWxlYnJpdHkgUGFnZSAtIFRoZSBNdXN0IA0KU2VlIE5VREUgQ2VsZWJyaXR5 IFNpdGUgLSBUaGUgVWx0aW1hdGUgQ2VsZWJyaXR5IENvbGxlY3Rpb24gPC9mb250PjwvZm9u dD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PGZv bnQgY29sb3I9IiNGRjRGQTciPjxmb250IHNpemU9NT4gPC9mb250PjwvZm9udD48Zm9udCBj b2xvcj0iIzAwODBGRiI+PGZvbnQgc2l6ZT01PjxzdHJvbmc+VU5MSU1JVEVEIEZSRUUgQUND RVNTIFRPIE9VUiBYWFggUkFURUQgDQpDSEFUIFJPT008L3N0cm9uZz48L2ZvbnQ+PC9mb250 PjwvcD4NCjxwIGFsaWduPWNlbnRlcj48Zm9udCBGQUNFPSJDb21pYyBTYW5zIE1TIj48Zm9u dCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT01PjxzdHJvbmc+R2V0IFlvdXIgQURVTFRT SUdIVFMgQUdFIFZFUklGSUNBVElPTiANCk1FTUJFUlNISVAgRk9SIE9OTFkgJDEyLjk1LiBJ dHMgR29vZCBGb3IgQW4gDQpFbnRpcmUgWWVhciBBbmQgR2V0cyBZb3UgVU5MSU1JVEVEIEZS RUUgQWNjZXNzIA0KSW50byAxODAwKyBPZiBUaGUgQmVzdCBBZHVsdCBTaXRlcyBPbiBUaGUg d2ViIEFsbCBZZWFyIA0KQWxsIEZyZWUuIE5ldmVyIFBheSBBbm90aGVyIERpbWUgVG8gR2V0 IEludG8gQSBTZXggU2l0ZSANCkFnYWluLjwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PGZvbnQg Y29sb3I9IiNGRkZGMDAiPjxmb250IHNpemU9NT48YnI+DQo8L2ZvbnQ+PC9mb250PjwvZm9u dD48Zm9udCBjb2xvcj0iI0ZGMDA4MCI+PGZvbnQgc2l6ZT03PjxzdHJvbmc+Qk9OVVMhISE8 YnI+DQo8L3N0cm9uZz48L2ZvbnQ+PGZvbnQgc2l6ZT01PjxzdHJvbmc+RlJFRSBYWFggQ2hh dCBSb29tcyEhITwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PC9wPg0KPHAgYWxpZ249Y2VudGVy PjxhIGhyZWY9Imh0dHA6Ly8yMDUuMjE2LjI0OS4xOC9jZ2ktc2hsL2RibWwuZXhlP3RlbXBs YXRlPS9hcy9zaWdudXAuZGJtJmFtcDtzaXRlPXN0MzUzMjUiPjxmb250IGNvbG9yPSIjODAw MDgwIj48Zm9udCBzaXplPTY+PHN0cm9uZz5JIFdBTlQgVEhFIFBBU1NXT1JEIE5PVzwvc3Ry b25nPjwvZm9udD48L2ZvbnQ+PC9hPjwvcD4NCjwvYm9keT4NCg0KPC9odG1sPg0K --_-=496
From: Michael Sitarzewski <michael@meetmeonline.com> Newsgroups: comp.sys.next.programmer Subject: Re: msql Adaptor Date: Wed, 23 Jul 1997 18:23:19 -0700 Organization: INTERNET AMERICA Message-ID: <3C414B12656D698C.2E74C7123349365E.B654644FA301DD97@library-proxy.airnews.net> References: <5qdtb6$657@nebula.mpn.com> <33CB23CA.41C6@mpip-mainz.mpg.de> NNTP-Proxy-Relay: library.airnews.net NNTP-Posting-Time: Wed Jul 23 18:23:22 1997 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Stefan Ried wrote: Is anybody out there working on a EOF Adaptor to mSQL 2.0 ???Man, I'd sure like to see this too. Does anyone have further information? Michael.
From: brian@spam.be.gone.interactiveinfo.com (bedwin) Newsgroups: comp.sys.next.programmer Subject: Re: NSThread event queue performance bug :( Date: Fri, 11 Jul 1997 03:15:04 -0400 Organization: Interactive Info Sys / What Software Message-ID: <brian-1107970315040001@ubppp-245-023.ppp-net.buffalo.edu> References: <wz7mf02lxu.fsf@w09.sfc.keio.ac.jp> In article <wz7mf02lxu.fsf@w09.sfc.keio.ac.jp>, penrose@w09.sfc.keio.ac.jp wrote: I wish I could help you, but I am working with my first threads on BeOS right now. I hope somebody helps you solve your problem. BTW, I thought your "CONTEST!" subject was quite humorous. > I have used mach threads just fine in other applications, but they are > cumbersome, inelegant, and the biggest bummer, they are not Openstep > compliant. I have thought also about implementing this with > distributed objects, but I keep thinking to myself that threads should be > easier than that. I should be able to run something in a seperate thread, > and when the thread exits, I should be able to drag an object on the screen > with full system performance. Right now, Openstep doesn't work that way. > I am leaning towards mach-threads, but I must remind you in the community > who actually managed to read this far, mach-threads are scorned all throughout > the Openstep documentation. > > Christopher Penrose > penrose@sfc.keio.ac.jp - bedwin brian@interactiveinfo.com
From: mmalcolm crawford <malcolm@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 23 Jul 1997 20:59:06 GMT Organization: P&L Systems Message-ID: <5r5raq$39t$1@ironhorse.plsys.co.uk> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> <199707212335441409135@accs-as08-dp15.dlls.grid.net> In-Reply-To: <199707212335441409135@accs-as08-dp15.dlls.grid.net> On 07/22/97, Garry Roseman wrote: >No, I don't know of any OpenStep applications that home users or graphic >professionals have been anxiously awaiting, and that's my point. > You asked, in effect, what applications NeXTSTEP had produced for which home users and graphics professional had been anxiously awaiting. I listed a few: From: mmalcolm crawford <malcolm@plsys.co.uk> Subject: Re: How to save Apple Date: 22 Jul 1997 21:19:09 GMT Organization: P&L Systems Message-ID: <5r384d$5nm$1@ironhorse.plsys.co.uk> Umm, the Web? For graphics professionals, Cambridge Animation's Animo: http://www.animo.com/ or Tailor (graphical PostScript editing tool -- OneVision's DigiScript also does this, but I think Tailor got there first?) --- > I guess I have to learn better how to be REALLY sarcastic. > Yes; I guess you should learn a bit more about Rhapsody's heritage too. Best wishes, mmalc. --
Sender: Izzy<Don't repley> Control: cancel <5r61cl$dbo@halley.pi.net> Message-ID: <cancel.5r61cl$dbo@halley.pi.net> Subject: cmsg cancel <5r61cl$dbo@halley.pi.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 23 Jul 1997 22:50:32 GMT EMP article removed by jem@xpat.com. Original Headers: From: Izzy<Don't repley> Subject: NUDE !!!!!! Newsgroups: comp.sys.next.programmer Path: ...!newsfeed.direct.ca!dispatch.news.demon.net!demon!bullseye.news.demon.net!demon!newsgate.unisource.nl!halley.pi.net!news Lines: 14
Sender: 18q34757@compuserve.com Control: cancel <5r6hru$nt1@news1-alterdial.uu.net> Message-ID: <cancel.5r6hru$nt1@news1-alterdial.uu.net> Subject: cmsg cancel <5r6hru$nt1@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 24 Jul 1997 03:23:48 GMT EMP article removed by jem@xpat.com. Original Headers: From: 18q34757@compuserve.com Subject: Get FREE SEX SITE..password is.. Newsgroups: comp.sys.next.programmer Path: ...!howland.erols.net!news-peer.sprintlink.net!news.sprintlink.net!Sprint!uunet!not-for-mail Lines: 99
From: Izzy<Don't repley> Newsgroups: comp.sys.next.programmer Subject: NUDE !!!!!! Date: 23 Jul 1997 22:42:29 GMT Organization: Double Hangman Message-ID: <5r61cl$dbo@halley.pi.net> You just have to see this! Its a page with nude girls. You will find more pics then in any Newsgroup!! http://www.geocities.com/Area51/Cavern/1802/nude.htm Don't tell your friends!!!
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33c9247f.0@news.unibe.ch> Date: 14 Jul 1997 00:43:10 GMT Control: cancel <33c9247f.0@news.unibe.ch> Message-ID: <cancel.33c9247f.0@news.unibe.ch> Sender: a18@a.a MMF cancelled by sef@kithrup.com
From: Timothy Luoma <luomat+next+usenet@luomatpeak.org> Newsgroups: comp.sys.next.programmer Subject: cc -pipe Date: Wed, 23 Jul 1997 23:56:53 -0400 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.NXT.3.96.970723234217.4017A-100000@cc344191-a.ewndsr1.nj.home.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I just built the pine suite (pine, pico, imapd, etc) quadfat for NS on a P-133 with 64 ram and a lot of swapspace free. First I ran it without the '-pipe' flag and got this: ./build nxt 1109.48s user 104.87s system 67% cpu 29:56.27 total as the results... it took about 30 minutes to compile.... So I ran it again (on a clean set of source code) this time WITH the -pipe flag (and no other changes) and got this: ./build nxt 1114.88s user 99.72s system 68% cpu 29:21.42 total and it took about 30 minutes to compile. I was hoping the -pipe was going to speed things up.... perhaps there is something about the pine config that made it not noticeably different.... Is there some other program that might help show if there is a difference or not? Is there some other way to speed compiles? Thanks TjL ps -- note: my From address should have a DOMAIN listed as 'luomat.peak.org'... if it doesn't then sendmail mucked with it and you'll have to add it, or mail me as luomat@peak.org
From: Evstathios Marinos <emarinos@marcon.de> Newsgroups: comp.sys.next.programmer Subject: Re: CVS and NIBs - again Date: Thu, 24 Jul 1997 07:35:49 +0200 Organization: MARCON, Karlsruhe (GERMANY) Message-ID: <33D6E9B5.D89FECBE@marcon.de> References: <33D49872.73EC@ergotech.com> <EDqKwz.Dtr@basil.icce.rug.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, are there compiled versions of the latest patched cvs 1.9, rcs 5.7 and diff 2.7 for Windows NT 4.0? If yes, where can I download them. Statis -- MARCON - Evstathios Marinos Consulting Evstathios Marinos | Phone : +49 721 37 71 78 Gartenstr. 2 | Fax : +49 721 37 71 79 76133 Karlsruhe (GERMANY) | E-Mail: em@marcon.de
From: Steve Watt <steve@watt.com.nospam> Organization: USENET spam abatement Sender: randy97 Date: 12 Jul 97 02:54:03 GMT Message-ID: <cancel.33c52b4f.1@pgh.nauticom.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33c52b4f.1@pgh.nauticom.net> Control: cancel <33c52b4f.1@pgh.nauticom.net> I have cancelled this article which had a BI of more than 20. The original subject was: }Subject: http://www.love.com
Date: Wed, 23 Jul 1997 20:09:06 -0400 From: joe.ragosta@dol.net (Joe Ragosta) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Message-ID: <joe.ragosta-ya02408000R2307972009060001@news.dol.net> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> <199707212335441409135@accs-as08-dp15.dlls.grid.net> <5r5raq$39t$1@ironhorse.plsys.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-transfer-encoding: 8bit In article <5r5raq$39t$1@ironhorse.plsys.co.uk>, mmalcolm crawford <malcolm@plsys.co.uk> wrote: > On 07/22/97, Garry Roseman wrote: > > >No, I don't know of any OpenStep applications that home users or graphic > >professionals have been anxiously awaiting, and that's my point. > > > You asked, in effect, what applications NeXTSTEP had produced for which home > users and graphics professional had been anxiously awaiting. > > I listed a few: > > For graphics professionals, Cambridge Animation's Animo: > http://www.animo.com/ > > or Tailor (graphical PostScript editing tool -- OneVision's DigiScript also > does this, but I think Tailor got there first?) > For a list of apps which have been announced for Rhapsody, see MacOS Rumors: http://rumors.netexpress.net/rhapsodyapps.shtml Not too bad for an OS which isn't even at DR1 yet. -- Regards, Joe Ragosta joe.ragosta@dol.net Visit the Complete Macintosh Web Site http://www.dol.net/~Ragosta/complmac.htm
From: gtupar@ctp.com (Georg Tuparev) Newsgroups: comp.sys.next.programmer Subject: Re: CVS and NIBs - again Date: 24 Jul 1997 14:19:44 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5r7oa0$kj8$1@concorde.ctp.com> References: <33D6E9B5.D89FECBE@marcon.de> In article <33D6E9B5.D89FECBE@marcon.de> Evstathios Marinos <emarinos@marcon.de> writes: > Hi, > > are there compiled versions of the latest patched cvs 1.9, rcs 5.7 and > diff 2.7 for Windows NT 4.0? If yes, where can I download them. Check: http://www.cyclic.com/ -- georg -- -- ------- /\/\ Georg Tuparev <georg_tuparev@ctp.com> | Currently in Dublin / /_ \ Cambridge Technology Partners 118/119 Lower Baggot Street \ / / Apollo House, Apollolaan 15 Dublin 2, Republic of Ireland \/\/ 1077 AB Amsterdam, The Netherlands Tel: +353(1)607-9083 Tel: +31(20)575-0492 Fax: +31(20)575-0500 WWW: http://www.ctp.com
From: Steve Watt <steve@watt.com.nospam> Organization: USENET spam abatement Sender: gtr@cyberbundle.net Date: 12 Jul 97 08:39:47 GMT Message-ID: <cancel.5q5mia$n1r$1@its.hooked.net> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5q5mia$n1r$1@its.hooked.net> Control: cancel <5q5mia$n1r$1@its.hooked.net> I have cancelled this article which had a BI of more than 20. The original subject was: }Subject: >Entrepreneurs Needed
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 24 Jul 97 12:26:46 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFFCFA91-37CBE0@141.214.134.235> References: <nagleEDIw6C.HGy@netcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: "John Nagle" <nagle@netcom.com> nntp://news.itd.umich.edu/comp.sys.next.advocacy, nntp://news.itd.umich.edu/comp.sys.mac.programmer.misc, nntp://news.itd.umich.edu/comp.sys.next.programmer On Fri, Jul 18, 1997 12:23 PM, John Nagle <mailto:nagle@netcom.com> wrote: > There's Prograph. But graphical programming doesn't scale well; >you can write a small program that way, but big programs turn into >acres of spagetti. MacTech had a recent interesting article on Spreadsheet 2000. It was written in Prograph and has something like 4000-5000 classes... rob -- <mailto: "Robert A. Decker" comrade@umich.edu> Listen to my Realaudio playlist:<http://hmrl.cancer.med.umich.edu/Rob/index.ssi> Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #10: "Wow. I'm a genius too. I think. BEEP." -Chris Elliott
From: Eric Smalling <Eric_Smalling@amrcorp.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Getting Prelude to Rhapsody Date: Thu, 24 Jul 1997 15:32:28 -0500 Organization: AMR/The SABRE Group Message-ID: <33D7BBDC.2A5A08DC@amrcorp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I remember someone saying you can sign on with Apple as a developer for $250 and you'd get, among other things, the Prelude to Rhapsody package. Can anyone verify this and repost exactly what to sign up for. Thanks! -- ____________________________________________________________________ Eric A. Smalling SABRE TECHNOLOGY SOLUTIONS Ft Worth, Texas USA --=== ------=== The Any views expressed are mine alone and are in no ----------- SABRE way the views of AMR or any of it's subsidiaries. ------=== Group --=== email:Eric_Smalling@amrcorp.com Corp. Web Site: http://www.sabre.com ____________________________________________________________________
From: jbruno@jive.tempo.bell-labs.com (John Bruno) Newsgroups: comp.sys.next.programmer Subject: serial ports Date: 24 Jul 1997 22:11:21 GMT Organization: Bell Laboratories, Lucent Technologies Message-ID: <5r8ju9$t3t10@news.research.bell-labs.com> Keywords: com1, com2 , serial port I am running NeXTStep 4.0 on a pentium PC system (Micron). I am trying to write a C program to access one of the serial ports. Listing the permissions on /dev/ttya shows that I have permission to open /dev/ttya. However, if I try "cat > /dev/ttya" it behaves as if there is no driver. I have been able to start a "getty" on ttya by modifying /etc/ttys and hanging up the init process. I am out of ideas. Can anyone help? Thanks, John Bruno Example of problem: jive: ls -l /dev/ttya crw-rw-rw- 1 root 11, 0 May 11 1996 /dev/ttya jive: cat > /dev/ttya /dev/ttya: No such device jive:
Sender: 18734757@compuserve.com Control: cancel <5r96lc$i7e@news1-alterdial.uu.net> Message-ID: <cancel.5r96lc$i7e@news1-alterdial.uu.net> Subject: cmsg cancel <5r96lc$i7e@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 25 Jul 1997 03:35:41 GMT EMP article removed by jem@xpat.com. Original Headers: From: 18734757@compuserve.com Subject: FREE SEX SITE Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!news-peer.sprintlink.net!news.sprintlink.net!Sprint!uunet!not-for-mail Lines: 15
Sender: 18734757@compuserve.com Control: cancel <5r96lp$i7g@news1-alterdial.uu.net> Message-ID: <cancel.5r96lp$i7g@news1-alterdial.uu.net> Subject: cmsg cancel <5r96lp$i7g@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 25 Jul 1997 03:33:28 GMT EMP article removed by jem@xpat.com. Original Headers: From: 18734757@compuserve.com Subject: GET FREE 2100 $ex-Web$Sites FREE! @? Newsgroups: comp.sys.next.programmer Path: ...!howland.erols.net!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!uunet!not-for-mail Lines: 99
Sender: 9876555554453@compuserve.com Control: cancel <5r9c3n$prs@news1-alterdial.uu.net> Message-ID: <cancel.5r9c3n$prs@news1-alterdial.uu.net> Subject: cmsg cancel <5r9c3n$prs@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 25 Jul 1997 05:03:54 GMT EMP article removed by jem@xpat.com. Original Headers: From: 9876555554453@compuserve.com Subject: Don't U pay 4 sex $ite password Newsgroups: comp.sys.next.programmer Path: ...!infeed1.internetmci.com!newsfeed.internetmci.com!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!uunet!not-for-mail Lines: 21
From: rainer@mathematik.uni-wuerzburg.de (Rainer =?ISO-8859-1?Q?Frohnh=F6fer)?= Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 25 Jul 1997 01:09:27 GMT Organization: University of =?ISO-8859-1?Q?W=FCrzburg,?= Germany Message-ID: <5r8uc7$el@lobotomy.urz.uni-wuerzburg.de> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> <nagleEDIw6C.HGy@netcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Cc: nagle@netcom.com In <nagleEDIw6C.HGy@netcom.com> John Nagle wrote: > bradley@apple.com (Bob Bradley) writes: > >In article <19970717190420809661@accs-as07-dp12.dlls.grid.net>, > The near future of routine programming seems to be Visual Basic, with > purchased objects for the hard parts. The Mac still doesn't have an answer > to that. Ahem. Palettes & InterfaceBuilder.app in OpenStep (no:OPENSTEP)? When (if?) the Java integration is finished things could really get interesting ... ------------------------------------- "Um Energie zu sparen, wird das Licht am Ende des Tunnels vorlaeufig abgeschaltet." rainer@mathematik.uni-wuerzburg.de (public key avaible at any key server near you ...)
From: Wesley Horner <wesman@azrael.uoregon.edu> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Getting Prelude to Rhapsody Date: 25 Jul 1997 06:05:42 GMT Organization: University of Oregon, Eugene Message-ID: <5r9fnm$9ur@pith.uoregon.edu> References: <33D7BBDC.2A5A08DC@amrcorp.com> It's true. That is one of the many benifits of being a developer. The website said somthing along the lines of while supplies last for the prelude package but they still have them I got mine a few weeks ago. go to http://www.devworld.apple.com the pdf's of the forms to sighn up are all there. In comp.sys.next.advocacy Eric Smalling <Eric_Smalling@amrcorp.com> wrote: > I remember someone saying you can sign on with Apple as a developer for > $250 and you'd get, among other things, the Prelude to Rhapsody > package. Can anyone verify this and repost exactly what to sign up for. > Thanks! > -- > ____________________________________________________________________ > Eric A. Smalling SABRE TECHNOLOGY SOLUTIONS > Ft Worth, Texas USA --=== > ------=== The > Any views expressed are mine alone and are in no ----------- SABRE > way the views of AMR or any of it's subsidiaries. ------=== Group > --=== > email:Eric_Smalling@amrcorp.com > Corp. Web Site: http://www.sabre.com > ____________________________________________________________________ -- ~~~~wesman@gladstone.uoregon.edu~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Vax a viscious creature known to eat 110AC and quotes through its *DCL*. Vax are usually found in groups of Vaxen called clusters where they lay in wait to ravage thier prey known as users.
Date: Thu, 24 Jul 1997 23:11:33 -0600 From: mlaster@galstar.com Subject: Patch to fix GNU diff 2.7 on OpenStep 4.2 Newsgroups: comp.sys.next.programmer Message-ID: <869801938.32573@dejanews.com> Organization: Deja News Posting Service While trying to install the latest difftools, rcs and cvs on my OpenStep box today, I was plagued by a random segmentation fault while trying to check a file in. I tracked it down to a problem with diff. It seems that it dies when you attempt to compare a file to standard input (as checkin does). After digging through the code for a while, I have come up with a workaround. I still don't understand exactly WHY this patch makes it work, but it does nonetheless (on my machine). If you have noticed similar problems with diff, give this patch a try and see if it fixes it for you. *** ../diffutils-2.7/io.c Thu Sep 22 11:47:00 1994 --- io.c Wed Jul 23 22:06:11 1997 *************** *** 669,674 **** --- 669,679 ---- int skip_test = always_text_flag | pretend_binary; int appears_binary = pretend_binary | sip (&filevec[0], skip_test); + /* weird OpenStep/Mach bugfix -- MCL */ + #ifdef NeXT + filevec[0].name = filevec[0].name; + #endif + if (filevec[0].desc != filevec[1].desc) appears_binary |= sip (&filevec[1], skip_test | appears_binary); else -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: willi@marvin.aaa-plus.com (Wilhelm Pitzeier) Newsgroups: comp.sys.next.programmer Subject: Fax to given number Date: 25 Jul 1997 09:46:27 GMT Organization: Customer of EUnet/PING Austria Message-ID: <5r9slj$lj9$1@news.Austria.EU.net> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi! How do I open a Fax panel with the name and the number of my company already in it? Any hints welcome, - Willi -- ******************* Dr. Wilhelm Pitzeier ******************* ***** AAA+ Software Forschungs- und Entwicklungs-GmbH ****** ****** A-1010 Vienna, Austria -- Rotenturmstrasse 22 ******* ******** E-Mail: willi@aaa-plus.com (NEXT/MIME/ISO) ******** * ISDN Voice: (+43 1) 5336665/32 **** Fax: (+43 1) 5336890 * ************************************************************
From: gtupar@ctp.com (Georg Tuparev) Newsgroups: comp.sys.next.programmer Subject: Re: Patch to fix GNU diff 2.7 on OpenStep 4.2 Date: 25 Jul 1997 09:59:30 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5r9te2$1qo$1@concorde.ctp.com> References: <869801938.32573@dejanews.com> I cake up with similar solution. But it's easier to use the /usr/bin/diff ;-) -- georg -- In article <869801938.32573@dejanews.com> mlaster@galstar.com writes: > While trying to install the latest difftools, rcs and cvs on my > OpenStep box today, I was plagued by a random segmentation fault > while trying to check a file in. I tracked it down to a problem > with diff. It seems that it dies when you attempt to compare a > file to standard input (as checkin does). After digging through > the code for a while, I have come up with a workaround. I still > don't understand exactly WHY this patch makes it work, but it > does nonetheless (on my machine). -- ------- /\/\ Georg Tuparev <georg_tuparev@ctp.com> | Currently in Dublin / /_ \ Cambridge Technology Partners 118/119 Lower Baggot Street \ / / Apollo House, Apollolaan 15 Dublin 2, Republic of Ireland \/\/ 1077 AB Amsterdam, The Netherlands Tel: +353(1)607-9083 Tel: +31(20)575-0492 Fax: +31(20)575-0500 WWW: http://www.ctp.com
From: rdi@interlog.com (Rick Innis) Newsgroups: comp.sys.next.programmer Subject: gcc 2.7.2.2 on NS 3.3 slab Followup-To: poster Date: 25 Jul 1997 08:01:56 -0400 Organization: InterLog Internet Services (416) 975-2655 info@interlog.com Message-ID: <5ra4jk$au7@shell1.interlog.com> I downloaded and installed the compiled version of gcc 2.7.2.2 from peak, and I'm having some interesting problems with it. Specifically,, when running configure for kaffe 0.8.4, I'm getting lots of errors like this: ld: multiple definitions of symbol ___builtin_vec_new /usr/local/lib/gcc-lib/m68k-next-nextstep3/2.7.2.2.f.2/libgcc.a(_op_vnew.o) definition of ___builtin_vec_new in section (__TEXT,__text) /lib/libsys_s.a(builtin_New.o) definition of absolute ___builtin_vec_new (value 0x50041ba) It looks like a library clash to me, but I'm not sure how to resolve it. Any advice would be appreciated. Thanks, --Rick.
From: smolny@pankow.mpiib-berlin.mpg.de (Bertram Smolny) Newsgroups: comp.sys.next.software,comp.sys.next.misc,comp.sys.next.programmer Subject: Q:SAMBA and faxing Date: 25 Jul 1997 10:13:49 GMT Organization: GWDG, Goettingen Message-ID: <5r9u8t$f69$1@gwdu19.gwdg.de> -- Hi, has anyone solved the problem: faxing by NXFax and exporting this service by Samba to faxing in a misc network ? plaese send e-mail. thanx. >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> Bertram Smolny Fon: ++49 30/2802-6154 MPI f. Infektionsbiologie Fax: ++49 30/2802-6406 Monbijoustr.2 D-10117 Berlin e-mail: smolny@mpiib-berlin.mpg.de >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
From: oschirr@abm12.abm.de (Oliver Schirrmeister) Newsgroups: comp.sys.next.programmer Subject: Re: C Question Date: 25 Jul 1997 12:21:14 GMT Organization: Nacamar Data Communications Message-ID: <5ra5nq$g04$1@news.nacamar.de> References: <5r5mqm$pdn$1@parlor.HiWAAY.net> Cc: kamran@tybrin3.hsv.tybrin.com In <5r5mqm$pdn$1@parlor.HiWAAY.net> Kamran Talai wrote: > Is there a way to declare an unsized array with elements of arbitrary size? > Specifically, I am trying to declare a pointer to an array of structures to > be malloc'ed during program execution. Is there a way to have the size of > the elements made known in the declaration so that an element can be pointed > to solely by its index number (without having to multiply the index number by > the sizeof the structure)? You can use an array of pointers to your structures. > I apologize if this question is too broad or basic for this forum. I think > Microsoft specific compilers do allow declaration of unsized arrays and so I > didn't want to be flamed in one of those forums. > > Thanks in advance for any and all replies. > >
From: mobbster@violet.ucsb.edu (Paul T. Mobbs) Newsgroups: comp.sys.next.programmer Subject: EOF databases? Date: 24 Jul 1997 23:24:22 -0700 Organization: University of California, Santa Barbara Message-ID: <5r9gqm$p71@violet.ucsb.edu> Can anyone suggest the simplest cheapest SQL database that will interface nicely with EOF? If it runs on Mach that is a definite plus, but if it runs on NT that is OK, too. Anybody have any luck with EOF + Sybase SQL Server on NT? Sybase is a big bloated bear if you ask me... and slow as all heck, y'know? Thanks in advance. -- ----------------------------------------------------------------- Paul Mobbs mobbster@ecimail.ucsb.edu UCSB Engineering
From: igb@patton-tm.East.Sun.COM (Iain Bason - SMCC High Performance Computing) Newsgroups: comp.sys.next.programmer Subject: Re: Is bcopy/memmove thread-safe? Date: 25 Jul 1997 14:50:08 GMT Organization: Sun Microsystems Inc. - BDC Message-ID: <5raef0$efc@walters.East.Sun.COM> References: <5qoflr$3ra$1@nnrp.cs.ubc.ca> <Ano5BvS00UzxA2prRs@andrew.cmu.edu> In article <Ano5BvS00UzxA2prRs@andrew.cmu.edu>, Charles William Swiger <cs4w+@andrew.cmu.edu> wrote: >Excerpts from netnews.comp.sys.next.programmer: 18-Jul-97 Is >bcopy/memmove thread-safe? by Chris Roehrig@cs.ubc.ca >> Under NEXTSTEP 3.3. The NeXT docs say that malloc "and its related >> functions" are thread-safe. Does this include memmove? > >I don't think so, no. Without knowing anything about the implementation of memmove under NEXTSTEP, I'd still say it's pretty likely that memmove itself is thread-safe. Any function that uses only local variables is inherently thread-safe, and I can't think of any reason why anyone would want to write memmove using static or global variables. However, memmove can't provide any protection to the data being moved. Iain
From: Rich Schroedel <richs@win.bright.net> Newsgroups: comp.sys.next.programmer Subject: Re: Rhapsody development options Date: Fri, 25 Jul 1997 11:23:19 -0600 Organization: BrightNet Wisconsin Message-ID: <33D8E103.908@win.bright.net> References: <5rag88$5ud$1@sps1.phys.vt.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Nathan Urban wrote: > > I currently have OPENSTEP for Mach 4.0 for Intel. I had been thinking of > upgrading to 4.2, but considering that Rhapsody is my intended future > audience, I'd rather prefer to get the Rhapsody Developer Release, > through the $250/yr Macintosh Associates plan. I'm not sure what is > included with that, though. > > 1) If I sign up for Macintosh Associates, can I get Rhapsody DR1 for > Intel, or only PowerMac? And am I guaranteed to get Rhapsody DR1 if I sign > up for the program, or is this a "while supplies last" thing? > > 2) Will Rhapsody DR1 come with the standard cc compiler, ProjectBuilder, > and InterfaceBuilder, or will they cost extra? > > (Has anyone noticed how difficult it is to find the answers to these > simple questions on devworld.apple.com?) Excellent questions! If you are posted an answer through private e-mail, please post it to the list. Thanks. -- Rich Schroedel "There is only one success... Ondossagon Software to live your life in your own way" richs@win.bright.net Christopher Marlowe
From: scott@doubleu.com (Scott Hess) Newsgroups: comp.sys.next.programmer Subject: Re: cc -pipe Date: 24 Jul 1997 13:01:13 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jul24082657@slave.doubleu.com> References: <Pine.NXT.3.96.970723234217.4017A-100000@cc344191-a.ewndsr1.nj.home.com> In-reply-to: Timothy Luoma's message of Wed, 23 Jul 1997 23:56:53 -0400 In article <Pine.NXT.3.96.970723234217.4017A-100000@cc344191-a.ewndsr1.nj.home.com>, Timothy Luoma <luomat+next+usenet@luomatpeak.org> writes: I just built the pine suite (pine, pico, imapd, etc) quadfat for NS on a P-133 with 64 ram and a lot of swapspace free. First I ran it without the '-pipe' flag and got this: ./build nxt 1109.48s user 104.87s system 67% cpu 29:56.27 total as the results... it took about 30 minutes to compile.... So I ran it again (on a clean set of source code) this time WITH the -pipe flag (and no other changes) and got this: ./build nxt 1114.88s user 99.72s system 68% cpu 29:21.42 total and it took about 30 minutes to compile. I was hoping the -pipe was going to speed things up.... perhaps there is something about the pine config that made it not noticeably different.... Since -pipe is really trading memory usage for disk usage, it might be that the quadfat nature of the compile got in the way. 64M should be plenty for this, assuming that you aren't running in fullcolor mode or something. I found swapping to be annoying with 32-bit color and 64M of RAM. Since I couldn't really tell the difference between 32-bit color and 16-bit color, and on my Matrox Millenium they both got the same benchmarks in NXBench, I dropped to 16-bit color. Also, keep in mind that the better your disk subsystem is, the less good -pipe is going to do you. Back when you had a 1.5MB/s subsystem on a NeXTstation, it made a lot more difference than when you've got a 5MB/s (or better) subsystem on a modern machine. My results are generally from 5% to 10% faster with -pipe than without. Is there some other program that might help show if there is a difference or not? BackSpace is the "standard" compile test. Is there some other way to speed compiles? Faster disks. Object files on a different disk than the source (if you have a choice, put the object files on the disk with better seek times. NS can cache the source files for reading, but object files have to be written). Faster CPU :-). More memory probably won't help much in this case. Later, -- scott hess <scott@doubleu.com> (606) 578-0412 http://www.doubleu.com/ <Favorite unused computer book title: The Idiots Guide to the Zen of Dummies in a Nutshell in Seven Days, Unleashed>
From: Wolfgang Baron <wbaron@ixpoint.de> Newsgroups: comp.sys.next.programmer Subject: Re: Patch to fix GNU diff 2.7 on OpenStep 4.2 Date: 25 Jul 1997 06:24:40 GMT Organization: iXpoint Informationssysteme GmbH, www.ixpoint.de Message-ID: <5r9gr8$1n0$1@ixpoint.de> References: <869801938.32573@dejanews.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Cc: mlaster@galstar.com In <869801938.32573@dejanews.com> mlaster@galstar.com wrote: > After digging through > the code for a while, I have come up with a workaround. I still > don't understand exactly WHY this patch makes it work, but it > does nonetheless (on my machine). > [...] You might just as well compile it with -O2 !!! It's faster and it works ... -- +++++++++++++++++++++++++++++++++++++++++++++++++ + Dipl.- Inform. Wolfgang Baron + + iXpoint Informationssysteme GmbH + + Daimlerstrasse 3, 76275 Ettlingen, Germany + + phone ++49 7243 3775-82 Fax ++49 7243 3775-77 + + email: wbaron@ixpoint.de (NeXTmail and MIME) + +++++++++++++++++++++++++++++++++++++++++++++++++ + pConnector - Network Printing for NEXTSTEP + + sca - the static code analyser + +++++++++++++++++++++++++++++++++++++++++++++++++
Newsgroups: comp.sys.next.programmer From: brianw@sounds.wa.com (Brian Willoughby) Subject: Help.store vs. bundle Message-ID: <EDwDBq.5wI.0.scream@sounds.wa.com> Organization: Sound Consulting, Bellevue, WA, USA Date: Fri, 25 Jul 1997 23:02:13 GMT I am building a bundle under NEXTSTEP 3.3 which has Help files in its English.lproj directory. The bundle works, but the Help files do not show up. Apparently, the problem is that ProjectBuilder and the standard NeXT Makefiles compress the Help files into a Help.store index, and these seem to be unloadable in a bundle situation. If I delete the Help.store file and manually copy the original Help files, then they are all added to the application when the bundle is loaded. I remember reading about someone else having similar problems, either in this newsgroup, or on the NeXTprog mailing list. Does anyone know about this? Is there a way to create a Help.store file which will be adding when a bundle is loaded? or is the correct approach to circumvent the Help compression and just copy the files? Thanks for any info, -- Brian Willoughby NEXTSTEP, OpenStep, Rhapsody Software Design Sound Consulting Apple Enterprise Alliance Partner NeXTmail welcome
Newsgroups: comp.sys.next.programmer,comp.sys.next.hardware From: brianw@sounds.wa.com (Brian Willoughby) Subject: Re: HELP with SCSI device problem Message-ID: <EDwE9L.63C.0.scream@sounds.wa.com> Organization: Sound Consulting, Bellevue, WA, USA References: <5qpgpt$sgs$2@excalibur.flash.net> <5quo2q$epf$1@enyo.uwa.edu.au> Date: Fri, 25 Jul 1997 23:22:33 GMT I missed the original article, but would still like to make a suggestion: Have you looked at the Generic SCSI Device Driver that NeXT provides? This is documented in the Man Pages under SG(4), and has been used to access SCSI scanners and SCSI drives with foreign file systems. Its much easier to use than writing an indirect driver, although the latter should also work. In <5qpgpt$sgs$2@excalibur.flash.net> daj@nwu.edu wrote: > I have a SCSI data acquisition device that was manufactured by Analogic, its model DASM-AD14. Technically it is a 14bit 2Mhz 8 channel acquisition device that has a 2Mb buffer and a SCSI interface. > [snip] > So to make a long story shorter than what it could be, is the above note true or is there some patch I can apply to allow me to write a device driver under Openstep 4.1? Can I get NEXTSTEP3.3 Developer and use it under OPENSTEP 4.1? Last, I'm still not clear in understanding why I can't just use /dev/rsd4c as the Analogic manual suggests. The requirement as Analogic has it in there manual is that their DASM-AD14 communicates through a character-special file (ie /dev/rsd4c). Clearly the device appears unwritable (in messages below), however it should be writable. Can anyone explain why I see this problem? > > Oh yeah, the last note is that Analogic no longer supports the device, it appeared to be to costly to produce so there wasnt enough demand. As a result I have virttually no help from Analogic. -- Brian Willoughby NEXTSTEP, OpenStep, Rhapsody Software Design Sound Consulting Apple Enterprise Alliance Partner NeXTmail welcome
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 25 Jul 1997 18:29:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFFEA5B2-402C7@206.165.44.29> References: <5rav8v$ek@shelob.afs.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit [I dared suggest that OpenDoc would be easier to use for the end-user than InterfaceBuilder] Gregory H. Anderson <Greg_Anderson@afs.com> replied: > After my 12-year-old daughter had a few weeks of training in HyperStudio > this year at school, she had no trouble conceptually with how to make IB > do exactly the same thing (dragging copies of things off palettes and > hooking them together). And with smart enough parts, there doesn't have > to be a whole lot more to it than that to build non-complex applications. > After all, Steve Jobs -- certainly NOT a programmer by anyone's standards > -- gave literally dozens of convincing demos in IB with only NeXT's basic > parts. You should see some of the stuff we've built internally. Er, yeesssss... Never used OpenDoc, now have we? Typical of the arrogance that I've come to expect of NeXT developers. So clueless about everything non-NeXT that they haven't a clue how clueless they are. [note: this is NOT to say that IB isn't the ultimate cool for what it does, but to suggest that "a few weeks training" is necessary to provide the same user-experience as OpenDoc is to demonstrate that the writer hasn't ANY idea what OpenDoc is like or what it does] ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: 07685@compuserve.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5rh3hi$bea@news1-alterdial.uu.net> Control: cancel <5rh3hi$bea@news1-alterdial.uu.net> Date: 28 Jul 1997 03:26:46 GMT Message-ID: <cancel.5rh3hi$bea@news1-alterdial.uu.net> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: scott@doubleu.com (Scott Hess) Newsgroups: comp.sys.next.programmer Subject: Re: #import not working properly? Date: 28 Jul 1997 06:00:47 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jul28013607@slave.doubleu.com> References: <B0002495-F62D55@141.214.134.235> In-reply-to: "Robert A. Decker"'s message of 26 Jul 97 22:02:56 -0400 In article <B0002495-F62D55@141.214.134.235>, "Robert A. Decker" <comrade@umich.edu> writes: If I have two headers, A.h and B.h: Here is A.h: ****************** #import "B.h" @interface A:NSObject B *staticB; @end ****************** and here is B.h: ******************* #import "A.h" @interface B:NSObject A *staticA; @end ****************** When I build I get errors in A.h saying B is an undefined type and errors in B.h saying that A is an undefined type. The way I understand how #import works is that it makes sure that a header is only included once, and probably also stops circular references to headers. So it looks like as it's working on A.h it imports B.h, then when it gets to B.h, it sees there's already an A.h included, but for some reason throws out both A.h and B.h?! This is an easy chestnut. When you're including A.h somewhere, it includes B.h. Since A.h is already being included, B.h doesn't include it again. At the point where B.h was imported, the class A had not yet been defined, and thus the class B cannot be defined. The solution is to either use id instead of (B *), or better yet to say @class A (or @class B) to define the class you need. Personally, I strive to keep the breadth of my imports to a minimum, because the more files you import, the more likely you are to find a conflict. So in the above I would likely have @class A replacing #import "A.h", and likewise with @class B replacing #import "B.h". That keeps A.h from being imported into every single import of B.h. Then, any importer of B.h which also needs A.h also imports A.h explicitely. If all my .m files have the same four #imports at the top, I wonder why and fix it (by correctly encapsulating things), or if I'm lazy I just put a file named "*Internals.h" or somesuch to do those imports for me. Later, -- scott hess <scott@doubleu.com> (606) 578-0412 http://www.doubleu.com/ <Favorite unused computer book title: The Idiots Guide to the Zen of Dummies in a Nutshell in Seven Days, Unleashed>
From: scott@doubleu.com (Scott Hess) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 28 Jul 1997 06:00:42 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jul28013018@slave.doubleu.com> References: <5rav8v$ek@shelob.afs.com> <AFFEA5B2-402C7@206.165.44.29> In-reply-to: "Lawson English"'s message of 25 Jul 1997 18:29:01 -0700 In article <AFFEA5B2-402C7@206.165.44.29>, "Lawson English" <english@primenet.com> writes: Typical of the arrogance that I've come to expect of NeXT developers. So clueless about everything non-NeXT that they haven't a clue how clueless they are. From the mouths of <censored>... -- scott hess <scott@doubleu.com> (606) 578-0412 http://www.doubleu.com/ <Favorite unused computer book title: The Idiots Guide to the Zen of Dummies in a Nutshell in Seven Days, Unleashed>
From: mmalcolm crawford <malcolm@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 28 Jul 1997 10:13:41 GMT Organization: P&L Systems Message-ID: <5rhrcl$p6i$1@ironhorse.plsys.co.uk> References: <33DA91BA.211E@afs.com> <B000476E-4B967@206.165.44.80> In-Reply-To: <B000476E-4B967@206.165.44.80> On 07/27/97, "Lawson English" wrote: >I'm sorry. I didn't realize that you could do something like surf the >internet , while typing a letter, and then print it out, using the current >web-page as the illustration for the letter, from WITHIN IB itself... > Well, ignorance has never stopped you making defamatory comments about NeXT's technology in the past, has it. There is no reason at all why you could not do this in principle; in practice I'm not sure how well advanced the HTMLView object is yet, or whether it will be included in the Rhapsody DR1 release. Note that web-browsing is only one activity which involves real-time updates, and distracts from the main point that, yes, "runtime linking of live data (including live updates over the Internet)" is supported by IB, and has been for a long time. I presume you can take a Reuters feed and display stock quotes in real time using uncompiled GX too? It doubtless translates all currency units to Yen in Pantone-269 on the fly as well? Best wishes, mmalc. --
From: 987655555453@compuserve.com Newsgroups: comp.sys.next.programmer Subject: Pa$$Word TO 1200 sex sites only $25! Date: 16 Jul 1997 08:21:29 GMT Organization: .PASSWORD INC... Message-ID: <5qi0a9$emg@news1-alterdial.uu.net> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary=_-=314 This is a multi-part message in MIME format. --_-=314 Content-type: text/html; name="Sex~1.htm" Content-transfer-encoding: base64 PCFET0NUWVBFIEhUTUwgUFVCTElDICItLy9JRVRGLy9EVEQgSFRNTC8vRU4iPg0KDQo8aHRt bD4NCg0KPGhlYWQ+DQo8dGl0bGU+U0VDUkVUIFBBU1NXT1JEIEZPUiBBRFVMVCBTSUdIVFM8 L3RpdGxlPg0KPG1ldGEgbmFtZT0iR0VORVJBVE9SIiBjb250ZW50PSJNaWNyb3NvZnQgRnJv bnRQYWdlIDEuMSI+DQo8L2hlYWQ+DQoNCjxib2R5IGJnY29sb3I9IiNDMEMwQzAiIHRleHQ9 IiM4MEZGRkYiIGxpbms9IiMwMDAwRkYiIHZsaW5rPSIjMDAwMEZGIiBhbGluaz0iI0ZGMDAw MCI+DQo8ZGl2IGFsaWduPWNlbnRlcj48Y2VudGVyPg0KPHRhYmxlIGJvcmRlcj01IGNlbGxw YWRkaW5nPTI+DQo8dHI+PHRkIGFsaWduPWNlbnRlciB3aWR0aD0xMDAlPjxmb250IGNvbG9y PSIjRkZGRjAwIj48Zm9udCBzaXplPTc+PGI+RlJFRSEgRlJFRSEgRlJFRSE8YnI+DQo8L2I+ PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT01PjxiPlRI QVQnUyBSSUdIVC4uLiBDT01QTEVURUxZIDwvYj48L2ZvbnQ+PGZvbnQgc2l6ZT02PjxiPkZS RUUhITwvYj48L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjMDAwMEZGIj48Zm9udCBzaXpl PTU+PGI+PGk+PGJyPg0KRU5URVIgVEhST1VHSCBBRFVMVFNJR0hUUyAmIzE2OTsgPC9pPjwv Yj48L2ZvbnQ+PGZvbnQgc2l6ZT02Pjxicj4NCjwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9 IiNGRkZGMDAiPjxmb250IHNpemU9Nz48Yj5JVCdTIEZSRUUhISEhITwvYj48L2ZvbnQ+PC9m b250PjwvdGQ+PC90cj4NCjwvdGFibGU+DQo8L2NlbnRlcj48L2Rpdj4NCjxwIGFsaWduPWNl bnRlcj48Zm9udCBGQUNFPSJDb21pYyBTYW5zIE1TIj48Zm9udCBjb2xvcj0iIzIyMTY1QSI+ PGZvbnQgc2l6ZT03PiA8L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjRkYwMDAwIj48Zm9u dCBzaXplPTc+T05FIFBBU1NXT1JEIDxmb250IEZBQ0U9IkNvbWljIFNhbnMgTVMiPjxicj4N CjwvZm9udD48L2ZvbnQ+PC9mb250Pjxmb250IGNvbG9yPSIjMjIxNjVBIj48Zm9udCBzaXpl PTU+IDwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9IiNGRkZGMDAiPjxmb250IHNpemU9NT48 c3Ryb25nPkdFVFMgWU9VIFVOTElNSVRFRCBGUkVFIEFDQ0VTUyBUTyBPVkVSIA0KMTgwMCsg T0YgVEhFIEhPVFRFU1QgQURVTFQgU0lURVMgPC9zdHJvbmc+PC9mb250PjwvZm9udD48L3A+ DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29sb3I9IiMwMDAwMDAiPjxmb250IHNpemU9NT48 c3Ryb25nPklGIFlPVSBBUkUgVElSRUQgT0YgUEFZSU5HIFRPIEdFVCBJTlRPIFNFWCANClNJ VEVTLi4uPC9zdHJvbmc+PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGRkZGRiI+PGZv bnQgc2l6ZT01PjxzdHJvbmc+IDwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PGZvbnQgY29sb3I9 IiMwMEZGRkYiPjxmb250IHNpemU9NT5PbmUgUGFzc3dvcmQgaXMgYWxsIHlvdSBuZWVkISEh PGJyPg0KPC9mb250PjwvZm9udD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgRkFDRT0i Q29taWMgU2FucyBNUyI+PGZvbnQgY29sb3I9IiMyMjE2NUEiPjxmb250IHNpemU9Nj4gPC9m b250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT02PjxzdHJvbmc+ TE9PSyBXSEFUIFlPVSBHRVQhPC9zdHJvbmc+PC9mb250PjwvZm9udD48L3A+DQo8cCBhbGln bj1jZW50ZXI+PGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PGZvbnQgY29sb3I9IiMyMjE2 NUEiPjxmb250IHNpemU9NT4gPC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iI0ZGRkYwMCI+ PGZvbnQgc2l6ZT01PjxzdHJvbmc+VU5MSU1JVEVEIEZSRUUgQUNDRVNTIFRPIE9WRVIgMTgw MCsgT0YgDQpUSEUgSE9UVEVTVCBBRFVMVCBTSVRFUzwvc3Ryb25nPiA8Zm9udCBGQUNFPSJD b21pYyBTYW5zIE1TIj48L2ZvbnQ+PC9mb250PjwvZm9udD48Zm9udCBjb2xvcj0iIzIyMTY1 QSI+PGZvbnQgc2l6ZT01PiA8L2ZvbnQ+PC9mb250PjwvcD4NCjxoMSBhbGlnbj1jZW50ZXI+ PGZvbnQgY29sb3I9IiNGRjAwMDAiPjxmb250IHNpemU9NT48c3Ryb25nPkp1c3QgQSBGZXcg T2YgVGhlIE92ZXIgMTgwMCsgWFhYIFJhdGVkIFNpZ2h0cyBUaGF0IA0KWW91IENhbiBFbnRl ciBDb21wbGV0ZWx5IEZyZWUhPGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PC9mb250Pjwv c3Ryb25nPjwvZm9udD48L2ZvbnQ+PC9oMT4NCjxoMSBhbGlnbj1jZW50ZXI+PGZvbnQgY29s b3I9IiM4MEZGMDAiPjxmb250IHNpemU9Nj48c3Ryb25nPlRFRU4gU0lURVM8L3N0cm9uZz48 L2ZvbnQ+PC9mb250PjwvaDE+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29sb3I9IiMwMDAw MDAiPjxmb250IHNpemU9ND4xMDAwIEhPVCBBTkQgTmFLRUQgLSBDZUxlQlJpVGlFUyBBTkQg VEVFTlMgLSAxMDAxIE51ZGUgRXJvdGljIA0KVGVlbnMgLSAyMCBBbWFUZWVucyAtIDYyMyBO dWRlIENlbGVicml0aWVzIEFuZCBUZWVucyAtIEFBQSBTTFVUWSANClRFRU5TIC0gQW1hdGV1 ciBUZWVucyAtIEFtYXRldXIgVGVlbnMgRnJvbSBBcm91bmQgVGhlIFdvcmxkIC0gQW1hdGV1 ciANClRlZW5zISAtIENvY2sgSHVuZ3J5IFRlZW5zIC0gQ3VtIERyaW5raW5nIFRlZW5zIC1G aXN0IEZ1Y2tpbmcgVGVlbnMgLSBIb3QgDQpOdWRlIFRlZW5zIC0gTm90aGluZyBCdXQgVGVl bnMgLSBUZWVucyBJbiBUcm91YmxlIC0gVHdhdCBTdWNraW5nIFRlZW5zIDwvZm9udD48L2Zv bnQ+PC9wPg0KPHAgYWxpZ249Y2VudGVyPjxmb250IGNvbG9yPSIjRkYwMDAwIj48Zm9udCBz aXplPTY+PHN0cm9uZz5BU0lBTiBTSUdIVFM8L3N0cm9uZz48L2ZvbnQ+PC9mb250PjwvcD4N CjxwIGFsaWduPWNlbnRlcj48Zm9udCBjb2xvcj0iIzAwMDAwMCI+PGZvbnQgc2l6ZT00PkFs bCBBc2lhbiBUZWVuIC0gQXNpYW4gQmFiZXMgLSBBc2lhbiBCYWJlcyBzaWFuIEJhYmVzISAt IEFzaWFuIEJlbGxlcyAtIEFzaWFuIA0KQm9uZGFnZSAtIEFzaWFuIEN1dGUgR2lybHMgLSBB c2lhbiBEZWxpdGVzIC0gQXNpYW4gRXJvcyAtIEFzaWFuIEZsb3dlcnMgLSANCkFzaWFuIEdv bGQgLSBZb3VuZyBBc2lhbiBXb21lbiAtIEFzaWFuIEhhcmRDb3JlIEFTSUFOIEhBUkRDT1JF IC0gDQpUSEUgTEVHRU5EIENPTlRJTlVFUyAtIEFTSUFOIEhvbG9zdWl0ZSAtIEFzaWFuIExv dmUgRG9sbHMgPC9mb250PjwvZm9udD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgY29s b3I9IiNGRjAwRkYiPjxmb250IHNpemU9Nj48c3Ryb25nPkNFTEVCUklUWSBTSUdIVFM8L3N0 cm9uZz48L2ZvbnQ+PC9mb250PjwvcD4NCjxwIGFsaWduPWNlbnRlcj48Zm9udCBjb2xvcj0i IzAwMDAwMCI+PGZvbnQgc2l6ZT00PiMxIE51ZGUgQ2VsZWJyaXR5IFBhZ2UgISAtIENlbGVi d29ybGQgQWNlJ3MgLSBYWFggQ2VsZWJyaXR5IFBpY3MgLSBDZWxlYnJpdHkgDQpDZW50cmFs IC0gQ2VsZWJyaXR5IENoaWNrIFBhdHRpZXMgLSBDZWxlYnJpdHkgQ3VtYm94IC0gQ2VsZWJy aXR5IFBpY3R1cmVzIA0KQ2VsZWJyaXR5IFBpY3R1cmVzIEdvbGQgU2l0ZSAtIENlbGVicml0 eSBTa2luIC0gQ2VsZWJyaXR5IFNraW4tQUJDIC0gRGEgUm9ja3MgDQpudWRlIGNlbGVicml0 eSBBcmNoaXZlIC0gTWFya3MgQ2VsZWJyaXR5IFBpY3R1cmVzIDIgLSBOdWRlIENlbGVicml0 eSBBcmNoaXZlIC0gDQpTdGFyTG9jYXRvciAtIENlbGVicml0eSBTdXBlclMgLSBUaGUgRGFp bHkgTnVkZSBDZWxlYnJpdHkgUGFnZSAtIFRoZSBNdXN0IA0KU2VlIE5VREUgQ2VsZWJyaXR5 IFNpdGUgLSBUaGUgVWx0aW1hdGUgQ2VsZWJyaXR5IENvbGxlY3Rpb24gPC9mb250PjwvZm9u dD48L3A+DQo8cCBhbGlnbj1jZW50ZXI+PGZvbnQgRkFDRT0iQ29taWMgU2FucyBNUyI+PGZv bnQgY29sb3I9IiNGRjRGQTciPjxmb250IHNpemU9NT4gPC9mb250PjwvZm9udD48Zm9udCBj b2xvcj0iIzAwODBGRiI+PGZvbnQgc2l6ZT01PjxzdHJvbmc+VU5MSU1JVEVEIEZSRUUgQUND RVNTIFRPIE9VUiBYWFggUkFURUQgDQpDSEFUIFJPT008L3N0cm9uZz48L2ZvbnQ+PC9mb250 PjwvcD4NCjxwIGFsaWduPWNlbnRlcj48Zm9udCBGQUNFPSJDb21pYyBTYW5zIE1TIj48Zm9u dCBjb2xvcj0iI0ZGMDAwMCI+PGZvbnQgc2l6ZT01PjxzdHJvbmc+R2V0IFlvdXIgQURVTFRT SUdIVFMgQUdFIFZFUklGSUNBVElPTiANCk1FTUJFUlNISVAgRk9SIE9OTFkgJDEyLjk1LiBJ dHMgR29vZCBGb3IgQW4gDQpFbnRpcmUgWWVhciBBbmQgR2V0cyBZb3UgVU5MSU1JVEVEIEZS RUUgQWNjZXNzIA0KSW50byAxODAwKyBPZiBUaGUgQmVzdCBBZHVsdCBTaXRlcyBPbiBUaGUg d2ViIEFsbCBZZWFyIA0KQWxsIEZyZWUuIE5ldmVyIFBheSBBbm90aGVyIERpbWUgVG8gR2V0 IEludG8gQSBTZXggU2l0ZSANCkFnYWluLjwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PGZvbnQg Y29sb3I9IiNGRkZGMDAiPjxmb250IHNpemU9NT48YnI+DQo8L2ZvbnQ+PC9mb250PjwvZm9u dD48Zm9udCBjb2xvcj0iI0ZGMDA4MCI+PGZvbnQgc2l6ZT03PjxzdHJvbmc+Qk9OVVMhISE8 YnI+DQo8L3N0cm9uZz48L2ZvbnQ+PGZvbnQgc2l6ZT01PjxzdHJvbmc+RlJFRSBYWFggQ2hh dCBSb29tcyEhITwvc3Ryb25nPjwvZm9udD48L2ZvbnQ+PC9wPg0KPHAgYWxpZ249Y2VudGVy PjxhIGhyZWY9Imh0dHA6Ly8yMDUuMjE2LjI0OS4xOC9jZ2ktc2hsL2RibWwuZXhlP3RlbXBs YXRlPS9hcy9zaWdudXAuZGJtJmFtcDtzaXRlPXN0MzUzMjUiPjxmb250IGNvbG9yPSIjODAw MDgwIj48Zm9udCBzaXplPTY+PHN0cm9uZz5JIFdBTlQgVEhFIFBBU1NXT1JEIE5PVzwvc3Ry b25nPjwvZm9udD48L2ZvbnQ+PC9hPjwvcD4NCjwvYm9keT4NCg0KPC9odG1sPg0K --_-=314
Subject: Re: LISP on Next Newsgroups: comp.sys.next.programmer References: <33CBC2F1.D1EEA930@cris.com> <5qvst0$29r$1@news.freedomnet.com> In-Reply-To: <5qvst0$29r$1@news.freedomnet.com> From: marco@sente.ch (Marco Scheurer) Organization: Sen:te Message-ID: <33dc9f42.0@epflnews.epfl.ch> Date: 28 Jul 97 13:31:46 GMT On 07/21/97, Bill Chin wrote: >krispos@cris.com wrote: >>I remember reading somewhere that the NeXT cubes had a built in LISP >>compiler. However apparently the color slabs do not, and I was [...] > >There are, however, many varieties of Lisp that are freely >available and can be compiled for NEXTSTEP/OPENSTEP. I personally >used XLISP-Plus. Many distributions can be embedded and called >through a C language API, so they can be integrated into an >Objective-C NEXTSTEP/OPENSTEP application. Another one is clisp. Comes with a NEXTSTEP front-end editor. Available from ftp://ma2s2.mathematik.uni-karlsruhe.de/pub/lisp/clisp -- Marco Scheurer (marco@sente.ch) Sen:te
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: How to save Apple Date: Mon, 28 Jul 1997 09:42:05 -0400 Organization: Carnegie Mellon, Pittsburgh, PA Message-ID: <wnr_6h200YUn06O=80@andrew.cmu.edu> References: <AFFE4E6B-52FD4@206.165.44.75> <5rav8v$ek@shelob.afs.com> <SouthSide-2607970238260001@bns.vip.best.com> In-Reply-To: <SouthSide-2607970238260001@bns.vip.best.com> Excerpts from netnews.comp.sys.next.advocacy: 26-Jul-97 Re: How to save Apple by Bob Bradley@kagi.com > > After my 12-year-old daughter had a few weeks of training in HyperStudio > > this year at school, she had no trouble conceptually with how to make IB > > do exactly the same thing (dragging copies of things off palettes and > > hooking them together). And with smart enough parts, there doesn't have > > to be a whole lot more to it than that to build non-complex applications. > > So are you suggesting the use of Interface Builder as a user-level tool? Absolutely. Obviously, you don't realize that a number of scientific/mathematic apps written for NEXTSTEP were not created by software developers, but people who understood their field (not computers) and found that IB and the NEXTSTEP development environment was powerful enough and easy enough to use that they could write elegant GUI programs to do what they wanted. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: dix.lorenz@hamburg.netsurf.de (Dix Lorenz) Newsgroups: comp.sys.next.programmer,comp.sys.next.misc Subject: C++ in framework? Date: Mon, 28 Jul 1997 17:15:09 +0200 Organization: EDV - Beratung Message-ID: <199707281715094377217@dip065-2.hamburg.netsurf.de> Hi, I would like to build a framework with a couple of C++-classes under OS Enterprise 4.2 for NT. I tried the same under OS4.1/Mach and it worked just like that. Under NT I get "unresolved external symbol" errors from the linker when trying to use the framework. If I take the source files from the framework and just use the exact same files in my application directly, it works without a problem. Does anybody have an idea what's going on? If it's of importance: I use the Prelude to Rhapsody. Thanks, Dix
From: acurylo@inmediapresents.nospam.com (Alex Curylo) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Mon, 28 Jul 1997 10:11:12 -0700 Organization: InMedia Presentations Message-ID: <acurylo-2807971011130001@van0229.tvs.net> References: <33DA91BA.211E@afs.com> <B000476E-4B967@206.165.44.80> <5rgrjh$o8u$1@news.xmission.com> In article <5rgrjh$o8u$1@news.xmission.com>, don@globalobjects.com wrote: > This is scary stuff, and few people mess with it, but the point still stands > that the dynamism of the environment allows for all sorts of crazy things > such as modifying running programs (even commercial apps you don't have > source to). Geez, and I thought ResEdit was a pretty neat idea back around the time this stuff first came out :) ---- Alex Curylo -- alex@witty.com
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: OpenDoc vs IB? Why "vs"? (Was Re: How to save Apple Date: 28 Jul 1997 10:16:00 -0700 Organization: Primenet Services for the Internet Message-ID: <B00226E0-6977D@206.165.44.11> References: <5rgrjh$o8u$1@news.xmission.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Don Yacktman <don@globalobjects.com> said: > Basically, IB palettes can do anything you can think of (just like an > Opendoc > part would). So it comes down to whether or not you can get the part you > need, which is the same for either platform. As a framework for putting > the > parts together, as noted earlier, IB is more tailored to developers. But > there's no reason to believe that a "user's IB" couldn't be made either. And > as also noted earlier, there are plenty of non-programmers out there who > find > IB quite easy to work with already. OK, so IB could be extended to do what OD does as a given (in turn, OD could be extended to do what IB does). However, OD is user-oriented from Day One. Which is my point. There's no need for palettes with OD (although they are very often useful in the OD context and some have provided palette parts, I believe). There's no need for 2-3 weeks of training to get a user up to speed with OD. You drag an icon from the disk into a window and that installs a new part automatically which is live by the time the disk stops grinding. You select and drag content from one open OD window to another and that content is copied to the new window, complete with dynamic linking to the underlying object code. You select and drag content from one part of an OD window to another part and that content is copied inside the second object (if it is a container part), either as raw data, or as a part-with-content, according to the Rules of Engagement of OpenDoc. Think of the UI of OpenDoc as being similar to PageMaker's layouts. It isn't just that you can select objects such as text boxes from palettes, but that you can then rearrange those objects in any order by selecting and dragging them around. And OD allows for infinite nesting of parts, each with its own viewport or set of multiple viewports, each with its own 3x3 transform matrix. While OD wasn't explicitly designed for GX, the design of ODF was such that when GX was rolled into Copland, ODF-Copland would be using GX to fulfill all the graphical and UI needs of OpenDoc [obligatory GX plug] because GX so nicely supports the concept of dynamically changing viewports (it is built into the GX implementation of viewports). All sorts of interesting demos of the UI of OpenDoc were devised, such as a 4x4 sliding puzzle that could embedd other OpenDoc parts, even active QuickTime movies. This made that trivial puzzle VERY difficult to solve. Another neat demo of OpenDoc's graphical user interface potential is found in the ODFDraw live object that no longer ships with the OS (how typically politically correct of Apple to not ship anything that actually allows a user to make use of the wiz-bang technology that ships with the Macintosh) but still is available through Hutching Software. ODFDraw allows one to make typical MacDraw 1.0-level drawings AND to embed OpenDoc parts in the same window as though they were drawn by ODFDraw. ODFDraw controls the borders and the clip-shapes and the layering, while the embedded parts provide the content of each shape, be-it an active QT movie, or a web browser, spreadsheet or word processor (or interactive 3D shoot'em up -think about that last for a while). Click on an ODFDraw object, and you get a color palette and other appropriate drawing controls & menus. Click on an embedded part's frame and you get the proper ODFDraw controls & menus for controlling a framed drawing object. Click on an embedded part's interior and you get the controls & menus appropriate for that kind of application (word processor/spreadsheet/web browser/etc). (note that I haven't played with ODFDraw in a while, so the details are from memory). Print an ODFDraw window and each part is charged with providing a version of its content suitable for printing in the designated position and clip-area. When people say that IB does what OD does, I wince. Not just because I believe that OD does one heck of a lot more in most non-programming-oriented ways, but because I believe that IB advocates have missed the point about OD: OD is a *user oriented* MiddleWare system. It provides, system-wide, a set of services and conventions and so on, that are available for the user. Here is a challenge for those who suggest that the Interface Builder would be the appropriate base upon which to build a Rhapsody-OpenDoc: Take a hypothetical jigsaw puzzle part and embedd the Cyberdog browser-part within it. Now surf to your favorite web-page and scramble the puzzle. Now continue surfing and try to unscramble the puzzle when you have no idea what it is that you are looking at. Now embed the scrambled web-browser part inside a spreadsheet, making it a dynamic illustration for the spreadsheet. Now make that functionality work in IB. The basic principle for creating a jigsaw puzzle part is the same as for creating a slider puzzle part. ODFDraw already demonstrates that irregular, overlapping parts are possible, while the slider part demonstrates that providing different views with different clip-shapes of the SAME part are possible. Marrying the two shouldn't prove very difficult at all. Now, are advocates of the position that IB could be extended to allow OD-like functionality willing to admit that this really wouldn't be all that easy, or even practical? Remember that containers in OpenDoc, whether jigsaw puzzles or spreadsheets, can be embedded in other ANY other container part (such as another instance of the jigsaw part which would display a scrambled version of the spreadsheet part containing the jigsaw part containing the scrambled web-browser part, all of which would print out as you would hope). Better would be to port the capabilities of OD to Rhapsody and make IB a specialized part or specialized suite of interacting parts. Better still, and what I advocate, why not take TWO steps backward and look at the capabilities of IB and see where they fill in gaps in the OD capabilities and create an OpenDoc 2.0 specification that implements ALL of the OD capabilities and those IB capabilities that would be of use outside the programming context? That way, the IB parts-suite under OpenDoc 2.0 would have all the capabilities of OpenDoc plus all the capabilities of IB AND other OD 2.0 parts would be able to make use of the current IB's ability to provide graphical representation & control of the linking of objects, but this time oriented for end-users for the linking of completely functional parts, instead of having a primarily programming-oriented tool? Wouldn't THIS be the best of all possible worlds? One could also work on the possibility that Yellow Box applications would have complete OD 2.0 functionality both as containers AND as contained parts. MUCH nicer scenario than what is being proposed for Rhapsody (that I have heard of), don't you agree? ++++++++++++++++++++++++++++++ This space left intentionally blank. +++++++++++++++++++++++++++++++++
From: Erik Doernenburg <erik@object-factory.REMOVE_ME.com> Newsgroups: comp.sys.next.programmer Subject: Changes in NSMutableDictionary's copyWithZone method? Date: 28 Jul 1997 15:45:01 GMT Organization: Object Factory GmbH (Germany) Message-ID: <5riept$ju8$1@leonie.object-factory.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Can anybody confirm that NeXT/Apple changed the behaviour of NSMutableDictionary's copyWithZone method from 4.1 to 4.2? As far as I understand it, the values (not the keys!) were just retained once more in 4.1 but are copied in 4.2. This is really bad if you store objects in the dictionary that cannot be copied (like EOs for example) but have to copy a dictionary that contains them. I am a bit reluctant to implement the following methods in all affected classes but I don't see another possibility. - copyWithZone:(NSZone *)zone { return [self retain]; } And if I want to store an NSLock? Categories...? erik -- Erik Dörnenburg -- OBJECT FACTORY -- Gesellschaft für Informatik und Datenverarbeitung mbH -- http://www.object-factory.com/~erik
From: Stephen Peters <speters@cygnus.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 28 Jul 1997 12:52:50 -0400 Organization: Cygnus Solutions Message-ID: <qdwwmb3yql.fsf@scuba.cygnus.com> References: <33DA91BA.211E@afs.com> <B000476E-4B967@206.165.44.80> <5rhrcl$p6i$1@ironhorse.plsys.co.uk> mmalcolm crawford <malcolm@plsys.co.uk> writes: > I presume you can take a Reuters feed and display stock quotes in > real time using uncompiled GX too? It doubtless translates all > currency units to Yen in Pantone-269 on the fly as well? Only while printing. ;-) -- Stephen L. Peters speters@cygnus.com PGP fingerprint: BFA4 D0CF 8925 08AE 0CA5 CCDD 343D 6AC6 "Poodle: The other white meat." -- Sherman, Sherman's Lagoon
From: Stephen Peters <speters@cygnus.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 28 Jul 1997 13:03:21 -0400 Organization: Cygnus Solutions Message-ID: <qdu3hf3y92.fsf@scuba.cygnus.com> References: <5rav8v$ek@shelob.afs.com> <AFFEA5B2-402C7@206.165.44.29> "Lawson English" <english@primenet.com> writes: > [I dared suggest that OpenDoc would be easier to use for the end-user than > InterfaceBuilder] If you'd said that, you might not have gotten the response you did. Instead, you said (article <AFFE4E6B-52FD4@206.165.44.75>): "OpenDoc required NO programming experience whatsoever to create runtime-linked applications called OpenDoc documents. To get comparable utility out of IB requires years of programming training and experience." That statement is totally stupid. You need no programming experience with OD if all the components are there -- you just link them together. I (and Greg) submit that if all the components are there you don't need years of programming training. Surprisingly, that's all Greg said. He didn't say IB was appropriate for users, just that you're exaggerating massively (again). Maybe I'm misunderstanding you. I'm still trying to figure out how GX can prevent tooth decay... -- Stephen L. Peters speters@cygnus.com PGP fingerprint: BFA4 D0CF 8925 08AE 0CA5 CCDD 343D 6AC6 "Poodle: The other white meat." -- Sherman, Sherman's Lagoon
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: More OpenDoc advocacy (was Re: How to save Apple Date: 28 Jul 1997 12:07:00 -0700 Organization: Primenet Services for the Internet Message-ID: <B00240BD-CAB18@206.165.44.11> References: <wnr_6h200YUn06O=80@andrew.cmu.edu> To: "Charles William Swiger" <cs4w+@andrew.cmu.edu> nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer, nntp://news.primenet.com/comp.sys.mac.advocacy MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Charles William Swiger <cs4w+@andrew.cmu.edu> said: Lawson English <english@primenet.com> said: > > So are you suggesting the use of Interface Builder as a user-level tool? > > Absolutely. Obviously, you don't realize that a number of > scientific/mathematic apps written for NEXTSTEP were not created by > software developers, but people who understood their field (not > computers) and found that IB and the NEXTSTEP development environment > was powerful enough and easy enough to use that they could write elegant > GUI programs to do what they wanted. OK, but that doesn't address the issue of whether or not IB would be suitable as a universally available tool for end users to create dynamic content, which is the point that I'm trying to make. OD documents use the information contained in the bento file to seek out and link together various OpenDoc editors when the document is opened. The end-user never has to worry about the programming details and seldom has to worry about linking the parts together so that they cooperatively share data back and forth. When I drag a webpage from Cyberdog to a word-processor part, all the linking of code is done automatically. If I drag a spreadsheet part on top of a charting part, the code-linking AND the data exchange is done automatically. This is an *end-user* oriented tool. It isn't meant to allow new kinds of applications in the traditional sense to be created. It is meant to allow users to combine user-oriented content in a transparent way. OD does all the code-linking in the background. To the user, embedding the webpage in the spreadsheet looks and feels exactly like embedding a drawing in a wordprocessor window in ClarisWorks. It is a USER-oriented tool, not a programming tool. The actual events taking place behind the scene may be similar or identical to what IB does, but the user need never worry about these issues for the vast majority of tasks that they will use OD for. As I advocated elsewhere, making an OD 2.0 specification that allows for more IB-like functions certainly makes sense, but I sincerely do NOT believe that IB should be used as the basis for creating an OD 2.0. Think of it this way: OpenDoc provides functionality along the lines of the AppKit, but for parts, instead of applications meant to stand alone (although virtually any OD part can be used as though it were a standalone application). Instead of talking about IB being the functional equivalent of OD (there is overlap, but not as much as you appear to believe -*I* believe), we should be talking about how to make the Yellow Box architecture completely component-based. Any app running in the Yellow Box should behave as though it were an OD part -able to embed or be embedded-in other parts. When Rhapsody apps are exported to non-Rhapsody systems, they should function as standalone applications. There should also be a provision for Yellow-box-only components that do NOT export to other systems. These would be lighter-weight parts along the lines of Java Applets, JavaBeans or the less feature-bloated OpenDoc parts (like I understand Brad Hutchings' parts to be like). Rather than saying "oh, gee, IB/NeXT can already do much/most of what OpenDOc does, but in a different way," why not acknowledge that OD has a great deal to offer Rhapsody as far as parts-nesting/interaction/GUI issues go, and work on designing a hybrid that combines the *complete* functionality of OD and the *complete* functionality of the AppKit and whatever IB has to offer that is appropriate for generic components? While not as easy-to-use/program-to/etc, the utility of OpenDOc and the ODF are greater for most end-users than the utility of the AppKit and IB in those areas where components are concerned. Why not merge the two frameworks using an embracing philosophy, rather than talking about "tacking on" *some* of OpenDoc's more trivial features? Rhapsody should become GREATER than MacOS + NeXT, not just be NeXT + one or two bits of MacOS. I want Rhapsody to become OpenDoc on steriods, not merely Just Another Framework that Microsoft will eventually imitate. Microsoft will NEVER imitate OpenDoc. OpenDoc allows Office-like products to be created out of masses of 3rd party products purchased seperately by the end-user. Microsoft can't afford to allow that to happen. OpenDOc was *designed *specifically** so that that could happen. Was the AppKit? Was IB? ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: kennel@nospam.lyapunov.ucsd.edu (Matt Kennel (Remove 'NOSPAM' to reply)) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 28 Jul 1997 19:45:55 GMT Organization: University of California at San Diego Message-ID: <slrn5tptnv.6eo.kennel@lyapunov.ucsd.edu> References: <EDxvJz.7CG.0.scream@sounds.wa.com> <AFFFCEB3-5639E@206.165.44.148> <tbrown-2607972239440001@ppp054.dialup.cmh.netset.com> On Sat, 26 Jul 1997 22:39:44 -0400, Ted Brown <tbrown@netset.com> wrote: :In article <AFFFCEB3-5639E@206.165.44.148>, "Lawson English" :<english@primenet.com> wrote: : :> :>I meant EXACTLY what I said: comparing OpenDoc to IB, you would need to :>write lots of code (or someone, somewhere would), in order to get the same :>user-experience from IB that end-users get from OpenDoc. : :The discussion wasn't really about IB vs. OpenDoc, it was more VB vs. IB. : :Someone wrote "lots of code" to get OpenDoc working, and someone else :wrote even more code to get the OpenDoc part working. The OpenDoc dev :tools supposedly sucked as well (I've had no firsthand experience with :them). I seem to recall that writing an Mac OpenDoc port didn't :automatically give you a UNIX/Window/OS/2 OD part either. (Or am I :wrong?) I question the fundamental premises of OpenDoc: * Is it even remotely feasible that end-users who are not familiar with anything like programming would be able to tie-together programs in sensible ways? * Is it particularly desireable to have a generic "document" concept in which one inserts 'parts' which are instantiated by different applications and somehow pray that some sort of useful global ''emergent behavior'' will automatically instantiate itself? There is yet another point. If you want to support 'components', then the programming language and system must make it EXTERMELY EASY for the implementor to turn objects into ''components''. The appropriate fundamental semantic concepts must be designed into the system from the beginning. Is this the case for OpenDoc? Doesn't seem like it. Is this the case for MS COM? Hell No. Is this the case for CORBA? Not particularly. Is this the case for Java RMI? Yes. {What is a Java Bean? Nothing barely heavier than an ordinary Java class.} Is this the case for Interface Builder? Yes. -- * Matthew B. Kennel/Institute for Nonlinear Science, UCSD - * "People who send spam to Emperor Cartagia... vanish! _They say_ that * there's a room where he has their heads, lined up in a row on a desk... * _They say_ that late at night, he goes there, and talks to them... _they *- say_ he asks them, 'Now tell me again, how _do_ you make money fast?'"
From: alanf@izzy.net Newsgroups: comp.sys.next.programmer Subject: inter-nib messaging Date: 28 Jul 1997 19:14:23 GMT Organization: "Comshare, Inc." Message-ID: <5rir2g$9fr$1@inet-prime.comshare.com> greetings ethereal composite mind, I'm having some weirdness in a 3.2 black dev environ: object A unarchives and is files owner of object B. When I try to message Object A from Object B (via an ID var connected to Files Owner), it appears to be messaging a different object altogether. Clues: 1) Console message: unrecognized -updateLocalNumberDisplay: message sent to instance (0x42028) of PhoneController class. 2) I'm using NXRunAlertPanel to display the Object ID's. The value of the id var in Object B that I've wired up to Files Owner is not the same as the "self" value displayed from Object A. I created a simple test from scratch, and the Object id's co-incided when displayed. Any ideas? Regards, Alan Frabutt (alanf@izzy.net) someone else having similar problems, either > in this newsgroup, or on the NeXTprog mailing list. Does anyone know > about this? Is there a way to create a Help.store file which will be > adding when a bundle is loaded? or is the correct approach to > circumvent the Help compression and just copy the files? Nope, loading a Help.store from a bundle doesn't seem to work, even though the 3.3 AppKit release notes seem to imply otherwise... :-( A possible workaround is to rename the help directory in your bundle project to something other than "Help", and add it to PB's "Other Resources". Disadvantage of this is that you lose the ability to associate help with UI elements in IB. Another way is to patch things up for an installed bundle in your Makefile.postamble, somewhat like this (WARNING: this is UNTESTED!): # Replace Help.store with uncompressed Help in installed bundle. after_install:: @bundle=$(DSTROOT)$(INSTALLDIR)/$(REL_PRODUCT_ROOT); \ $(RM) -f $$bundle/$(LANGUAGE).lproj/Help.store; \ $(FASTCP) $(LANGUAGE).lproj/Help $$bundle/$(LANGUAGE).lproj Hope this helps, Tom. -- __/__/__/__/ Tom Hageman <tom@basil.icce.rug.dev.null.nl> [NeXTmail/Mime OK] __/ __/_/ IC Group <tom@icgned.dev.null.nl> (work) __/__/__/ <<SPAMBLOCK: remove .dev.null to reply>> __/ _/_/ Confused? You won't be after the NeXT episode.
Sender: gary099g@erols.com Control: cancel <5qkd4e$kod@news1-alterdial.uu.net> Message-ID: <cancel.5qkd4e$kod@news1-alterdial.uu.net> Subject: cmsg cancel <5qkd4e$kod@news1-alterdial.uu.net> From: jem@xpat.com Newsgroups: comp.sys.next.programmer Date: 17 Jul 1997 06:12:41 GMT EMP article removed by jem@xpat.com. Original Headers: From: gary099g@erols.com Subject: Get FREE SEX SITE..password is Newsgroups: comp.sys.next.programmer Path: news.dti.ad.jp!...!newsfeed.direct.ca!news-feed1.tiac.net!uunet!not-for-mail Lines: 99
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer,comp.sys.next.misc Subject: Re: C++ in framework? Date: 28 Jul 1997 18:48:42 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5ripia$jal$2@owl.slip.net> References: <199707281715094377217@dip065-2.hamburg.netsurf.de> In comp.sys.next.programmer Dix Lorenz <dix.lorenz@hamburg.netsurf.de> wrote: > Hi, > I would like to build a framework with a couple of C++-classes under OS > Enterprise 4.2 for NT. I tried the same under OS4.1/Mach and it worked > just like that. Under NT I get "unresolved external symbol" errors from > the linker when trying to use the framework. If I take the source files > from the framework and just use the exact same files in my application > directly, it works without a problem. > Does anybody have an idea what's going on? > If it's of importance: I use the Prelude to Rhapsody. You might try adding "-undefined suppress" to the framework link options. -- Steve Dekorte - OpenStep consultant - San Francisco
From: Jay Hill <volfreak@macconnect.com> Newsgroups: comp.sys.next.programmer Subject: Newbie Seeking Development Information Date: Mon, 28 Jul 1997 15:07:12 -0400 Organization: Oak Ridge National Lab, Oak Ridge, TN Message-ID: <33DCEDE0.4155@macconnect.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit All, Please forgive my newbieness. I am fairly computer literate (have been a Mac user for 10+ years, little DOS and UNIX for 15+ years (though not much UNIX for last 8), and Windows for 5+) and have been developing database applications for Windows for the past ~5 years (MSVisual FoxPro, PowerBuilder, and other miscellaneous development applications) as well as integrating/customizing office applications. I would like to get more involved in developing for Rhapsody. Early this year, I joined the Apple Developer Program. I do not have a background in C or C++. I have some experience with OO techniques and philosphies. I got the OpenStep tutorial and have read through most of the document. I'm still learning the syntax but I'm not having too much difficulty figuring it out. I also got the OpenStep/OO documentation available (includes description and detail on ObjC) from NeXT's web site, as well as Foundation and App Kit documentation. I have just started reading the OpenStep documentation. I have a background in systems (mostly communications and computer-controlled data/signals processing) and business. I understand the business models well so I feel I have a good grasp as to when/where systems and software can be of benefit. What I need is more experience (which will come with work) and knowledge (which is why the request) on the actual development tools and methodology. What suggestions can you provide (such as)? 1. What other documentation should I look at? I saw a reference to (IIRC) "Developing Business Applications with OpenStep". I have the ref printed out at home so can't give exact name. 2. What, if any, courses should I persue? 3. Any other good bits of info you feel up to offering. Thanks in advance for any and all responses. Could you please email responses to "volfreak@macconnect.com" as I don't make it to the newsgroups that frequently? -- Jaybird B-) "The larger the island of knowledge, the longer the shore of imagination." -- Unkown
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 28 Jul 1997 14:48:00 -0700 Organization: Primenet Services for the Internet Message-ID: <B0026670-2F568@206.165.43.190> References: <qdu3hf3y92.fsf@scuba.cygnus.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Stephen Peters <speters@cygnus.com> said: > "OpenDoc required NO programming experience whatsoever to > create runtime-linked applications called OpenDoc > documents. To get comparable utility out of IB requires years > of programming training and experience." > > That statement is totally stupid. You need no programming experience > with OD if all the components are there -- you just link them > together. I (and Greg) submit that if all the components are there > you don't need years of programming training. > I presume that you meant that you need no programming experience with *IB* [not OD] if all the components are there. There's no *conscious* "linking" required with OpenDoc. You just drag and drop a part and the linking takes place behind the scenes. There are behaviors BUILT INTO OpenDoc that every ODF-using part provides that IB doesn't provide and there is no way that anyone could take off-the-shelf IB components and get certain OD-like behaviors within IB without adding that behavior using Obj-C (Which would require "years of programming experience"). Of course, the same is true with OpenDoc for those behaviors that IB provides that OD doesn't provide, but I submit that for the average end-user, OpenDoc (and the framework Apple designed for programmers to use -ODF) provides far more interesting behaviors, _gratis_, than IB does, for the non-trained end-user. OpenDOc uses virtually the same user interface as System 7.x does. "Linking" in OPenDoc is accomplished merely by dragging a part or document into an open OpenDoc window. This is far easier than what IB requires (from what everyone has said -you don't require 2 weeks experience to learn how to drag an icon into an open window). OTOH, IB provides (from what little I've seen) lots of extra behaviors that OD doesn't provide that WOULD be useful for power-users in certain [non-programming] situations. > Surprisingly, that's all Greg said. He didn't say IB was appropriate > for users, just that you're exaggerating massively (again). > He claimed that his daughter, with only a few weeks of training, was able to use IB productively [in an OD-like way]. OD requires virtually no training. It was meant to be intuitive to non-technical end-users. In fact, for most tasks, the same behavior that accomplishes something in a Works application is all that is needed to accomplish the same kind of thing in OD. If you know how to use ClarisWorks or Microsoft Works, than you know how to use a large portion of OpenDoc. In fact, aside from some advanced features, I'd say that 90% of all of OD's core functionality is available without reading the manual or even seeing a demonstration. Just double click on a stationary file and drag other stationary files into it and you're on your way. Experiment with doing things and you can discover most of the rest of the functionality of OpenDoc itself as you go. > Maybe I'm misunderstanding you. I'm still trying to figure out how GX > can prevent tooth decay... Mmmm... Maybe by preventing build-up of stress while creating and manipulating relatively complex graphical objects so that you don't grind your teeth at night as much? How do you implement a paint-dragging routine in GX? create a brush shape (2-3 lines of code, even in hypercard) track the brush shape (5-6 lines of code, although it gets more complicated if you allow for the less-than-perfect sampling that mouse-tracking does -Bresenham's line-drawing is trivial but requires a relatively large number of lines) How do you implement a dynamic brush shape? Add a few lines of code to modify the ink/style/transform objects contained in the brush shape whilest in the mouse-tracking loop. More than a little slow in HyperCard due to the callback overhead, but still useful for prototyping the algorithms and it looks cool. ++++++++++++++++++++++++++++++ HOW DO I SUBSCRIBE TO GX-TALK? --------------------------------- To subscribe to GX-TALK proceed as follows: Send E-mail to: <gx-talk-request@aimed.org> with anything in the subject line and the following command as the first (and only) line of the message body: SUBSCRIBE +++++++++++++++++++++++++++++++++
From: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.next.programmer Subject: Re: Newbie Seeking Development Information Date: 28 Jul 1997 17:03:30 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5rj1f2$pja$1@sps1.phys.vt.edu> References: <33DCEDE0.4155@macconnect.com> [Posted to comp.sys.next.programmer, cc'ed to original poster.] In article <33DCEDE0.4155@macconnect.com>, Jay Hill <volfreak@macconnect.com> wrote: > Early this year, I joined the Apple Developer Program. I do not have a > background in C or C++. I have some experience with OO techniques and > philosphies. I got the OpenStep tutorial and have read through most of > the document. I'm still learning the syntax but I'm not having too much > difficulty figuring it out. I also got the OpenStep/OO documentation > available (includes description and detail on ObjC) from NeXT's web > site, as well as Foundation and App Kit documentation. I have just > started reading the OpenStep documentation. > What I need is more experience > (which will come with work) and knowledge (which is why the request) on > the actual development tools and methodology. > 3. Any other good bits of info you feel up to offering. After thoroughly reading and re-reading the OpenStep API (there's lots of good design philosophy embodied in it!), read through some of the example source code that comes with OPENSTEP/Rhapsody, as well as the mini-examples in NeXTAnswers. Other than that, you probably don't need a whole lot else specifically on OpenStep (though there are probably a few books you might want to pick up; I'm not current on OpenStep books), so I'll just give a few more recommendations on resources for object-oriented design: Study dynamic concepts such as forwarding, NSInvocations, protocol conformance, class clusters, +poseAs:, and bundles. Read more about dynamism at http://www.misckit.com/DynamicOO/. Buy this book: _Design Patterns: Elements of Reusable Object-Oriented Software_ (http://st-www.cs.uiuc.edu/users/patterns/DPBook/DPBook.html). The examples are mostly C++, but this book is unmatched in teaching about object-oriented design. Lots of the patterns in the book are implemented even better in Obj-C (such as Obj-C forwarding in NSProxy for the Proxy pattern, the NSNotification implementation of the Observer pattern, the NSMatrix +setCellClass: and -setPrototype: implementations of the Prototype pattern, etc.) Read more about design patterns, such as at http://st-www.cs.uiuc.edu/cgi-bin/wikic/wikic.
From: Todd White <toddw@telexiscorp.com> Newsgroups: comp.sys.next.programmer Subject: Threads and NSAutoreleasePools Date: Mon, 28 Jul 1997 19:45:05 -0400 Organization: Telexis Corporation Message-ID: <33DD2F01.D1F5FB47@telexiscorp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Does anybody know how to use NSThreads with NSAutoreleasePools? It seems that, though you can create an autorelease pool then detach a new thread, you cannot create another pool and then release it. Here's an example: - (void) someMethod { NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init]; .... do stuff.... [NSThread detachNewThreadSelector: @selector(anotherMethod) toTarget: [[[self class] alloc] init] withObject: nil]; [pool release]; } - (void) anotherMethod { NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init]; .... do stuff.... [NSThread detachNewThreadSelector: @selector(thirdMethod) toTarget: self withObject: nil]; [pool release]; } - (void) thirdMethod { NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init]; .... do stuff.... [pool release]; // Inner-most exit point. } When the code reaches the inner-most "[pool release]" it says that I'm sending release to an already freed object. OK, so the Foundation Kit manual says that NSThreads send release to all of the autorelease pools that have been declared within the thread. That shouldn't apply here. If I remove the [pool release] from the thirdMethod, it works OK, but only to give me the same error in "anotherMethod" (one level up). It seems that NSAutoreleasePool is NOT thread-safe. Can anyone provide any proof that I am wrong (please - I really hope you can)? Thanks, Todd White Telexis Corporation
From: uzs90z@uni-bonn.de (Michael Schuerig) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Tue, 29 Jul 1997 01:06:34 +0200 Organization: RHRZ - University of Bonn (Germany) Message-ID: <1997072901063415024@rhrz-isdn3-p23.rhrz.uni-bonn.de> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> <nagleEDIw6C.HGy@netcom.com> <5r8uc7$el@lobotomy.urz.uni-wuerzburg.de> <AFFE4E6B-52FD4@206.165.44.75> <EDxvJz.7CG.0.scream@sounds.wa.com> <AFFFCEB3-5639E@206.165.44.148> <tbrown-2607972239440001@ppp054.dialup.cmh.netset.com> <slrn5tptnv.6eo.kennel@lyapunov.ucsd.edu> Matt Kennel (Remove 'NOSPAM' to reply) <kennel@nospam.lyapunov.ucsd.edu> wrote: > * Is it even remotely feasible that end-users who are not familiar > with anything like programming would be able to tie-together > programs in sensible ways? Yes. If the "programs" present themselves in the way OpenDoc stationery does. Anyone who's able to use ClarisWorks can master that as well. > * Is it particularly desireable to have a generic "document" concept > in which one inserts 'parts' which are instantiated by different > applications and somehow pray that some sort of useful global > ''emergent behavior'' will automatically instantiate itself? Yes. To me that is. Your mileage may vary, of course. Michael --- Michael Schuerig mailto:uzs90z@uni-bonn.de http://www.uni-bonn.de/~uzs90z/
From: dix.lorenz@hamburg.netsurf.de (Dix Lorenz) Newsgroups: comp.sys.next.programmer,comp.sys.next.misc Subject: Re: C++ in framework? Date: Tue, 29 Jul 1997 03:28:59 +0200 Organization: EDV - Beratung Message-ID: <199707290328596592668@dip062-2.hamburg.netsurf.de> References: <199707281715094377217@dip065-2.hamburg.netsurf.de> <5ripia$jal$2@owl.slip.net> Steve Dekorte <dekorte@slip.net> wrote: > In comp.sys.next.programmer Dix Lorenz <dix.lorenz@hamburg.netsurf.de> wrote: > > Hi, > > I would like to build a framework with a couple of C++-classes under OS > > Enterprise 4.2 for NT. I tried the same under OS4.1/Mach and it worked > > just like that. Under NT I get "unresolved external symbol" errors from > > the linker when trying to use the framework. If I take the source files > > from the framework and just use the exact same files in my application > > directly, it works without a problem. > > Does anybody have an idea what's going on? > > If it's of importance: I use the Prelude to Rhapsody. > > You might try adding "-undefined suppress" to the framework link options. I am afraid I didn't make it too clear: the framework builds, the app doesn't. But your tip got me to finishing the build of my app. Horrible crash. I am almost afraid the linker is correct when it complains. So probably the building of the framework is the culprit. Do you have any other ideas? Thanks, Dix
From: rbarris@quicksilver.com (Rob Barris) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: OpenDoc vs IB? Why "vs"? (Was Re: How to save Apple Date: Mon, 28 Jul 1997 16:23:46 -0700 Organization: Quicksilver Software, Inc. Message-ID: <rbarris-ya023280002807971623460001@206.82.216.1> References: <5rgrjh$o8u$1@news.xmission.com> <B00226E0-6977D@206.165.44.11> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit In article <B00226E0-6977D@206.165.44.11>, "Lawson English" <english@primenet.com> wrote: > Take a hypothetical jigsaw puzzle part and embedd the Cyberdog browser-part > within it. Now surf to your favorite web-page and scramble the puzzle. Now > continue surfing and try to unscramble the puzzle when you have no idea > what it is that you are looking at. Now embed the scrambled web-browser > part inside a spreadsheet, making it a dynamic illustration for the > spreadsheet. Now make that functionality work in IB. Now make me care. Rob Barris Quicksilver Software Inc. rbarris@quicksilver.com * Opinions expressed not necessarily those of my employer *
From: "Jay Hill" <volfreak@macconnect.com> Newsgroups: comp.sys.next.programmer Subject: Re: Newbie Seeking Development Information Date: 28 Jul 97 20:59:09 -0400 Organization: None Message-ID: <B002B8AD-98ED@206.80.166.87> References: <199707282154.OAA07059@idiom.com> Mime-Version: 1.0 Content-Type: text/enriched; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable <SMALLER>Thanks John, I appreciate the reccomendations (not the first time I've heard take C and not ++). I'll go look up the info. Again thanks, Jay Hill "The larger the island of knowledge, the longer the shore of imagination." Unknown --------------------------------------------------- This message was created and sent using the Cyberdog Mail System --------------------------------------------------- </SMALLER> ng the Cyberdog Mail System --------------------------------------------------- </SMALLER>
From: Greg_Anderson@afs.com (Gregory H. Anderson) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 29 Jul 1997 14:23:21 GMT Organization: Anderson Financial Systems Inc. Message-ID: <5rkucp$k2b@shelob.afs.com> References: <B0026670-2F568@206.165.43.190> "Lawson English" <english@primenet.com> writes > [I] claimed that [my] daughter, with only a few weeks of training, was > able to use IB productively [in an OD-like way]. OD requires virtually > no training. It was meant to be intuitive to non-technical end-users. I made no such claim. What I said was that my daughter learned how to use HyperStudio at school. Using those concepts and knowledge, she was able to apply what she already knew about instantiating buttons, hooking things up, etc. to InterfaceBuilder with NO additional "training." I know you have claimed that HyperCard is an end-user tool, so this implies that anyone who understands HyperCard "programming" can understand IB also. -- Gregory H. Anderson | "We're in the land of the blind, Visionary Ophthalmologist | selling working eyeballs, and they Anderson Financial Systems | balk at the choice of color." -- Tony greg@afs.com (NeXTmail OK) | Lovell, on Mac user reactions to NeXT
From: michael.b.moore@msfc.nasa.gov (Michael B. Moore) Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: More OpenDoc advocacy (was Re: How to save Apple Date: Tue, 29 Jul 97 07:45:50 GMT Organization: NASA/MSFC Message-ID: <michael.b.moore.1220549990D@news.msfc.nasa.gov> References: <wnr_6h200YUn06O=80@andrew.cmu.edu> <B00240BD-CAB18@206.165.44.11> In Article <B00240BD-CAB18@206.165.44.11>, "Lawson English" <english@primenet.com> wrote: >I want Rhapsody to become OpenDoc on steriods, not merely Just Another >Framework that Microsoft will eventually imitate. > >Microsoft will NEVER imitate OpenDoc. OpenDoc allows Office-like products >to be created out of masses of 3rd party products purchased seperately by >the end-user. Microsoft can't afford to allow that to happen. OpenDOc was >*designed *specifically** so that that could happen. > >Was the AppKit? Was IB? > You stated the case for OD in this post very well, Lawson. This part I particularly agree with. I still have dreams of seeing something like OD come to fruition; it may be a case of technology way ahead of its time. Someday, I don't know when, but someday, the OD model will be how *all* shrink-wrapped software will be written. Mike Moore NASA/MSFC
From: rog@ohm.york.ac.uk (Roger Peppe) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] WDEFs & Rhapsody? Date: 29 Jul 1997 14:36:33 GMT Organization: Department of Electronics, University of York, UK. Sender: rp9@york.ac.uk Message-ID: <5rkv5h$7ar$1@netty.york.ac.uk> References: <869482436.23064@dejanews.com> <EDo7rE.9EC@cam-ani.co.uk> On Mon, 21 Jul 1997 13:21:13 GMT, Ian Stephenson <ians@cam-ani.co.uk> wrote: > If you really did want to radically change a window, you subclass it. This > is generally not a good idea, and interface builder stops being helpfull, > but it allows you complete control over the windows behaviour. why then were Windows the only objects that pre-Openstep Interface Builder allowed the programmer to subclass directly ? i've found that subclassed windows are very useful, when you want to *add* to behaviour rather than modify it. when programming a simple panel for example, it's very useful to be able to reduce it to a single class responsible for most of that panel's behaviour, and for communicating with the nib file's owner, etc. there is no reason why a Window can't act as its own delegate... rog.
From: "Mitch" <Pahli@cww.de> Newsgroups: adass.iraf.programming,alt.comp.shareware.programmer,alt.msdos.programmer,comp.msdos.programmer,comp.os.msdos.programmer,comp.programming,comp.programming.contests,comp.sys.next.programmer,gfido.beruf.programmierer Subject: NEED HELP (How can I remove the protection of my HD) Date: 29 Jul 1997 19:25:00 GMT Organization: Telemedia News Server Message-ID: <01bc9c55$8f88ed40$535047c3@cityweb.cww.de> Hi, I´ve a big problem with my hard disk. I have two hard disks (<C> and <D>). <D> doesn´t work for one year. FDISK recognize it with the addition "PROTECT". I can´t read the disk, I can´t write to the disk and it´s impossible to format the disk. Can you give me a tip how to manage it? How could I remove the protection of the HD? Please help me !!!! Bye, Mitch
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 29 Jul 1997 12:15:00 -0700 Organization: Primenet Services for the Internet Message-ID: <B0039427-1AF54@206.165.44.33> References: <5rkucp$k2b@shelob.afs.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Gregory H. Anderson <Greg_Anderson@afs.com> said: [my mistake snipt] > I made no such claim. What I said was that my daughter learned how to use > HyperStudio at school. Using those concepts and knowledge, she was able > to > apply what she already knew about instantiating buttons, hooking things > up, etc. to InterfaceBuilder with NO additional "training." I know you > have claimed that HyperCard is an end-user tool, so this implies that > anyone who understands HyperCard "programming" can understand IB also. I misunderstood, sorry. However, the interface for OpenDoc is far easier than HyperCard's, also. To "hook things up" means to drag them into a window. Things that were designed to use input from other things generally only require a drag'n'drop also. You have "live" cut and paste to create a live link (although IB's techniques might make more sense here), and so on. My point about OD is that it is NOT a programming tool. It is a middleware system that provides a universal plug-in service. It was designed to be a no-brainer to use. And plug-ins can be nested arbitrarily using OpenDoc. OpenDOc provides the user-interface conventions for arbitrary nesting of parts. As I said, the issue should not be "can IB do what OD does?" but "How can we provide, *UNIVERSALLY*, OpenDoc services within Rhaposdy?" It's all very well to talk about "run mode" in IB. But OpenDoc is ALWAYS in "run mode" and in "builder mode" simultaneously. ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer,comp.sys.next.misc Subject: Re: C++ in framework? Date: 29 Jul 1997 18:52:00 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5rle4g$a2c$1@owl.slip.net> References: <199707281715094377217@dip065-2.hamburg.netsurf.de> <5ripia$jal$2@owl.slip.net> <199707290328596592668@dip062-2.hamburg.netsurf.de> Dix Lorenz wrote: > Under NT I get "unresolved external symbol" errors from What symbols does it list with this message? -- Steve Dekorte - OpenStep consultant - San Francisco
From: kefera@sekmet.com Newsgroups: comp.sys.next.programmer Subject: $5000 Credit Card, Low APR Date: Tue, 29 Jul 1997 12:54:16 Organization: Simply Internet, San Diego Message-ID: <5rli3g$f48@frog.inetworld.net> Because of our arrangement with OFFSHORE BANKS- Now you can get your own UNSECURED MAJOR CREDIT CARD at interest rates as low as 6.95%, GUARANTEED! *Accepted at millions of locations world-wide, this is a MAJOR Bank Credit Card you are very familiar with! *No Credit Check! *No Social Security Number needed for approval! *Even Bakruptcy-No Problem! *$5,000 Credit Limit, minimum! *Unsecured Line of Credit of$15,000.! *BUSINESS paying $4800.00 monthly, with FREE WEB PAGE! * Could earn up to $460,000 yr! I am also available to answer all your questions! For more info,send an email to my autoresponder, Cashforyou@unitedmall.com Dr. David Alan
From: gary-nospam-@screaming.org (Gary W. Longsine) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Getting Prelude to Rhapsody Date: 29 Jul 1997 21:18:02 GMT Organization: Save the Skeet Foundation Message-ID: <5rlmma$5uf$1@news.platinum.com> References: <33D7BBDC.2A5A08DC@amrcorp.com> Cc: Eric_Smalling@amrcorp.com In <33D7BBDC.2A5A08DC@amrcorp.com> it appeared that Eric Smalling wrote: > I remember someone saying you can sign on with Apple as a developer for > $250 and you'd get, among other things, the Prelude to Rhapsody > package. Can anyone verify this and repost exactly what to sign up for. Oops! looks like I lied. Check out this site. Look under "getting started" where it seems you can in fact order the Prelude to Rhapsody bundle for about that price. http://devworld.apple.com/dev/prelude.html /gary -- Gary W. Longsine, Systems Engineer | ____/| OpenStep MachOS PLATINUM technology, inc. | \ o.O| Objective-C l_o_n_gsine@platinum.com (NeXTmail | =(_)= the Dock (Can i have his spam?) & MIME) |. U Elegance is Relevant.
From: Wolfgang Baron <wbaron@ixpoint.de> Newsgroups: comp.sys.next.programmer Subject: Re: CVS and NIBs - again Date: 29 Jul 1997 11:18:51 GMT Organization: iXpoint Informationssysteme GmbH, www.ixpoint.de Message-ID: <5rkjir$24s$1@ixpoint.de> References: <33D6E9B5.D89FECBE@marcon.de> <5r7oa0$kj8$1@concorde.ctp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Hi all, In <5r7oa0$kj8$1@concorde.ctp.com> Georg Tuparev wrote: > {{\xurl http://www.cyclic.com/ > }<} > http://www.cyclic.com/ > > -- georg -- I have been following this thread, however I do not have the older postings. The oldest posting I have is by Jim Redman of 22nd of July. My company uses OPENSTEP both on mach and NT. I would like to get CVS working under ProjectBuilder control on either of these systems, if not both. Is there any way to make CVS work with Projectbuilder under OPENSTEP 4.2 or OPENSTEP Enterprise ? I have successfully installed CVS on both systems, but I cannot get it to work with ProjectBuilder's SCM bundle. Creating a working directory fails on both systems for differing reasons. Using CVS manually works on both systems, but that is not what I want. The symptoms are: On OPENSTEP/Mach Paths are not correctly converted between \ and /, which causes dreadful directories to be created in /. On OPENSTEP Enterprise CVS seems to unpack only the global directory. Interfaces (and probably subprojects) are missing after having created the work directory. If I use CVS manually, it works (thus the CVS binary works). If you have any ideas on how to get around these problems or you have old postings containing solutions, please send me a copy. Georg Tuparev is on holiday, so I need someone else to send me the material. Thanks, -- ++++++++++++++++++++++++++++++++++++++++++++++++ + Dipl.- Inform. Wolfgang Baron + + iXpoint Informationssysteme GmbH + + Daimlerstrasse 3, 76275 Ettlingen, Germany + + phone ++49 7243 3775-0 Fax ++49 7243 3775-77 + + email: wbaron@ixpoint.de (NeXTmail and MIME) + ++++++++++++++++++++++++++++++++++++++++++++++++ + pConnector - Network Printing for NEXTSTEP + + sca - the static code analyser + ++++++++++++++++++++++++++++++++++++++++++++++++
From: search@search30.com Subject: Win The Web Wars Newsgroups: comp.sys.next.programmer Organization: e-health Message-ID: <33de7126.0@news1.abac.com> Date: 29 Jul 97 22:39:34 GMT HOW CAN YOU BECOME SUCCESSFUL ON THE INTERNET? The Secret is Don't get lost in the Search Engines Do you want to know the secrets to getting the greatest number of hits by BEST POSITIONING YOUR WEB SITE ON THE SEARCH ENGINES? No problem, our *complete* scientific study, shows exactly how! In fact, the program is so powerful, it's dangerous to your competitors ...PLUS.... Two (2) absolutely NEW tricks just created by our staff on July 8, 1997 be the first to use them http://www.e-earth.net/reports.htm
From: "Benjamin Smith" <benjs@ix.netcom.com> Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: More OpenDoc advocacy (was Re: How to save Apple Date: 29 Jul 97 20:49:08 -0400 Organization: Netcom Message-ID: <B00407CD-106DF2@207.93.51.100> References: <michael.b.moore.1220549990D@news.msfc.nasa.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit nntp://nntp.ix.netcom.com/comp.sys.mac.programmer.misc, nntp://nntp.ix.netcom.com/comp.sys.next.programmer, nntp://nntp.ix.netcom.com/comp.sys.mac.advocacy >You stated the case for OD in this post very well, Lawson. This part I >particularly agree with. I still have dreams of seeing something like OD >come to fruition; it may be a case of technology way ahead of its time. >Someday, I don't know when, but someday, the OD model will be how *all* >shrink-wrapped software will be written Mike, I absolutely agree and really wish people would be more forward thinking at times. OD is not failed technology because of itself, it really has to do with who introduced it, when it was introduced and what big manufacturers embraced it (or actually did not embrace it). I think many big outfits did not embrace because it would have exposed just how inefficient and bloated their software is. And it also opened the door for small nonames to get big and save consumers money. But OD is great technology and it will show up in essence sometime in the future, if by another name. Ben S.
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: search@search30.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33de7126.0@news1.abac.com> Control: cancel <33de7126.0@news1.abac.com> Date: 29 Jul 1997 22:36:12 GMT Message-ID: <cancel.33de7126.0@news1.abac.com> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.lang.objective-c,comp.sys.next.programmer Subject: weird warnings Date: 29 Jul 97 23:04:50 -0400 Organization: University of Michigan ITD News Server Message-ID: <B00427A0-696AB2@141.214.134.235> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: "Multiple recipients of list" <webobjects@omnigroup.com> nntp://news.itd.umich.edu/comp.sys.next.programmer I'm working on a compiled webobjects application. Right now I'm working in Project Builder. I have 8 classes I've created myself. When I build the project I get, at this time, 90 warnings and no errors. One consistent warning is: ****** /NextLibrary/Frameworks/WebObjects.framework/Headers/WebObjects.h could not use precompiled header '/NextLibrary/Frameworks/EOAccess.framework/Headers/EOAccess.p', because: header 'EOAccess/EOSchemaGeneration.h' overrideden, /NextLibrary/Frameworks/EOAccess.framework/Headers/EOSchemaGeneration.h vs./NextLibrary/Frameworks/EOAccess.framework/PrivateHeaders/EOSchemaGenera tion.h (within the precomp) ****** How do I go about fixing that? The other warnings don't seem as consistent, unless I delete the derived files everytime I build. Here's an example of some of the errors: ****** _GCFleet.m_ 'GCShip' does not respond to 'name' 'GCShip' does not respond to 'retain' ****** When I look at both GCShip.h and GCShip.m, my 'name' message looks perfectly fine. It's correct in that I don't have a retain message, which brings up another question, do I have to override 'retain' for my own classes? I also get the same error with the message autorelease. Here's what part of my GCShip header and implemention files look like: ****** GCShip.h ************************* #import "GC.h" @class GCLocation, GCPlayer, GCFleet, GCComponent, GCSystem @interface GCShip : NSObject { NSString *name; // other variables } (NSString *) name; // other methods @end ****** end GCShip.h ********************* ****** GCShip.m ************************* #import "GCShip.h" unsigned static int=0; @implementation GCShip : NSObject - (id) initWithLocation:(GCLocation *)loc andOwner:(GCPlayer *)theOwner andSystem:(GCSystem *)theSystem andComponent:(GCComponent *)firstComponent andPath:(NSString *)path { [super init]; // other stuff name = @"unnamed ship"; return self; } - (NSString *) name { return name; } @end ***** end GCShip.m *************************** Does anyone know what's going on? Why am I getting warnings saying methods don't exist that really do? Sorry if this is a basic question, but I new to Objective-C. rob -- <mailto: "Robert A. Decker" comrade@umich.edu> Listen to my Realaudio playlist:<http://hmrl.cancer.med.umich.edu/Rob/index.ssi> Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #10: "Wow. I'm a genius too. I think. BEEP." -Chris Elliott
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5rli3g$f48@frog.inetworld.net> Date: 30 Jul 1997 00:14:46 GMT Control: cancel <5rli3g$f48@frog.inetworld.net> Message-ID: <cancel.5rli3g$f48@frog.inetworld.net> Sender: kefera@sekmet.com Spam cancelled by sef@kithrup.com
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.lang.objective-c,comp.sys.next.programmer Subject: Re: weird warnings Date: 30 Jul 1997 04:14:40 GMT Organization: WolfWare Message-ID: <5rmf3g$31g$1@vader.wolfware.ipc.net> References: <B00427A0-696AB2@141.214.134.235> In-Reply-To: <B00427A0-696AB2@141.214.134.235> On 07/29/97, "Robert A. Decker" wrote: > > The other warnings don't seem as consistent, unless I delete the derived >files everytime I build. They may not seem consistent because a warning does not cause the .o file to fail to be generated. For example - assume you have MyClass.m which when built generates some warnings but still results in a "successful" MyClass.o file being generated. Next time you build make realizes that MyClass.m source has not changed since the last time MyClass.o was generated, sees no reason to re-compile MyClass.m and so does not encounter the code that resulted in the warnings being produced the first time. If you "clean" the project in-between then the O-files always get deleted and so all of the source is re-built every time so the warnings are "consistent". >****** >_GCFleet.m_ > 'GCShip' does not respond to 'name' > 'GCShip' does not respond to 'retain' >****** My first guess was that in GCFleet.m you forgot to #import GCShip.h - this will often result in warnings such as the above. But from looking at the code you posted I can see this isn't the case in this instance. > Here's what part of my GCShip header and implemention files look like: >****** GCShip.h ************************* >#import "GC.h" >@class GCLocation, GCPlayer, GCFleet, GCComponent, GCSystem > >@interface GCShip : NSObject >{ > NSString *name; > // other variables >} >(NSString *) name; Unless that was just a typo in the paste this is your problem. That's supposed to be: - (NSString *) name; (Note the leading "-" indicating it's an instance method as opposed to a "+" for a class method.) > When I look at both GCShip.h and GCShip.m, my 'name' message looks >perfectly fine. It's correct in that I don't have a retain message, I'm not sure what you're thinking here when you say "it's correct in that I don't have a retain" message. >which >brings up another question, do I have to override 'retain' for my own >classes? I also get the same error with the message autorelease. The general answer to your question is that you should absolutely never override retain. Or autorelease. Or release. Leave them alone. Don't touch them. Or else :-) (Actually there are a couple of very bizarre special cases where it might be useful to override these methods... until you can explain the zen of the existing retain/release scheme in your sleep, backwards I wouldn't advise it though.) -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: Steve Kellener <skellener@earthlink.net> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Mon, 28 Jul 1997 21:31:27 -0700 Organization: EarthLink Network, Inc. Message-ID: <33DD71CE.185D@earthlink.net> References: <AFEFAFFC-20EF8@206.165.44.6> <LTaylor7-1407971300290001@vic-ca1-19.ix.netcom.com> <19970717190420809661@accs-as07-dp12.dlls.grid.net> <bradley-1807970122490001@bns.vip.best.com> <199707212335441409135@accs-as08-dp15.dlls.grid.net> <5r5raq$39t$1@ironhorse.plsys.co.uk> <joe.ragosta-ya02408000R2307972009060001@news.dol.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Wanna save Apple? Buy a Mac. STEVE K.
From: schack@skyler.arc.ab.ca (Brian Schack) Newsgroups: comp.sys.next.programmer,comp.sys.next.misc Subject: Re: C++ in framework? Followup-To: comp.sys.next.programmer Date: 30 Jul 1997 08:22:33 -0600 Organization: Alberta Research Council, Calgary Alberta, Canada Sender: schack@skyler.arc.ab.ca Message-ID: <vb204g62mu.fsf@skyler.arc.ab.ca> References: <199707281715094377217@dip065-2.hamburg.netsurf.de> <5ripia$jal$2@owl.slip.net> <199707290328596592668@dip062-2.hamburg.netsurf.de> In-reply-to: dix.lorenz@hamburg.netsurf.de's message of Tue, 29 Jul 1997 03:28:59 +0200 >>>>> "Dix" == Dix Lorenz <dix.lorenz@hamburg.netsurf.de> writes: Dix> I am afraid I didn't make it too clear: the framework builds, Dix> the app doesn't. But your tip got me to finishing the build Dix> of my app. Horrible crash. I am almost afraid the linker is Dix> correct when it complains. So probably the building of the Dix> framework is the culprit. Dix> Do you have any other ideas? This sounds like a problem we struggled with for a while, caused by the stupidity of the NT linker. Basically functions and global variables are not exported properly from frameworks. At runtime, when the framework tries to access a global variable, it uses an incorrect address, and you get a "horrible crash". To get around it, you must use a Windows heinosity, namely the __declspec(dllimport) and __declspec(dllexport) declarations. More details can be found at: http://www.next.com/NeXTanswers/HTMLFiles/2462.htmld/2462.html under the questions: "Why do I get a segmentation fault when trying to use a bundle?" "How do I export symbols other than class names from my framework?" Hope this helps. Brian -- ---------------------------------------------------------------------------- Brian Schack |mailto:schack@arc.ab.ca | "I don't want to achieve Alberta Research Council |http://www.arc.ab.ca | immortality through my 6815 8th St NE | | work ... I want to achieve Calgary, Alberta |ph: (403) 297-7564 | it through not dying." Canada T2E 7H7 |fax: (403) 297-2339 | - Woody Allen ----------------------------------------------------------------------------
Date: Wed, 30 Jul 1997 09:50:13 -0600 From: ajmas@bigfoot.com Subject: Interface Control under Rhapsody / Open Step Newsgroups: comp.sys.mac.programmer.help,comp.sys.next.programmer Message-ID: <870260285.24089@dejanews.com> Organization: Deja News Posting Service Under OpenStep or Rhapsody, when you write your program do you actually try to separate yourself from how the high level user interface, as defined by the system is designed? The reason I am asking this is if I decide I would like to use OpenStep as a base, would it be possible at an extension level to redfine what the interface looked like and still have the applications working as normal. For example, if I decided that I did not like my interface having its menus at the top, would I be able to write a routine that drew the menu bar along the side or in a window? I do realise that there is no need to do this in normal operations, but I would be interested in experimenting, for my own benefit on interfaces that have been suggested either in Sci-Fi or by design artists. Thanks Andre "Where do you want to go today? Umm, forwards with innovation, though I take it you would have to copy the idea first" -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: acurylo@inmediapresents.nospam.com (Alex Curylo) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Wed, 30 Jul 1997 09:32:50 -0700 Organization: InMedia Presentations Message-ID: <acurylo-3007970932520001@van0407.tvs.net> References: <5rav8v$ek@shelob.afs.com> <AFFEA5B2-402C7@206.165.44.29> <qdu3hf3y92.fsf@scuba.cygnus.com> In article <qdu3hf3y92.fsf@scuba.cygnus.com>, Stephen Peters <speters@cygnus.com> wrote: > Maybe I'm misunderstanding you. I'm still trying to figure out how GX > can prevent tooth decay... Because if I read any more #@%#^@^#$%!!! pointless whining about decisions about as cast in stone as anything can be in this industry, I'm going to hunt down all the GX partisans and kick their faces in, and since they won't have any teeth left after that tooth decay will not be a problem. ---- Alex Curylo -- alex@witty.com
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: 30 Jul 1997 09:11:00 -0700 Organization: Primenet Services for the Internet Message-ID: <B004BA8A-2C7C7@206.165.44.36> References: <5rfuit$4ko$1@nntp2.ba.best.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit > Assuming you had the component objects to do so, yes, which is the > whole point. IB does not manipulate pretty pictures of objects. It > manipulates real-life honest-to-God objects that are no different than > the 'real thing'. > Yes, I know. However, OpenDoc is not an application (well it is, but it is meant to be useful system wide), and regardless, IB doesn't provide the kind of user-interface support that OD does, nor does it handle the kinds of automatic linking that OD does. That is, it doesn't determine what kind of components created the content that was saved last time around and auto-link those components together when you open a document. Nor does it determine what alternative components might be able to display the same data (or portions thereof) if the original components are not available. In a nutshell: even if all the right components are available, would you advocate replacing Microsoft Office with IB itself? OD was designed for just that. ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: acurylo@inmediapresents.nospam.com (Alex Curylo) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: How to save Apple Date: Wed, 30 Jul 1997 09:48:31 -0700 Organization: InMedia Presentations Message-ID: <acurylo-3007970948310001@van0407.tvs.net> References: <EDxvJz.7CG.0.scream@sounds.wa.com> <AFFFCEB3-5639E@206.165.44.148> <tbrown-2607972239440001@ppp054.dialup.cmh.netset.com> <slrn5tptnv.6eo.kennel@lyapunov.ucsd.edu> In article <slrn5tptnv.6eo.kennel@lyapunov.ucsd.edu>, kennel@NOSPAMlyapunov.ucsd.edu wrote: > I question the fundamental premises of OpenDoc: > > * Is it even remotely feasible that end-users who are not familiar > with anything like programming would be able to tie-together > programs in sensible ways? Well, no, but that's a bit disingenuous, because end-users are not able to figure out the pagination features of Word (took me 45 minutes to talk my mom through that yesterday), let alone do anything that actually takes some conceptualization skills. > * Is it particularly desireable to have a generic "document" concept > in which one inserts 'parts' which are instantiated by different > applications and somehow pray that some sort of useful global > ''emergent behavior'' will automatically instantiate itself? Not sure if it's desirable, but it is pretty cool. At the 7.6 press intro I think I mentioned before, there was us with WebPainter, the Wav people, and the Charts & Numbers people, and yeah, we dropped everything in, mixed and matched, and it pretty much worked and it was pretty cool. However, there were a LOT of interface issues that nobody had worked out. Moving vs. resizing vs. editing of an embedded part was clunky at best ... and I had to have it explained to me how it should work, and despite what you may hear I am generally not a complete idiot so if I couldn't figure out on my own how something works, chances are the average end-user (and definitely the mom-type end-user) would have no hope whatsoever. So whenever I hear people whining about the demise of OpenDoc, my immediate reaction is "Yeah, right, and how many commercial apps did YOU port to OpenDoc, buddy? So why again do you think you know what you're talking about then?" ---- Alex Curylo -- alex@witty.com
From: acurylo@inmediapresents.nospam.com (Alex Curylo) Newsgroups: comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer,comp.sys.mac.advocacy Subject: Re: More OpenDoc advocacy (was Re: How to save Apple Date: Wed, 30 Jul 1997 09:58:55 -0700 Organization: InMedia Presentations Message-ID: <acurylo-3007970958550001@van0407.tvs.net> References: <michael.b.moore.1220549990D@news.msfc.nasa.gov> <B00407CD-106DF2@207.93.51.100> In article <B00407CD-106DF2@207.93.51.100>, "Benjamin Smith" <benjs@ix.netcom.com> wrote: > times. OD is not failed technology because of itself, it really has to do > with who introduced it, when it was introduced and what big manufacturers > embraced it (or actually did not embrace it). OD is failed technology because Novell didn't deliver a Windows version on time. And IBM didn't deliver the OS/2 version on time. And OS/2 went nowhere anyway. Without the cross-platform abilities, its attractiveness lessened. I think many big outfits did > not embrace because it would have exposed just how inefficient and bloated > their software is. OD is failed technology because there is no business model to get rich off small efficient components. There is to get rich off inefficient bloated suites. Without a clear economic motivator, its attractivess lessened immensely. And it also opened the door for small nonames to get big > and save consumers money. But OD is great technology and it will show up in > essence sometime in the future, if by another name. Here's a stunning prediction for you: Component technology will never show up at the end-user level in any obvious fashion. However, component technology will underly the next generation of software suites. But it will be a development tool and a way to ease runtime resource requirements, not the mix and match parts by the user that the more wild-eyed frothers predict. Hell, at the 7.6 press intro I keep mentioning, there were only three of us with user-oriented productivity parts, and the first thing the suits started talking about was how to bundle all of us as a suite... ---- Alex Curylo -- alex@witty.com
From: dix.lorenz@hamburg.netsurf.de (Dix Lorenz) Newsgroups: comp.sys.next.programmer Subject: Re: C++ in framework? Date: Wed, 30 Jul 1997 20:11:17 +0200 Organization: EDV - Beratung Message-ID: <1997073020111715407511@dip104-2.hamburg.netsurf.de> References: <199707281715094377217@dip065-2.hamburg.netsurf.de> <5ripia$jal$2@owl.slip.net> <199707290328596592668@dip062-2.hamburg.netsurf.de> <vb204g62mu.fsf@skyler.arc.ab.ca> Brian Schack <schack@skyler.arc.ab.ca> wrote: > This sounds like a problem we struggled with for a while, caused by > the stupidity of the NT linker. Basically functions and global > variables are not exported properly from frameworks. At runtime, when > the framework tries to access a global variable, it uses an incorrect > address, and you get a "horrible crash". To get around it, you must > use a Windows heinosity, namely the __declspec(dllimport) and > __declspec(dllexport) declarations. Well, I have read up on DLL-creation under Windows now and I had thought this would be it, too, especially as it is explained in great length everywhere. Unfortunately, I don't have global variables or functions in my C++-framework, only classes and their methods (occasionaly a static method, but I think "unresolved: __vt$8DLStream" means the constructor, which clearly is defined). Which doesn't mean I haven't tried the "__declspec(dllexport)" for my classes. I have, but it didn't change anything. > > More details can be found at: > > http://www.next.com/NeXTanswers/HTMLFiles/2462.htmld/2462.html > > under the questions: > > "Why do I get a segmentation fault when trying to use a bundle?" > "How do I export symbols other than class names from my framework?" I had stumbled over this document before and it doesn't really help in my case. Either it answers problem at run-time or some special cases which have nothing to do with my VERY special case. Aargh! > Hope this helps. > > Brian Thanks anyway, Dix
From: Sajid Ahmed the Peaceman <peaceman@capital.net> Newsgroups: adass.iraf.programming,alt.comp.shareware.programmer,alt.msdos.programmer,comp.msdos.programmer,comp.os.msdos.programmer,comp.programming,comp.programming.contests,comp.sys.next.programmer,gfido.beruf.programmierer Subject: Re: NEED HELP (How can I remove the protection of my HD) Date: Wed, 30 Jul 1997 15:23:57 -0400 Organization: Logical Net Message-ID: <33DF94CD.729F@capital.net> References: <01bc9c55$8f88ed40$535047c3@cityweb.cww.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit To: Mitch <Pahli@cww.de> Mitch wrote: > > Hi, I´ve a big problem with my hard disk. I have two hard disks (<C> and > <D>). <D> doesn´t work for one year. FDISK recognize it with the addition > "PROTECT". I can´t read the disk, I can´t write to the disk and it´s > impossible to format the disk. Can you give me a tip how to manage it? How > could I remove the protection of the HD? > > Please help me !!!! > > Bye, Mitch Check your cmos settings. You should be able to get into it by pressing the del key when your computer first starts, or if you have a Phoenix Bios, ctrl-alt-esc, ctrl-alt-s, or ctrl-alt-ins . Peaceman
These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Marcel Waldvogel and Netfuture.ch.