This is Prog-06.gz in view mode; [Up]
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: <12797864532834@digifix.com> Date: 1 Jun 1997 03:57:29 GMT Organization: Digital Fix Development Message-ID: <26791865137630@digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - ISV company pages - ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news: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. news: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. news: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. news: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. news: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. news: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!!! news: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. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news: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. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news: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 The ftp sites ============= ftp://ftp.next.peak.org - The main site for North American submissions formerly ftp.cs.orst.edu ftp://ftp.informatik.uni-muenchen.de: - (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next - Peanuts mirror in the US ftp://terra.stack.urc.tue.nl - (Dutch NEXTSTEP User Group) 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: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Sun, 1 Jun 1997 00:57:56 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <EnYE5IS00iV1Q1Sokq@andrew.cmu.edu> References: <5mndjq$l7d$1@news.digifix.com> <AFB4A410-12B51B@206.165.44.84> Excerpts from netnews.comp.sys.next.advocacy: 30-May-97 Re: GX OOP class proposal &.. by "Lawson English"@primene >> Certainly not. However Lawson's GX Jihad is hardly GNU. GNU >> has some credibility, and even Stallman isn't as wacko as you. > > Is it wacko to believe that a retained mode interface is superior to an > immediate-mode graphics engine? Yes. Both models have advantages and disadvantages, and you can easily think of common circumstances where one is better suited for a particular role than the other. > Is it wacko to believe that Apple has made a BIG mistake in embracing DPS > over GX/Taligent? Yes. DPS works really well, and there are a large number of successful, best-of-breed applications which were written using DPS. Going by the standard of "are there any successful, best-of-breed apps available?", both GX and Taligent failed to deliver. > Is it wacko to believe that calling for like-minded individuals to work on > providing a GX-based text-handling class library useable for 20,000,000 > end-users is a worthwhile thing to do? Yes. GX, while it had some interesting ideas, was flawed in many respects to the point that essential functionality like printing didn't work. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc,comp.sys.mac.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 31 May 1997 23:26:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB66892-56630@206.165.44.88> References: <wnY7T7G00UhBA1ud5D@andrew.cmu.edu> nntp://news.primenet.com/comp.sys.next.programmer, nntp://news.primenet.com/comp.sys.mac.programmer.misc, 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: [color state grouping for optimization] > This is a relatively minor optimization issue, and it's bizzare that you > keep bringing this up as if it were a telling point. Going over the > details of how you optimize DPS drawing doesn't seem to have much > relevance. In the context of saying that one can simply lift algorithms used in a retained graphics engine and plunk them into an immediate mode engine that renders by making calls over a communications channel and expect equivalent performance automatically, I think it is *quite* relevant. There's a LOT of state-changes that can occur during the rendering of a GX layout shape and I doubt that algorithms that are fast using the retained-state info of GX are going to be fast when used in the APpKit. It goes both ways, of course. People have pointed out that the structure of the data found in the GX Layout shape may not be a good fit for the data structures used in text-oriented apps. OTOH, I know that DTP/word-processing apps using GX have been implemented and are shipping in various countries around the world, so it obviously isn't impossible. ++++++++++++++++++++ YAS [Yet Another Sig]
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc,comp.sys.mac.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 1 Jun 1997 06:38:28 GMT Organization: Digital Fix Development Message-ID: <5mr5d4$2ma$1@news.digifix.com> References: <wnY7T7G00UhBA1ud5D@andrew.cmu.edu> <AFB66892-56630@206.165.44.88> In-Reply-To: <AFB66892-56630@206.165.44.88> On 05/31/97, "Lawson English" wrote: >Charles William Swiger <cs4w+@andrew.cmu.edu> said: > >[color state grouping for optimization] > >> This is a relatively minor optimization issue, and it's bizzare that you >> keep bringing this up as if it were a telling point. Going over the >> details of how you optimize DPS drawing doesn't seem to have much >> relevance. > >In the context of saying that one can simply lift algorithms used >in a retained graphics engine and plunk them into an immediate >mode engine that renders by making calls over a communications >channel and expect equivalent performance automatically, I think >it is *quite* relevant. > You're wrong. The GX stuff we are talking about here is line-layout, specifically the positioning of various glyphs and such.. All that information is available on the AppKit side of the equation.... all of that positioning can be handled without traversing the communication channel to the DPS server. -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 31 May 1997 10:41:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB5B519-1DA6E@206.165.44.43> References: <5mpbja$d55@lehi.kuentos.guam.net> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer, nntp://news.primenet.com/comp.sys.mac.programmer.misc MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit crobato@kuentos.guam.net said: In <5mosb1$9v6@proxye1.san.rr.com>, "Ed Deans" <eadeans@ibm.net> writes: > > > >Sure it does but you should remember that BOTH were supported. GX was > >supposed to be the future so developers were supposedly fixing their apps > >to support it. Apple let GX die. Even with GX not have been wild adopted, > >developers had been told since its debut to plan for it--now it's plan for > >DPS, which is lacking in places that GX was not. > > > > This GX dying thing is a big clique by now. GX printing had little > support when everyone is standardizing on Display Postscript. GX > graphics routines however, are rolled into the Rhapsody graphics engine. > > "Everyone is standardizing on Display PostScript..." What everyone? Under the current Rhapsody graphics model, unless you need to do something unusually fast, you use the C wrappers to code to a QuickDraw-like model with floating point coordinates. There's also a Bezier curve class, but that is also a non-PS model. Where's the "standarizing on DPS" there? Also, what "everyone?" Unless you've committed to developing for Rhapsody-only, you're still using QuickDraw or GX on the MacOS side. (unless you happen to have a spare Display GhostScript-like interpreter lying around for use on the MacOS like Adobe, MacroMedia and a few other companies may). The point of the contest is two-fold: 1) to provide Rhapsody-text functionality (maybe even using the same class/method structure) to MacOS developers 2) to point out that implementing Rhapsody-text functionality using GX is easier than implementing GX-like typographical features using DPS + APpKit. <hint> <hint> Dr. Amelio... ------------------------------------------------------------------------- "Apple is handcuffed to a burning building and has started to chew off its right leg to survive." - David Brady referring to rumored Apple cuts. -------------------------------------------------------------------------
From: pretzl@pobox.com (Allan Peretz) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: Sun, 01 Jun 1997 13:44:08 GMT Organization: Planet Digital Network Technologies Message-ID: <33917b98.71663816@news.pdnt.com> References: <5m2u4j$2b5$3@news1.ucsd.edu> <338b6960.76768104@news.pdnt.com> <338B8919.6351@earthlink.net> <338B8EA4.1D57@earthlink.net> <AlvinKoh-2805971604500001@fsng172mac.ops.sing.paging.mot.com> <5mhvfv$k1i$1@news.digifix.com> <33908F5C.3B0A@wootech.com> On Sat, 31 May 1997 13:51:17 -0700, "Joseph A. Woo" <jwoo@wootech.com> wrote: >I have to agree with the above statement. It is ONLY $250.00 to become >an Apple Developer program member and is well worth the cost. You get >loads of stuff from Apple and OpenStep(minus WebObjects) for free. > >What are all of you complaining about?? GO PAY YOUR $250 AND STOP >COMPLAINING. Geez, those of you who want everything for free are being >rediculous. If you can't pay $250 now, save it and then pay if it is >that important to you. BTW: If you join the Apple Developer Program >now, then you will get Rhapsody Developer release and all other >pre-release software(as long as you sign the Apple NDA). Mr. Woo: You are incorrect. Joining the Dev. Program now will probably *not* get you a copy of the prelude. The supplies are extremely limited and are not anticipated to last as long as the 10 day wait for a Dev. Program membership. Also, I'm not a developer (unless you consider providing database and accounting system solutions to the business community development) so I don't see why I should have to join this program just to buy a reasonably priced copy of OS 4.2. Allan Peretz Information Technology Specialist Clifton Gunderson, L.L.C.
Newsgroups: comp.sys.next.programmer From: George Pipkin <gpp8p@Virginia.edu> Subject: Master/Detail in Rhapsody Prelude: an update Content-Type: text/plain; charset=iso-8859-1 Message-ID: <3391861E.707C1AA8@Virginia.edu> Sender: usenet@murdoch.acc.Virginia.EDU Organization: University of Virginia Mime-Version: 1.0 Date: Sun, 1 Jun 1997 14:24:30 GMT Like many other people, I have been playing around with the Rhapsody Prelude disks. I set up a simple database, and used the EO Modeler and the Interface Builderto create a simple application. Last week I noticed that the software didn't work as I anticipated - in particular, when I dragged a relationship from EO Modeler to a window, it didn't create a table object for the detail entity. Moreover, then I attempted to do this "manually" as described in the manual, the relationship didn't appear to be available in the Inspector to make the appropriate connection. Since then I have continued fooling around with in in odd hours, and I discovered that this phenemon appears to take place only when the databases you are working with are already populated with data. I have tested this on two different installations of the disk, and with trwo different adaptors: the Sybase adaptor, and the ODBC adaptor going to SQL anywhere. If the databases are non populoated with test daa, all appears to work as adveretised. I am surprised that I have not seen much mention of this on the various mailing lists since it would appear that this problem would prevent anyone from successfully going through the tutorials provided in the documentaion - particularly te Movie/Studio database if they had loaded the data before trying the tutorial as the documentation instructs them to. I would be interested to know if anybody had successfully done this. What I am attempting to determine is if this is a problem related to Sybase or adaptors going against Sybase or if this is something more general. Does this happen with Oracle users ? - George Pipkin
From: pressrelease@usa.net Newsgroups: comp.sys.next.programmer Subject: VISIT OUR SITE AND RECEIVE A $99.00 GIFT Date: 30 May 1997 19:55:43 GMT Organization: ABC, LLC Message-ID: <5mnbbv$7so$4@its.hooked.net> ------------------>>>>> FREE GIFTS!! -------------------->>>>>>>> FREE GIFTS!!! Come visit our WEB SITE & Receive $99.00 worth of FREE GIFTS! http://www.cyberbundle.com
From: mmalcolm crawford <m.crawford@shef.ac.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 1 Jun 1997 16:13:11 GMT Organization: University of Sheffield, UK Message-ID: <5ms72n$49$1@bignews.shef.ac.uk> References: <5mo5jb$t71$1@news.digifix.com> <AFB501C4-1CCD2@206.165.44.25> <5moesg$36b$1@news.digifix.com> <5mosb1$9v6@proxye1.san.rr.com> In-Reply-To: <5mosb1$9v6@proxye1.san.rr.com> On 05/31/97, "Ed Deans" wrote: > >>Lawson English <english@primenet.com> said: > >> > >>Talk about silly questions. Rhapsody doesn't give MacOS users > >>ANYTHING except (maybe) an upgrade path sometime in the NeXT 18-24 > >>months to obtain these oh-so-important buzzwords that were the > >>reason why they bought a Mac in the first place. > > You're pretty much right from what we know about Rhapsody. Where's the > innovation that will push Rhapsody into the fore of the market's mind so > Apple will survive? > Umm, was this rhetorical sarcasm? In case it wasn't, how about: (Innovation as far as MacOS goes): Preemptive multitasking; multi-threading; multi-user capability... Unified imaging model; OO development environment; unequalled cross-platform compatibility... and so on. > > Given developers faith that Apple can and will ship a new OS > >before the end of the Century? > > Even on its deathbed, the worst case for Copland was a ship date before Jan > 1, 2000. > Again it's not clear what you're implying here -- in case you're suggesting Rhapsody won't ship until then, I'm prepared to lay a bet that it will ship to the general public prior to end of 1998 (developers will get their hands on it by end Q3 this year) -- and this is a conservative estimate. Best wishes, mmalc. --
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5mnbbv$7so$4@its.hooked.net> Date: 1 Jun 1997 16:20:56 GMT Control: cancel <5mnbbv$7so$4@its.hooked.net> Message-ID: <cancel.5mnbbv$7so$4@its.hooked.net> Sender: pressrelease@usa.net Spam cancelled by sef@kithrup.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: GX OOP class proposal & contest (was Re: GX game graphics Date: 1 Jun 1997 10:50:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB708E2-1AD6B@206.165.44.34> References: <EnYE5IS00iV1Q1Sokq@andrew.cmu.edu> 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 Charles William Swiger <cs4w+@andrew.cmu.edu> said: Lawson English <english@primenet.com> said: > > Is it wacko to believe that a retained mode interface is superior to an > > immediate-mode graphics engine? > > Yes. Both models have advantages and disadvantages, and you can easily > think of common circumstances where one is better suited for a > particular role than the other. > Fair enough. The solution is to implement a hybrid that uses the best of both worlds. The current implementation of DPS does NOT look like it would be a good basis for such a hybrid engine. > > Is it wacko to believe that Apple has made a BIG mistake in embracing > DPS > > over GX/Taligent? > > Yes. DPS works really well, and there are a large number of successful, > best-of-breed applications which were written using DPS. > > Going by the standard of "are there any successful, best-of-breed apps > available?", both GX and Taligent failed to deliver. Ready, Set, Go, GX consistently gets high marks in reviews and is the only Chinese DTP system for the Mac that I'm aware of. The Arabic-localized version will be out shortly, I've heard. Tailor GX was implemented using GX. Typistry was implemented using GX. LightningDraw GX gets high marks in its price-range. Electrifier has been named one of the top 32 Netscape plug-ins, I understand (a GX-based plug-in). Others exist, I'm sure... > > > Is it wacko to believe that calling for like-minded individuals to work on > > providing a GX-based text-handling class library useable for 20,000,000 > > end-users is a worthwhile thing to do? > > Yes. GX, while it had some interesting ideas, was flawed in many > respects to the point that essential functionality like printing didn't > work. Well, in fact, I've got GX installed on my system. Some non-GX apps require me to temporarily disable GX Printing before they'll print. Others will attempt to patch "Classic Printing" and will crash the system if I attempt to print from them (these I tend to avoid using whenever possible). Others will convert QuickDraw constructs into GX shapes and print just fine. Never had a problem with a GX-specific app printing. Which means that all one needs to do to "fix" GX printing is convert it to shared libraries that provide a standardized interface to "Classic Printing" for GX-savvy apps rather than trying to replace Classic Printing with GX. Once that is done, the problems with GX Printing automatically go away because only those apps designed to look for it will interact with it in any way. That means that both GX and NON-GX apps that can print to the GX Printing model can still gain benefits from doing so. One could even provide *auxilliary* info as a set of resources in a given Classic Print driver that could be used to provide many of the current GX Printing services, again, without interfering in any way with those apps that aren't happy with GX Printing. Those problems that exist with GX Printing are completely fixable using my suggestion above and would provide great benefit to any System 7.x users with enough RAM to handle GX Printing (which is quite a few, these days). The vast majority of users will be using MacOS for many, MANY years. To cripple the current capabilities rather than fixing them suggests that Apple management is wearing blinders because enhancements to the CURRENTLY SHIPPING OS are what is going to save Apple long enough for Rhapsody to even ship. -------------------------------------------------------------------------- Any sufficiently advanced magic will be indistinguishable from technology -my corollary to Clarke's Law (AFAIK Mercedes Lackey got it from me at a World Fantasy Convention) --------------------------------------------------------------------------
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc,comp.sys.mac.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 1 Jun 1997 11:01:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB70B51-23FBA@206.165.44.34> References: <5mr5d4$2ma$1@news.digifix.com> nntp://news.primenet.com/comp.sys.next.programmer, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.mac.advocacy MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Scott Anguish <sanguish@digifix.com> said: Lawson English <english@primenet.com> said: > >In the context of saying that one can simply lift algorithms used > >in a retained graphics engine and plunk them into an immediate > >mode engine that renders by making calls over a communications > >channel and expect equivalent performance automatically, I think > >it is *quite* relevant. > > > > You're wrong. > > The GX stuff we are talking about here is line-layout, > specifically the positioning of various glyphs and such.. > > All that information is available on the AppKit side of the > equation.... all of that positioning can be handled without traversing > the communication channel to the DPS server. This doesn't address the question: what is a GX glyph? GX text shapes can apply typefaces, which include *layers* of predefined variations of a given glyph, applied on a glyph-by-glyph basis to the formatted text of the GX Layout shape. The resultant hybrid glyph info is almost certainly stored as bitmaps within the layout shape. While you could create a new PS font for each style-run that uses typefaces in the layout shape, that would be a bit time-consuming, I suspect, although it might save space in the long run if a single style-run were used for every glyph. Then there's the issue of constantly changing faces, fonts, languages, sytles, etc. In the most pathological case, that info could change with every glyph, making the communications latency a major issue instead of a minor one. +++++++++++++++++++++++++++++++++ ImagineAsig (tm)
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.next.software,comp.sys.mac.advocacy Subject: Re: Prelude WebObjects serial number? Date: 1 Jun 97 14:36:11 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFB73964-8E6749@141.214.134.235> References: <5mnj48$n2h$1@news.digifix.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: "Scott Anguish" <sanguish@digifix.com> nntp://news.itd.umich.edu/comp.sys.next.programmer, nntp://news.itd.umich.edu/comp.sys.next.software, nntp://news.itd.umich.edu/comp.sys.mac.advocacy On Fri, May 30, 1997 6:08 PM, Scott Anguish <mailto:sanguish@digifix.com> wrote: > I believe its on the back of the envelope that the CD came >in.. > I'd suggest immediately transferring it to the CD using a >Sharpie magic marker.. I, and a coworker that was at the conference with me, made the mistake of immediately transferring the WebObjects cd to the plastic carrying case that the other Prelude cds were in. Does anyone know if the serial numbers are unique to each cd? There were a couple other people from the U of MI that were at the conference. I could try to use their number. rob -- <mailto: "Robert A. Decker" comrade@umich.edu> <http://hmrl.cancer.med.umich.edu/Rob/index.ssi> Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #2: "Has anyone seen Chris? I have some last minute instructions for the scene where he wrestles the evil monkey." -Get a Life
From: klui@cup.hp.com (Ken Lui) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 1 Jun 1997 19:33:32 GMT Organization: Hewlett-Packard Company Message-ID: <5msiqc$dbp$1@hpax.cup.hp.com> References: <4095467485.51634735@softarc.com> <AFB4698E-4F5EF@206.165.44.84> In article <AFB4698E-4F5EF@206.165.44.84>, Lawson English <english@primenet.com> wrote: >GX, on the other hand, DOES support bi-directional text, even in the >simplest of its text-shapes. This feature isn't specific to GX, because GX isn't required to do bidirectional text handling available via hebrew language kit. Ken -- Ken Lui, klui@cup.hp.com 19111 Pruneridge Avenue M/S 44UR Enterprise Systems Division Cupertino, CA 95014-0795 USA Open Warehouse Team 1.408.447.3230 FAX 1.408.447.1053 Views within this message may not be those of the Hewlett-Packard Company
From: klui@cup.hp.com (Ken Lui) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 1 Jun 1997 19:51:18 GMT Organization: Hewlett-Packard Company Message-ID: <5msjrm$dg5$1@hpax.cup.hp.com> References: <5mngel$p6t@shelob.afs.com> <AFB4D089-11167@206.165.44.62> In article <AFB4D089-11167@206.165.44.62>, Lawson English <english@primenet.com> wrote: >And how well does it handle bi-directional text? Arabic? Vertical Chinese? >Vertical English (for that matter)? The wonderful thing with objects is once Rhapsody supports bi-directional and vertical text, all apps will have 'em. That capability shouldn't be a concern for AFS because Apple has said they will roll much of Mac's int'l capabilities into Rhapsody. I would welcome vertical text, but it's not a prerequisite unless one's doing traditional manuscripts. A lot of data is done right-to-left, top- to-bottom these days, in China and Japan. Though it would be good to have top-to-bottom, right-to-left as well as right-to-left, top-to-bottom. Ken -- Ken Lui, klui@cup.hp.com 19111 Pruneridge Avenue M/S 44UR Enterprise Systems Division Cupertino, CA 95014-0795 USA Open Warehouse Team 1.408.447.3230 FAX 1.408.447.1053 Views within this message may not be those of the Hewlett-Packard Company
From: klui@cup.hp.com (Ken Lui) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 1 Jun 1997 19:59:21 GMT Organization: Hewlett-Packard Company Message-ID: <5mskap$dkg$1@hpax.cup.hp.com> References: <5mnd95$l45$1@news.digifix.com> <AFB49EC1-1175B4@206.165.44.84> In article <AFB49EC1-1175B4@206.165.44.84>, Lawson English <english@primenet.com> wrote: > NeXT's Internationalization is far superior to MacOS's >and GX's from what NeXT developers are constantly claiming. > >Does anyone but me fail to note the logic-error in the above assertion >about NeXT's superior Internationalization support? I'd say NeXT's foundation for achieving i18n is superior to that available on Macintosh today. Feature-wise, the Mac is ahead, but these features are available without GX. Ken -- Ken Lui, klui@cup.hp.com 19111 Pruneridge Avenue M/S 44UR Enterprise Systems Division Cupertino, CA 95014-0795 USA Open Warehouse Team 1.408.447.3230 FAX 1.408.447.1053 Views within this message may not be those of the Hewlett-Packard Company
From: cstory@research.canon.com.au Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: Mon, 02 Jun 1997 07:56:11 +1000 Organization: Canon Information Systems Research Australia Message-ID: <3391EFFB.552E@research.canon.com.au> References: <5m2u4j$2b5$3@news1.ucsd.edu> <338b6960.76768104@news.pdnt.com> <338B8919.6351@earthlink.net> <338B8EA4.1D57@earthlink.net> <AlvinKoh-2805971604500001@fsng172mac.ops.sing.paging.mot.com> <5mhvfv$k1i$1@news.digifix.com> <33908F5C.3B0A@wootech.com> <33917b98.71663816@news.pdnt.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit > You are incorrect. Joining the Dev. Program now will probably *not* > get you a copy of the prelude. The supplies are extremely limited and > are not anticipated to last as long as the 10 day wait for a Dev. > Program membership. Also, I'm not a developer (unless you consider > providing database and accounting system solutions to the business > community development) so I don't see why I should have to join this > program just to buy a reasonably priced copy of OS 4.2. If you're not a developer, you don't need it. The point of the Prelude is to introduce developers to the tools that will be included in Rhapsody DR (Developer's Release). It's to encourage software development for Rhapsody. If you're not a developer, you don't really fit in. Apple isn't pushing NextStep. BTW, the notice I got about the Prelude was that it was limited to members of the seed program, a subset of the developer program (although I suspect a dense subset). Cliff
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.lang.objective-c,comp.sys.next.programmer,comp.sys.next.advocacy Subject: Books? Date: 1 Jun 97 19:33:44 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFB77F1A-9EC97B@141.214.134.235> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit nntp://news.itd.umich.edu/comp.sys.next.programmer, nntp://news.itd.umich.edu/comp.sys.next.advocacy Can anyone recommend some good Objective C and OpenStep programming books? I couldn't find any at the Addison Wesley web site, but I would like books of their quality (It's gotten to the point where I really don't trust anyone but AW). I would like an Objective C programming language reference book. I would also like some OpenStep books. I would particularly like to have a book that has all the classes in alphabetical order and a brief description of their methods and with tons of sample code (a perfect example of this is the AW book called 'Java Class Libraries - an Annotated Reference'). 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: alex@WebIS.net (Alex Kac) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Sun, 01 Jun 1997 20:44:04 -0600 Organization: Web Information Solutions---Interactive and database web design studio Message-ID: <alex-0106972044040001@192.168.1.3> References: <EnYE5IS00iV1Q1Sokq@andrew.cmu.edu> <AFB708E2-1AD6B@206.165.44.34> In article <AFB708E2-1AD6B@206.165.44.34>, "Lawson English" <english@primenet.com> wrote: : Charles William Swiger <cs4w+@andrew.cmu.edu> said: : Lawson English <english@primenet.com> said: : > Yes. DPS works really well, and there are a large number of successful, : > best-of-breed applications which were written using DPS. : > : > Going by the standard of "are there any successful, best-of-breed apps : > available?", both GX and Taligent failed to deliver. : : Ready, Set, Go, GX consistently gets high marks in reviews and is the only : Chinese DTP system for the Mac that I'm aware of. The Arabic-localized : version will be out shortly, I've heard. Tailor GX was implemented using : GX. Typistry was implemented using GX. LightningDraw GX gets high marks in : its price-range. Electrifier has been named one of the top 32 Netscape : plug-ins, I understand (a GX-based plug-in). Others exist, I'm sure... None of those are best of breed. *I cannot use Read, Set Go in any form (GX or not) to do any of my work well. PageMaker or Quark please. *Tailor first came out on NeXT and had many problems with GX... *LightningDraw again does not have the feature set of either Illustrator or Freehand. Definately NOT best of breed, though it DID have nice features related to transparency. *I hate electrifier. It is a horrible little plugin that does nothing good on the web. I have not seen one web site that uses it that I have remarked, wow, that looks cool. I refuse to go to any site that asks me to use it. : : Well, in fact, I've got GX installed on my system. Some non-GX apps require : me to temporarily disable GX Printing before they'll print. Others will : attempt to patch "Classic Printing" and will crash the system if I attempt : to print from them (these I tend to avoid using whenever possible). Others : will convert QuickDraw constructs into GX shapes and print just fine. : : Never had a problem with a GX-specific app printing. You know, that means nothing. Of course apps written for GX will work fine, but 99% of programs aren't. And there are definately NO programs written that use GX that are worth using when quality is on the line. -- Web Information Solutions develops interactive and database driven websites. For more information, go to <http://www.WebIS.net>
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.lang.objective-c,comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Books? Date: 2 Jun 1997 04:00:19 GMT Organization: Digital Fix Development Message-ID: <5mtggj$rr3$1@news.digifix.com> References: <AFB77F1A-9EC97B@141.214.134.235> In-Reply-To: <AFB77F1A-9EC97B@141.214.134.235> On 06/01/97, "Robert A. Decker" wrote: > > Can anyone recommend some good Objective C and OpenStep programming >books? I couldn't find any at the Addison Wesley web site, but I would like >books of their quality (It's gotten to the point where I really don't trust >anyone but AW). > > I would like an Objective C programming language reference book. > > I would also like some OpenStep books. I would particularly like to have >a book that has all the classes in alphabetical order and a brief >description of their methods and with tons of sample code (a >perfect example of this is the AW book called 'Java Class Libraries >- an Annotated Reference'). > There is a list of the available reading materials at http://www.stepwise.com/Resources/Books The quality of the two OpenStep books that are currently out there is very high, Developing Business Apps on OpenStep from Nik Gervae and Pete Clark... and Openstep for Enterprises by Nancy Craighill. The Preface of the first is on Stepwise at that location.. Other than those references, get the PDF documents from www.next.com , the quality of NeXT's documentation is good to excellent, especially on Obj-C.. -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 04:07:50 GMT Organization: Digital Fix Development Message-ID: <5mtgum$s25$1@news.digifix.com> References: <EnYE5IS00iV1Q1Sokq@andrew.cmu.edu> <AFB708E2-1AD6B@206.165.44.34> In-Reply-To: <AFB708E2-1AD6B@206.165.44.34> On 06/01/97, "Lawson English" wrote: >Charles William Swiger <cs4w+@andrew.cmu.edu> said: >Lawson English <english@primenet.com> said: > >> > Is it wacko to believe that a retained mode interface is superior to an >> > immediate-mode graphics engine? >> >> Yes. Both models have advantages and disadvantages, and you can easily >> think of common circumstances where one is better suited for a >> particular role than the other. >> > >Fair enough. The solution is to implement a hybrid that uses the >best of both worlds. The current implementation of DPS does NOT >look like it would be a good basis for such a hybrid engine. > Then kiss Apple goodbye... They don't have the extra time for this. Especially when the returns would be little if any. >> > Is it wacko to believe that Apple has made a BIG mistake in embracing >> DPS >> > over GX/Taligent? >> >> Yes. DPS works really well, and there are a large number of successful, >> best-of-breed applications which were written using DPS. >> >> Going by the standard of "are there any successful, best-of-breed apps >> available?", both GX and Taligent failed to deliver. > >Ready, Set, Go, GX consistently gets high marks in reviews and is >the only Chinese DTP system for the Mac that I'm aware of. The >Arabic-localized version will be out shortly, I've heard. Tailor >GX was implemented using GX. Typistry was implemented using GX. >LightningDraw GX gets high marks in its price-range. Electrifier >has been named one of the top 32 Netscape plug-ins, I understand >(a GX-based plug-in). Others exist, I'm sure... > FreeHand.. Illustrator... PageMaker... PhotoShop... Canvas.... PowerCadd... All best of breed apps, the first four are MUST HAVE apps for DTP... none require GX. The list of high-quality visual apps that DON'T use GX is long and distinguished... That doesn't mean the others aren't good apps, but it does mean that the Mac can succeed without GX... -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc,comp.sys.mac.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Followup-To: comp.sys.next.advocacy,comp.sys.mac.advocacy Date: 2 Jun 1997 04:13:34 GMT Organization: Digital Fix Development Message-ID: <5mth9e$s2b$1@news.digifix.com> References: <5mr5d4$2ma$1@news.digifix.com> <AFB70B51-23FBA@206.165.44.34> In-Reply-To: <AFB70B51-23FBA@206.165.44.34> Note that I've adjusted the followups to remove the next..programmer group.. Its safe to assume that they don't want to hear more of Lawson's GX rant.... On 06/01/97, "Lawson English" wrote: > >Scott Anguish <sanguish@digifix.com> said: >Lawson English <english@primenet.com> said: > >> >In the context of saying that one can simply lift algorithms used >> >in a retained graphics engine and plunk them into an immediate >> >mode engine that renders by making calls over a communications >> >channel and expect equivalent performance automatically, I think >> >it is *quite* relevant. >> > >> >> You're wrong. >> >> The GX stuff we are talking about here is line-layout, >> specifically the positioning of various glyphs and such.. >> >> All that information is available on the AppKit side of the >> equation.... all of that positioning can be handled without traversing >> the communication channel to the DPS server. > >This doesn't address the question: what is a GX glyph? > Who gives a Rat's Ass(TM)? >GX text shapes can apply typefaces, which include *layers* of >predefined variations of a given glyph, applied on a glyph-by-glyph >basis to the formatted text of the GX Layout shape. The resultant >hybrid glyph info is almost certainly stored as bitmaps within >the layout shape. While you could create a new PS font for each >style-run that uses typefaces in the layout shape, that would be >a bit time-consuming, I suspect, although it might save space in >the long run if a single style-run were used for every glyph. > You are making wild assumptions on how GX is implementing its workings, absolutely NONE of which has any basis of how they actually decide to position the various components.. >Then there's the issue of constantly changing faces, fonts, >languages, sytles, etc. In the most pathological case, that info >could change with every glyph, making the communications latency >a major issue instead of a minor one. > No it doesn't.... To position the glyphs does NOT necessarily require round trips to the windowserver... -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: Timothy Luoma <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: Dual dev 3.3 & 4.1 Followup-To: poster Date: Sun, 1 Jun 1997 20:38:18 -0700 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.SUN.3.96.970601203335.6249A-100000@kira> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Has anyone here setup a dual-developer system where they can develop 3.3 & 4.1? I used the 'dual developer' setup by Christian Starkjohann <cs@hal.kph.tuwien.ac.at> but it didn't seem to quite work -- under 3.3 I get weird message that Christian thought might indicate cc thought it was being debugged rather than used as a compiler... Next token is 59 (';') Reducing via rule 162 (line 998), -> maybe_attribute state stack now 0 1 3 29 73 190 321 304 324 Entering state 444 Next token is 59 (';') Reducing via rule 158 (line 975), declarator maybeasm maybe_attribute -> initdcl state stack now 0 1 3 29 73 190 321 Entering state 442 Reducing via rule 151 (line 948), initdecls ',' initdcl -> initdecls state stack now 0 1 3 29 73 Entering state 190 Next token is 59 (';') Shifting token 59 (';'), Entering state 320 Reducing via rule 13 (line 274), typed_declspecs setspecs initdecls ';' -> datadef state stack now 0 1 3 Entering state 26 and they just go on and on and on.... I am also getting weird files such as these: asroot.c.combine asroot.c.cse asroot.c.cse2 asroot.c.dbr asroot.c.flow asroot.c.fppc asroot.c.greg asroot.c.jump asroot.c.jump2 asroot.c.loop asroot.c.lreg asroot.c.rtl asroot.c.sched asroot.c.sched2 asroot.c.stack Compiling under 4.1 works fine.... Anyone got a good guess? Replies via email especially appreciated, will post solution once I figure it out (with attribs to those who guide the process, of course) TjL -- TjL <luomat@peak.org> / http://www.peak.org/~luomat/next/ "The best things in life are made into inferior versions and bundled with the latest Microsoft systems" NS/OS users: My 'other sites' page has been entirely reworked
From: leonvs@occam.com (Leon von Stauber) Newsgroups: comp.lang.objective-c,comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Books? Date: 2 Jun 1997 05:22:11 GMT Organization: Occam's Razor Message-ID: <5mtla3$go8$2@hackberry.zilker.net> References: <AFB77F1A-9EC97B@141.214.134.235> Cc: comrade@umich.edu In <AFB77F1A-9EC97B@141.214.134.235> "Robert A. Decker" wrote: > > Can anyone recommend some good Objective C and OpenStep programming >books? I couldn't find any at the Addison Wesley web site, but I would like >books of their quality (It's gotten to the point where I really don't trust >anyone but AW). An excellent place to start is http://www.stepwise.com/Resources/Books/ > I would like an Objective C programming language reference book. Good luck. The only dedicated Obj-C book I've used is one by Pinson & Weiner (published by Addison/Wesley), but I wouldn't recommend it. ____________________________________________________________________ Leon von Stauber http://www.occam.com/leonvs/ Occam's Razor, Game Designer <leonvs@occam.com> PSW Technologies, System Administrator <leonvs@pswtech.com> MIDS, Web Developer <leonvs@mids.org> "We have not come to save you, but you will not die in vain!"
From: rex@mit.edu (Eric King) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Mon, 02 Jun 1997 00:37:44 -0500 Organization: All USENET -- http://www.Supernews.com Message-ID: <rex-0206970037450001@accs-as17-dp10.nwrk.grid.net> References: <5mn8j9$ol3@shelob.afs.com> <AFB48643-BB4B9@206.165.44.84> <5mndjq$l7d$1@news.digifix.com> In article <5mndjq$l7d$1@news.digifix.com>, sanguish@digifix.com (Scott Anguish) wrote: ) Then again, Mac developers who are going to deploy on Rhapsody )are likely busy learning the system and designing killer apps instead )of obsessing on writing APIs around a dead imaging system. Suppose a developer wants to create a killer multimedia app for Rhapsody. Obviously they should consider QTML, but chances are Apple's going to be providing a different (and probably much cleaner) OO interface to Quicktime 3.0 for Rhapsody. Sure they could rough out the interface with the existing OpenStep tools, but without QTML, it's going to be hard to proceed at the moment. The situation worsens if the developer wants to use QT 3.0's nifty new vector animation track. It supports antialiasing, transparency, and other neat stuff but requires its data to be in a certain format. Unfortunately, popular graphics apps like Illustrator and Freehand don't currently support this format and no announcements have been made as to whether they'll support it in the future; its quite a bit removed from their own PS-based formats. The 'preferred' way of producing vector shapes in this format is by using a translator for shapes in that accursed GX format. Furthermore there isn't an API for creating or manipulating these shapes either. Just the data format. Since, in your words, GX is 'Dead,' how is a developer supposed to create a killer app using this latest bit of nifty Apple tech? I'm sure there are all sorts of nifty 'killer' apps that could be made with this system. Ranging from low-bandwidth Web page animators to After Effects-like compositing apps to games. There's a lot of potential here. Unfortunately, DPS and OpenStep are amazingly useless in this instance. It's *far* easier to convert a gx-like shape to Postscript than the other way around. Further there isn't a high-level 2D graphics framework in OpenStep. There isn't even the simple abstraction of a shape. Since GX graphics isn't being integrated and apparently neither are the Taligent classes (though they're now going to be included with Java) an app developer is faced with the rather daunting task of rolling their own graphics library. That's a few months of extra coding and debugging which could otherwise be spent adding more features or shortening the time until release. If a developer wants to develop an insanely great graphics app for Rhapsody, they're probably going to have to wait a while. There's not enough to work with at the moment. ) Yes. GX is Dead. Ironically, the only apps that have announced support for Quicktime's new vector media layer are dependent on GX Graphics. I guess other developers shouldn't bother trying to support the new API because one of the best tools available for working with it is 'Dead' How long until Apple develops something to fill the void left by GX Graphics? 1 year? 2? 3? ) Its unlikely that many additional developers are going to be )creating frameworks for a rarely installed graphics sub-system that is )DEAD. The beauty of GX, is that to do what Lawson asked wouldn't require *many* only a handful working part time. (And there are certainly more than a handful hanging out on the gxlist) Apple's done the vast majority of work that needs to be done already. If you consider the Taligent's stuff, Apple's already written a better system than what they're planning for Rhapsody. Personally, I think it's very telling that Sun, a long-time OpenStep and DPS proponent chose to license Taligent's 2D Graphics and Text frameworks for Java. I guess they felt that their OpenStep implementation and the PS-like Bravo code from Adobe didn't contribute enough functionality in those areas. -Eric -- Excerpt from Marianne Moore's 'The Pangolin' This near artichoke with head and legs and grit-equipped gizzard, the night miniature artist engineer is, yes, Leonardo da Vinci's replica- impressive animal and toiler for whom we seldom hear.
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: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 00:41:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB7CB89-DFCD@206.165.44.52> References: <5mtgum$s25$1@news.digifix.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Scott Anguish <sanguish@digifix.com> said: > > FreeHand.. Illustrator... PageMaker... PhotoShop... > > Canvas.... PowerCadd... > > All best of breed apps, the first four are MUST HAVE apps for > DTP... none require GX. The list of high-quality visual apps that > DON'T use GX is long and distinguished... > > That doesn't mean the others aren't good apps, but it does > mean that the Mac can succeed without GX... I understand that Jobs suggested that Adobe's lack of commitment to Rhapsody was a "3rd party opportunity." If GX *facilitates* the creation of such apps (and it does), wouldn't it make sense for Apple to put a LEETLE more work into a retained mode engine for Rhapsody (not to mention MacOS)? ++++++++++++++++++++++++++++++ HOW DO I SUBSCRIBE TO GX-TALK? --------------------------------- To subscribe to GX-TALK proceed as follows: Send E-mail to: <gx-talk-request@aimed.org> with anything in the subject line and the following command as the first (and only) line of the message body: SUBSCRIBE +++++++++++++++++++++++++++++++++
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc,comp.sys.mac.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 00:43:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB7CC0D-FEF4@206.165.44.52> References: <5mth9e$s2b$1@news.digifix.com> nntp://news.primenet.com/comp.sys.next.programmer, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.mac.advocacy MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Scott Anguish <sanguish@digifix.com> said: > > >Then there's the issue of constantly changing faces, fonts, > >languages, sytles, etc. In the most pathological case, that info > >could change with every glyph, making the communications latency > >a major issue instead of a minor one. > > > > No it doesn't.... To position the glyphs does NOT > necessarily require round trips to the windowserver... But if the one-way trip is made many, MANY times in a document, as can happen with multi-lingual, multi-etc text, then even that "trivial" optimization issue becomes less-than-trivial. -------------------------------------------------------------------------- Any sufficiently advanced magic will be indistinguishable from technology -my corollary to Clarke's Law (AFAIK Mercedes Lackey got it from me at a World Fantasy Convention) --------------------------------------------------------------------------
From: mmalcolm crawford <malcolm@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 10:22:32 GMT Organization: P&L Systems Message-ID: <5mu6t8$hbs$1@ironhorse.plsys.co.uk> References: <5mtgum$s25$1@news.digifix.com> <AFB7CB89-DFCD@206.165.44.52> In-Reply-To: <AFB7CB89-DFCD@206.165.44.52> On 06/02/97, "Lawson English" wrote: >Scott Anguish <sanguish@digifix.com> said: >> FreeHand.. Illustrator... PageMaker... PhotoShop... >> >> All best of breed apps, the first four are MUST HAVE apps for >> DTP... none require GX. The list of high-quality visual apps that >> DON'T use GX is long and distinguished... >> >I understand that Jobs suggested that Adobe's lack of commitment to >Rhapsody was a "3rd party opportunity." If GX *facilitates* the creation of >such apps (and it does) > Maybe it doesn't, or isn't necessary on OpenStep, cf Create... PasteUp... TIFFany... Best wishes, mmalc. --
From: mmalcolm crawford <malcolm@plsys.co.uk> Newsgroups: comp.lang.objective-c,comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Books? Date: 2 Jun 1997 13:01:37 GMT Organization: P&L Systems Message-ID: <5mug7h$jvs$1@ironhorse.plsys.co.uk> References: <AFB77F1A-9EC97B@141.214.134.235> In-Reply-To: <AFB77F1A-9EC97B@141.214.134.235> On 06/02/97, "Robert A. Decker" wrote: > > Can anyone recommend some good Objective C and OpenStep programming >books? > Others have already given places to look; I'd just like to put in a word for NeXT's documentation, which is freely available (online at least, you could also buy hard copy, cf www.next.com) and *very* good. Read through that first before going to the bookshop -- the other references available are probably not quite what you want as a "beginner". Best wishes, mmalc. --
From: stephane@lysis.ch (Stephane Corthesy) Newsgroups: comp.sys.next.programmer Subject: Re: NSTabView? Date: 2 Jun 97 14:41:51 GMT Organization: Lysis S.A. Message-ID: <3392dbaf.0@news.planet.ch> References: <weesh-3005970658040001@user-37kbm6n.dialup.mindspring.com> Cc: weesh@mindspring.com In <weesh-3005970658040001@user-37kbm6n.dialup.mindspring.com> Kenneth H. Wieschhoff wrote: > Is there a tabbed view of some sort available? > > ->Ken > I made a brand new version, following the work done in the MiscKit. I've sent it to misckit, but it seems it will not appear in the next release. I will resend it to Don. If you want the new palette, send me an email. Stephane -- "L'ordinateur n'a pas l'intelligence qu'on lui prete, mais celle qu'on lui donne." Stephane Corthesy Lysis S.A. Rue des Cotes de Montbenon 8 CH-1003 Lausanne Switzerland Tel. +41.21.312.91.91 Fax +41.21.312.93.43 E-mail: stephane@lysis.ch (NeXTMail welcome)
From: Greg_Anderson@afs.com (Gregory H. Anderson) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 15:30:35 GMT Organization: Anderson Financial Systems Inc. Message-ID: <5muour$9kg@shelob.afs.com> References: <5mth9e$s2b$1@news.digifix.com> Scott Anguish writes > Note that I've adjusted the followups to remove the next..programmer > group. Its safe to assume that they don't want to hear more of > Lawson's GX rant.... Sorry Scott, I restored it because I'm about to make a programming point. The One Whose Name I Cannot Type wrote: > >Then there's the issue of constantly changing faces, fonts, > >languages, sytles, etc. In the most pathological case, that info > >could change with every glyph, making the communications latency > >a major issue instead of a minor one. and Scott replied: > No it doesn't.... To position the glyphs does NOT > necessarily require round trips to the windowserver... You can take out the "necessarily", because it just DOESN'T. Font info is cached locally in the AppKit, so determining character bounding boxes, glyph advancement, and kerning adjustments does NOT ping the windowserver during the course of text layout. To draw, you moveto a starting location, then just send a succession of xyshow operators, with setcolor or setfont operations mixed in when some characteristic of the run changes. None of these operations require a round trip. I mention xyshow because it raises an interesting point that has been overlooked by the Agent Provocateur. PostScript is entirely neutral on the topic of text direction. EVERY text handling operation accepts and/or returns x AND y offsets, which may be position or negative. As far as PS is concerned, text may run in ANY direction, even diagonally! From the standpoint of font metrics, when you ask how far the current point moves for a given glyph, Western fonts return positive x's and 0 y's. Hebrew fonts return negative x's. Chinese fonts return 0 x's, but positive y's. What's important is whether a particular text engine is prepared to adjust for these factors, and can translate other paragraph layout attributes -- like tracking and inter-line spacing -- for reasonable usage in alternate geometries. But now we're into the domain of "did the author care about non-Western cases". It has NOTHING to do with PostScript itself. -- 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: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy Subject: Rhapsody webobjects problem Date: 2 Jun 97 14:49:01 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFB88DE5-DE6E60@141.214.134.235> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: rhapsody-dev-feedback@apple.com, webobjects@omnigroup.com nntp://news.itd.umich.edu/comp.sys.next.advocacy I understand that we don't receive support for our Rhapsody cd's. Can you forward this to the rhapsody evangelist if you can't help me? I've been trying to get WebObjects (which I received at WWDC) running. However, it appears that some libraries aren't installed. I've reinstalled webobjects, openstep user, and openstep developer and the libraries still aren't present. Here's the errors in console that I get: Software Version OPENSTEP 4.2 for Mach (Lantern5V) probing for DOS dyld: /NextDeveloper/Apps/WebObjectsBuilder.app/WebObjectsBulder can't open the library: /NextLibrary/Frameworks/EOAccess.framework/Versions/B/EOAccess (No such file or directory. errno = 2) dyld: /NextDeveloper/Apps/WebObjectsBuilder.app/WebObjectsBulder can't open the library: /NextLibrary/Frameworks/EOAccess.framework/Versions/B/EOAccess (No such file or directory. errno = 2) dyld: /NextDeveloper/Apps/WebObjectsBuilder.app/WebObjectsBulder can't open the library: /NextLibrary/Frameworks/EOAccess.framework/Versions/B/EOAccess (No such file or directory. errno = 2) dyld: /NextDeveloper/Apps/WebObjectsBuilder.app/WebObjectsBulder can't open the library: /NextLibrary/Frameworks/EOAccess.framework/Versions/B/EOAccess (No such file or directory. errno = 2) I have no directory named 'EOAccess.framework'. Where should this have been installed from? 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: bettis@inetnebr.com (Mr. Jeremy Bettis) Newsgroups: comp.sys.next.software,comp.sys.next.programmer Subject: Re: OS Enterprise for NT? Date: 2 Jun 1997 14:19:52 -0500 Organization: Internet Nebraska Message-ID: <5mv6co$s7b$1@falcon.inetnebr.com> References: <5mf6uu$ghq$1@bvadm.bv.tek.com> <5mi0dn$t0i$1@falcon.inetnebr.com> <5mi0m5$kev$1@news.digifix.com> <5miet8$kvv$1@falcon.inetnebr.com> <5miu3i$146$1@news.digifix.com> NNTP-Posting-User: bettis sanguish@digifix.com (Scott Anguish) writes: >On 05/28/97, Mr. Jeremy Bettis wrote: >>sanguish@digifix.com (Scott Anguish) writes: >>> Its important to note also that when Rhapsody ships the >>>Runtime costs drop to ZERO. >> >>On WinNT? Deploying a product on Rapsody is going to be just >>as hard as NeXTSTEP was. No one will want it unless they can run >>all of their favorite Winders programs on the same computer. >> > I question this assertion... but I believe this is only made >because you don't understand what the Runtime is... This may only be an issue of nominclature, but I do understand what the runtime is. I am not sure I understand what Rapsody is. Correct me if this is wrong. OpenStep Enterprise is a runtime/development kit under Windows NT. Rapsody is an operating system that will run on PowerPC and i386 computers, be based on some MACH/BSD 4.something kernel. > First off, the Runtime costs I was discussing are for NT. Good, I did not relalise that you were saying that the OpenStep Enterprise runtimes go to $0 when Rapsody ships.
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 13:45:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB8835E-2AA2E3@206.165.44.70> References: <5mu6t8$hbs$1@ironhorse.plsys.co.uk> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit mmalcolm crawford <malcolm@plsys.co.uk> said: Lawson English <english@primenet.com> said: > >I understand that Jobs suggested that Adobe's lack of commitment to > >Rhapsody was a "3rd party opportunity." If GX *facilitates* the creation > of > >such apps (and it does) > > > Maybe it doesn't, or isn't necessary on OpenStep, cf > > Create... PasteUp... TIFFany... > Have those been reviewed in a Mac-oriented magazine or mainstream DTP magazine? How were they received by the reviewer? -------------------------------------------------------------------------- Any sufficiently advanced magic will be indistinguishable from technology -my corollary to Clarke's Law (AFAIK Mercedes Lackey got it from me at a World Fantasy Convention) --------------------------------------------------------------------------
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 21:06:51 GMT Organization: Digital Fix Development Message-ID: <5mvclb$jb2$1@news.digifix.com> References: <5mtgum$s25$1@news.digifix.com> <AFB7CB89-DFCD@206.165.44.52> In-Reply-To: <AFB7CB89-DFCD@206.165.44.52> On 06/01/97, "Lawson English" wrote: >Scott Anguish <sanguish@digifix.com> said: > >> >> FreeHand.. Illustrator... PageMaker... PhotoShop... >> >> Canvas.... PowerCadd... >> >> All best of breed apps, the first four are MUST HAVE apps for >> DTP... none require GX. The list of high-quality visual apps that >> DON'T use GX is long and distinguished... >> >> That doesn't mean the others aren't good apps, but it does >> mean that the Mac can succeed without GX... > >I understand that Jobs suggested that Adobe's lack of commitment >to Rhapsody was a "3rd party opportunity." Yep, thats what it said. > If GX *facilitates* the >creation of such apps (and it does), wouldn't it make sense for >Apple to put a LEETLE more work into a retained mode engine for >Rhapsody (not to mention MacOS)? No. Because apps that offer that functionality can easily be available for Unified without it. GX is not required to do any of this stuff. -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 21:09:28 GMT Organization: Digital Fix Development Message-ID: <5mvcq8$jf9$1@news.digifix.com> References: <5mth9e$s2b$1@news.digifix.com> <5muour$9kg@shelob.afs.com> In-Reply-To: <5muour$9kg@shelob.afs.com> On 06/02/97, Gregory H. Anderson wrote: >Scott Anguish writes >> Note that I've adjusted the followups to remove the next..programmer >> group. Its safe to assume that they don't want to hear more of >> Lawson's GX rant.... > >Sorry Scott, I restored it because I'm about to make a programming point. > Fair enough... >The One Whose Name I Cannot Type wrote: >> >Then there's the issue of constantly changing faces, fonts, >> >languages, sytles, etc. In the most pathological case, that info >> >could change with every glyph, making the communications latency >> >a major issue instead of a minor one. > >and Scott replied: >> No it doesn't.... To position the glyphs does NOT >> necessarily require round trips to the windowserver... > >You can take out the "necessarily", because it just DOESN'T. Font >info is cached locally in the AppKit, so determining character >bounding boxes, glyph advancement, and kerning adjustments does >NOT ping the windowserver during the course of text layout. Of course *I* knew this. There is nothing to prevent some idiot trying to malign DPS to say that it does though... -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 14:35:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB88EFE-2D5E03@206.165.44.70> References: <5muour$9kg@shelob.afs.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.advocacy MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Gregory H. Anderson <Greg_Anderson@afs.com> said: [not necessarily requiring round-trips to DPS] > You can take out the "necessarily", because it just DOESN'T. Font info > is cached locally in the AppKit, so determining character bounding boxes, > glyph advancement, and kerning adjustments does NOT ping the > windowserver > during the course of text layout. To draw, you moveto a starting location, > then just send a succession of xyshow operators, with setcolor or setfont > operations mixed in when some characteristic of the run changes. None of > these operations require a round trip. > Never said that they did. However, they require one trip per setting. GX's retained mode design requires one trip per LINE OF TEXT (although a font renderer is likely called more frequently -one or more times per glyph). > I mention xyshow because it raises an interesting point that has been > overlooked by the Agent Provocateur. PostScript is entirely neutral on > the topic of text direction. EVERY text handling operation accepts and/or > returns x AND y offsets, which may be position or negative. As far as PS > is concerned, text may run in ANY direction, even diagonally! Great, but each direction-setting requires sending that offset through the DPS channel. From the > standpoint of font metrics, when you ask how far the current point moves > for a given glyph, Western fonts return positive x's and 0 y's. Hebrew > fonts return negative x's. Chinese fonts return 0 x's, but positive y's. > And? GX retains that info within the shape. There's a single API call to handle ALL the kerning and so on. The communications between the font renderer and the internal GX cache can be made as optimal as you like because all of the info is available on a per-string basis. > What's important is whether a particular text engine is prepared to adjust > for these factors, and can translate other paragraph layout attributes -- > like tracking and inter-line spacing -- for reasonable usage in alternate > geometries. But now we're into the domain of "did the author care about > non-Western cases". It has NOTHING to do with PostScript itself. Excuse? What about English fonts like GX Hoefler Italic (GX fonts are going to be useable in Rhapsody, remember?) where one of the alternative feature sets for capital letters has a cursive "Q" with a tail that goes all the way to the end of the next letter or where the "X" in "GX" might wrap around an touch or almost touch the "!" to the right, FROM the right? How does one handle the situation where you do NOT want overlapping glyphs and are willing to sacrifice ornamentation without the need for custom kerning? GX allows one to set a flag to prevent ornamental glyphs from touching each other by reverting to a less ornmented version of the glyph, when available. How would you handle this using PS? This is a standard-issue Roman-script MacOS font, BTW, that ships with every Macintosh, even if APple forgot to ship any demonstration of its appearance. Sounds like a bit of round trip communication needed, even for the simple case of deciding which alternative capital "Q" or "X" to use. And let's not forget about contextual glyphs that might need auto-tweeking for much the same reason. Are you going to keep a version of DPS hit-testing on the AppKit side, now? ++++++++++++++++++++++++++++++ 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 +++++++++++++++++++++++++++++++++
Date: 2 Jun 1997 21:38:38 GMT From: clewis@ferret.ocunix.on.ca (Chris Lewis) Sender: steve edwards<custservice@tulsastage.com> Message-ID: <cancel.5mm1v2$l8o@news2.cais.com> Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5mm1v2$l8o@news2.cais.com> Control: cancel <5mm1v2$l8o@news2.cais.com> OKLAHOMA spam cancelled by clewis@ferret.ocunix.on.ca Original Subject: oklahoma special effects Total spams this type to date: 32 Total this spam type for this user: 32 Total this spam type for this user today: 3 Originating site: cais.com Complaint addresses: noc@cais.com abuse@cais.com postmaster@webzone.net
From: untulis@ng.netgate.net (Jason Untulis) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Followup-To: comp.sys.mac.advocacy,comp.sys.mac.programmer.misc Date: 3 Jun 1997 00:32:53 GMT Organization: Starfleet Headquarters Message-ID: <5mvonl$cnn@ss.netgate.net> References: <5mn8j9$ol3@shelob.afs.com> <AFB48643-BB4B9@206.165.44.84> Lawson English (english@primenet.com) wrote: : Lurkers can decide relative merit of posters based on relative value and : tone of content, I think. And by reputation as well... If you've never seen Lawson before, make sure to visit dejanews and the semper.fi archives for some of Lawson's "tone", since Lawson is the de facto reason why semper.fi became a moderated list... And Lawson, make sure you come back here in a year with your download/usage stats for your GX XFCN. I'm sure by then they'll be in the hundreds... -- Jason Untulis, Ravenous Media Consumer /\ / untulis@netgate.net <http://www.netgate.net/~untulis/> \ /==\ / untulis@netcom.com protected by spamgard[tm] \/v2.4\/ untulis@leland.stanford.edu PGP key soon (C) <http://www.netgate.net/~untulis/copyright.html>
From: Dennis Munsie <munsie@earthlink.net> Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody package Date: Mon, 02 Jun 1997 20:12:18 -0600 Organization: ChodeSoft Message-ID: <33937D7E.573E@earthlink.net> References: <5m2u4j$2b5$3@news1.ucsd.edu> <33853BCB.595B@ix.netcom.com> <weesh-2705970810450001@user-37kbv65.dialup.mindspring.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit > > Who qualifies for this offer? > > ALL levels of Macintosh Developer Program and Apple Media Program members > worldwide: > > - who have signed non-disclosure agreements > - who did not already receive the OpenStep tools as part of their WWDC > bundle > Okay, here's my question. What is so 'secret' about OpenStep 4 that we need a NDA to get a hold of it? I have no real intrest in getting a copy of an OS that runs only on Intel, but if I did, why wouldn't something like this be available to anyone who is on Apple's Developer Mailings? BTW, what exactly does the extra $100 a year get compared to the basic mailing at $150? All that I could find was that it would get you access to their testing labs, access to their developer question line (with a per question fee), and hardware discounts, something that really doesn't help the small time developers (like me) that bought their Mac first, and then decided to develop for it. Dennis
From: planetary <kris@xmission.xmission.com> Newsgroups: comp.lang.objective-c,comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Books? Date: 2 Jun 1997 16:24:33 -0600 Organization: XMission Internet (801 539 0900) Message-ID: <5mvh71$ifn@xmission.xmission.com> References: <AFB77F1A-9EC97B@141.214.134.235> <5mug7h$jvs$1@ironhorse.plsys.co.uk> In comp.sys.next.advocacy mmalcolm crawford <malcolm@plsys.co.uk> wrote: : On 06/02/97, "Robert A. Decker" wrote: : > : > Can anyone recommend some good Objective C and OpenStep programming : >books? : > : Others have already given places to look; I'd just like to put in a word for : NeXT's documentation, which is freely available (online at least, you could : also buy hard copy, cf www.next.com) and *very* good. Read through that : first before going to the bookshop -- the other references available are : probably not quite what you want as a "beginner". I've written some prototype "use case" doc for EOF programming. It is designed for beginners who are climbing the learning curve of EOF. It is task-based and provides links to conceptual material that intermediate programmers will wind useful. If there's a huge demand for it, I might make the prototype available, and even finish a chunk of it. Send me some email if you're interested. ...................kris -- Kristopher Magnusson kris@xmission.com (no NeXTmail, please) --------------------------------------------------------------------------- Contains freshness saver packet. DO NOT EAT.
Newsgroups: comp.sys.next.programmer From: frank@ali.bc.ca (Frank Pang) Subject: DO question: trapping client deaths Message-ID: <EB68yM.9I0@gateway.ali.bc.ca> Sender: nobody@gateway.ali.bc.ca Organization: A.L.I. Technologies, Inc. Date: Mon, 2 Jun 1997 23:23:09 GMT In NS3.3: I have a program using DO where a client can send messages for the server to do some fetching and processing. In the meantime, there's a possibility that the client dies before the server is able to send a reply to the client. In this scenario, the server will try to send a message to a dead client, causing [myNXConnection run] to exit. Is this an exception condition? It doesn't seem to raise exceptions. What's a clean way to handle client deaths / invalid connections? And where can I get some sample code for this? Thanks. -- Frank Pang, frank@ali.bc.ca Software Developer A.L.I. Technologies Ltd., (NeXT & MIME accepted) 95-10551 Shellbridge Way, Richmond, BC, 279-5422 (ext. 366) Canada V6X 2W9
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: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 21:47:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB8F437-B8A51@206.165.44.80> References: <5n04a1$re7$1@news.digifix.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 > >Integration with QuickTime. Etc. > > > In Rhapsody? QuickTime 3D is expected to be one of the first > Apple APIs on Rhapsody... How's that DPS interaction with QT proceeding? I can play QT movies into a GX offscreen viewport and manipulate each frame as just another bitmap shape, using any and all GX transparency, clipping and 3D effects, before sending it onscreen. ++++++++++++++++++++++++++++++ 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: beauvois@usa.net Newsgroups: comp.sys.next.programmer Subject: Strangeness... Date: Tue, 03 Jun 1997 02:31:11 -0500 Organization: University of Tennessee Message-ID: <beauvois-0306970231110001@amour.la.utk.edu> I'm having a real bad day with this one. It deals with this statement: if (dictionary2) [dictionary2 release]; The original code, up to this point: - (NSMutableDictionary *)createDictionaryForView:(id)aView hasSubviews:(BOOL)sub { NSMutableDictionary *dictionary2; NSString *aString2; NSMutableArray *newArray4; if (dictionary2) [dictionary2 release]; ... } The error from the debugger is as follows: -[testView createDictionaryForView:hasSubviews:] (self=0xfb608, _cmd=0x8c92, aView=0xf74e8, sub=1 '\001') at testView.m:279 Program generated(1): Memory access exception on address 0x1 (protection failure). 0x50069c8 in objc_msgSend () The testView class is a subclass of View, and "dictionary2" is a pointer to an as of yet unallocated NSMutableDictionary. The compiler gives me the following list of warnings, one of which deals with the above "dictionary2" : testView.m: In method `windowSubviews:' testView.m:48: warning: `swapArray' may be used uninitialized in this function testView.m: In method `viewSubviews:forDictionary:andArray:' testView.m:136: warning: `moreSubs' may be used uninitialized in this function testView.m:138: warning: `refer' may be used uninitialized in this function testView.m:139: warning: `subs2' may be used uninitialized in this function testView.m:140: warning: `newArray2' may be used uninitialized in this function testView.m:140: warning: `swapArray2' may be used uninitialized in this function testView.m:141: warning: `newDictionary' may be used uninitialized in this function testView.m:142: warning: `loopFlag' may be used uninitialized in this function testView.m: In method `createDictionaryForView:hasSubviews:' testView.m:272: warning: `dictionary2' may be used uninitialized in this function testView.m: In method `createMutableArrayCase:' testView.m:302: warning: `newArray2' may be used uninitialized in this function testView.m: In method `shadowView:' testView.m:368: warning: `ref' may be used uninitialized in this function testView.m:370: warning: `previousArray2' may be used uninitialized in this function testView.m:370: warning: `newArray3' may be used uninitialized in this function (btw, NONE of these are global; All are defined within actual methods. ) For the record, here are their types. Note that not all are objects (see loopFlag), yet the warning is the same for each: swapArray (NSMutableArray *) moreSubs (List *) refer (NSString *) subs2 (NSNumber *) newArray2 (NSMutableArray *) swapArray (NSMutableArray *) newDictionary (NSMutableDictionary *) loopFlag (BOOL) dictionary2 (NSMutableDictionary *) newArray2 (NSMutableArray *) ref (NSString *) previousArray2 (NSMutableArray *) newArray3 (NSMutableArray *) Also, the compiler seems to discriminate without symmetry: Some of the above warnings are called while several other variables are declared on the same line as the one being cited. There are also a great many variables (objects included) that it doesn't complain about, and the above list includes only those w/warnings. I don't know if I compiled using optimization (I'll check - I used whatever PB defaults to ), but even so - why wouldn't ALL uninitialized vars be flagged then ? I guess this boils down to two problems, both of which I am clueless about. Is there some connection between the two ? Has anyone generated this same error ? Please help ! Regards -- CB
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 3 Jun 1997 03:50:25 GMT Organization: Digital Fix Development Message-ID: <5n04a1$re7$1@news.digifix.com> References: <5mvclb$jb2$1@news.digifix.com> <AFB89A0D-2FF75B@206.165.44.70> In-Reply-To: <AFB89A0D-2FF75B@206.165.44.70> On 06/02/97, "Lawson English" wrote: >Scott Anguish <sanguish@digifix.com> said: >Lawson English <english@primenet.com> said: > >> > If GX *facilitates* the >> >creation of such apps (and it does), wouldn't it make sense for >> >Apple to put a LEETLE more work into a retained mode engine for >> >Rhapsody (not to mention MacOS)? >> >> >> No. Because apps that offer that functionality can easily be >> available for Unified without it. >> >> GX is not required to do any of this stuff. > >Never said that it was. Nor is DPS. Nor is anything more powerful than >SMallTalk's Pen class. > >The question is time-to-market for such apps. > Yes it is.... Including the OS... Attempting to integrate a GX or Taligent imaging model would delay the OS. I don't hold any of your delusions that GX makes time-to-market for Apps that much quicker. If it did, there would have been better adoption on the Mac OS to date. >GX provides functionality that DPS + AppKit doesn't and won't for quite >some time. > That is not something thats worthy of debate at this point. GX isn't well adopted on MacOS. That isn't changing. >E.G. full implementation of GX Typography. >Full implementation of GX >Printing. Which gives the user what in practical terms? >Full implementation of GX transparency. 3D perspective. Yeah... gotta have that 3D perspective in the imaging model. Geez Lawson... thats just foolish. >Integration with QuickTime. Etc. > In Rhapsody? QuickTime 3D is expected to be one of the first Apple APIs on Rhapsody... -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: Chuck_Esterbrook@orcacomputer.com Newsgroups: comp.sys.next.software,comp.sys.next.programmer Subject: Re: OS Enterprise for NT? Date: 2 Jun 1997 13:04:37 GMT Organization: Virginia Tech, Blacksburg, Virginia Message-ID: <5mugd5$2e0$1@solaris.cc.vt.edu> References: <5mf6uu$ghq$1@bvadm.bv.tek.com> <5mi0dn$t0i$1@falcon.inetnebr.com> <5mi0m5$kev$1@news.digifix.com> sanguish@digifix.com (Scott Anguish) wrote: > > > > Its important to note also that when Rhapsody ships the > Runtime costs drop to ZERO. > Why don't they drop the runtime costs to ZERO right now? I've got software to ship for 4.2 NT and that would certainly help. Also, how are they going to handle Display PostScript? I believe they currently pay Adobe a license fee. Just curious if anyone has some info about these things. -- Chuck Esterbrook, Software Eng. http://www.orcacomputer.com/~chuck
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 15:22:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB89A0D-2FF75B@206.165.44.70> References: <5mvclb$jb2$1@news.digifix.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Scott Anguish <sanguish@digifix.com> said: Lawson English <english@primenet.com> said: > > If GX *facilitates* the > >creation of such apps (and it does), wouldn't it make sense for > >Apple to put a LEETLE more work into a retained mode engine for > >Rhapsody (not to mention MacOS)? > > > No. Because apps that offer that functionality can easily be > available for Unified without it. > > GX is not required to do any of this stuff. Never said that it was. Nor is DPS. Nor is anything more powerful than SMallTalk's Pen class. The question is time-to-market for such apps. GX provides functionality that DPS + AppKit doesn't and won't for quite some time. E.G. full implementation of GX Typography. Full implementation of GX Printing. Full implementation of GX transparency. 3D perspective. Integration with QuickTime. Etc. ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 15:27:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFB89B44-304068@206.165.44.70> References: <5mvcq8$jf9$1@news.digifix.com> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.advocacy MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Those whose named won't be mentioned because that's what they called me, said: > > > >You can take out the "necessarily", because it just DOESN'T. Font > >info is cached locally in the AppKit, so determining character > >bounding boxes, glyph advancement, and kerning adjustments does > >NOT ping the windowserver during the course of text layout. > > Of course *I* knew this. There is nothing to prevent some > idiot trying to malign DPS to say that it does though... But as I pointed out, GX typography includes such complicated glyphs that hit-testing may be required to decide which version of a glyph to use. Idiots (whose names won't be mentioned) that don't bother to read the f-ing manual when debating topics that they know nothing about generally make even more mistakes than those (whose names won't be mentioned) that DO bother to read the manual, but merely misunderstood those big two- and three-syllable words... I mean, there are *illustrations* available in HTML format online that demonstrate this graphically. One need not be able to read -only observe that a version of a glyph might overlap with another in many common circumstances. ++++++++++++++++++++++++++++++ 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: dental@precipice.com (Rick Sanford) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Rhapsody webobjects problem Date: 2 Jun 1997 22:49:52 GMT Organization: Dental Records (R) Message-ID: <5mvimg$16s@news1-alterdial.uu.net> References: <AFB88DE5-DE6E60@141.214.134.235> Cc: comrade@umich.edu In <AFB88DE5-DE6E60@141.214.134.235> "Robert A. Decker" wrote: > I've been trying to get WebObjects (which I received at WWDC) running. > However, it appears that some libraries aren't installed. I've reinstalled > webobjects, openstep user, and openstep developer and the libraries still > aren't present. Here's the errors in console that I get: > > Software Version OPENSTEP 4.2 for Mach (Lantern5V) > probing for DOS > dyld: /NextDeveloper/Apps/WebObjectsBuilder.app/WebObjectsBulder can't open > the library: /NextLibrary/Frameworks/EOAccess.framework/Versions/B/EOAccess > (No such file or directory. errno = 2) > > I have no directory named 'EOAccess.framework'. Where should this have > been installed from? > > thanks, > rob > you'll likely need to install EOF in addition to the webobjects, openstep user, and openstep developer you have now. -- Rick Sanford Dental Records(R) dental@precipice.com NeXTMAIL welcome http://www.dentalrecords.com
From: mmalcolm crawford <malcolm@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 2 Jun 1997 22:52:18 GMT Organization: P&L Systems Message-ID: <5mvir2$kh$1@ironhorse.plsys.co.uk> References: <5mvclb$jb2$1@news.digifix.com> <AFB89A0D-2FF75B@206.165.44.70> In-Reply-To: <AFB89A0D-2FF75B@206.165.44.70> On 06/03/97, "Lawson English" wrote: >GX provides functionality that DPS + AppKit doesn't and won't for quite >some time. > Not on Rhapsody. >E.G. full implementation of GX Typography. Full implementation of GX >Printing. Full implementation of GX transparency. 3D perspective. >Integration with QuickTime. Etc. > With the exception of typography, which we are told is to be folded into Rahpsody anyway, I'm not convinced that any of these features is essential to a good word processor capable of multi-language support. YMMV. Best wishes, mmalc. --
From: dental@precipice.com (Rick Sanford) Newsgroups: comp.lang.objective-c,comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Books? Date: 2 Jun 1997 23:10:25 GMT Organization: Dental Records (R) Message-ID: <5mvjt1$16s@news1-alterdial.uu.net> References: <AFB77F1A-9EC97B@141.214.134.235> <5mtla3$go8$2@hackberry.zilker.net> Cc: leonvs@occam.com In <5mtla3$go8$2@hackberry.zilker.net> Leon von Stauber wrote: > In <AFB77F1A-9EC97B@141.214.134.235> "Robert A. Decker" wrote: > > > > I would like an Objective C programming language reference book. > > Good luck. The only dedicated Obj-C book I've used is one by Pinson & > Weiner (published by Addison/Wesley), but I wouldn't recommend it. > > ____________________________________________________________________ > Leon von Stauber > yeah, did anyone actualy build the Swamp Runner application? I've never seen it anywhere and I sure couldn't finish it... -- Rick Sanford Dental Records(R) dental@precipice.com NeXTMAIL welcome http://www.dentalrecords.com
From: John Kuszewski <johnk@spork.niddk.nih.gov> Newsgroups: comp.sys.next.programmer Subject: widgets for visual pipes? Date: Tue, 03 Jun 1997 04:45:13 -0400 Organization: Nat'l Insts of Health Message-ID: <3393D999.441B@spork.niddk.nih.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, I seem to remember an old NeXTSTEP program that let users build C-shell scripts visually. IIRC, it let you draw something that looked like +-------------+ +-------------+ | ls -l | | cat bar | +-------------+ +-------------+ | | | -------------+ | | +-------------+ | cat | +-------------+ | | | +-------------+ | grep foo | +-------------+ | | | +-------------------+ | nawk -f linecount | +-------------------+ My question is actually about the widgets used to create this sort of interface. Could anyone describe how painful this sort of thing is to build in NeXTSTEP/Rhapsody? I'd like to build a program that uses a similar "visual" approach to allow nontechnical users to build simple database queries. Thanks for any advice/observations y'all can offer me! -- _____________ | ___/_ | |/ / -- /\ // /-- || || / /|| || || / / || || ||/ / || John Kuszewski || |/ /| || johnk@spasm.niddk.nih.gov || / /|| || \/ / / || \/ that's MISTER protein G to you! |/__/| | /_________| My parents went to Zaire and all I got was this lousy retrovirus.
From: Bob Hathaway <bhathaway@sigs.com> Newsgroups: comp.sys.next.programmer,comp.lang.oberon,comp.object.logic,comp.lang.scheme,comp.lang.beta Subject: Object Magazine Online - New June Issue - FREE NEW ONLINE WEB JOURNAL Followup-To: comp.object Date: 2 Jun 1997 21:31:23 -0500 Organization: Object Magazine Online Message-ID: <5mvvlr$ngk@shoga.wwa.com> Summary: Object Magazine Online - FREE NEW WEB JOURNAL Keywords: Free WWW OO Object-Oriented Journal OBJECT MAGAZINE ONLINE ====================== OBJECT MAGAZINE ONLINE HYPERTEXT JOURNAL FREE NEW MONTHLY OBJECT-ORIENTED FORUM FORMERLY: OBJECT CURRENTS Location: http://www.sigs.com/omo Formerly: http://www.sigs.com/objectcurrents (Archive Site) Editor-In-Chief: Bob Hathaway <bhathaway@sigs.com> Issues: January 1996 (OCJ) thru May 1997 New Issue: June 1 (Now Available) Next Issue: July 1 Publisher: SIGS: Web Apps (NEW), Object Magazine, C++ Report, JOOP/ROAD, Object Expert, Smalltalk Report, X Journal, Java Report, Object Buyer's Guide, ... CONFERENCES: Object Expo/Java Expo/Web Apps Solutions, C++ World, OOP, Smalltalk Solutions, ... This is an invitation to join us at Object Magazine Online and view, engage in, and participate in the latest in object-oriented technology using the newest in information technology, the WWW. Object Magazine Online is a complete new free monthly journal with original Feature Articles, Columns, and Departments along with several *new* articles from Object Magazine and the various SIGS journals. OBJECT MAGAZINE ONLINE ARTICLES We are accepting original Articles to present in OMO which include honorarium and the opportunity to publish. Our World Class Columnists include: Watts Humphrey: SEI Process Director, CMM & PSP Inventor Bertrand Meyer: Eiffel, OO Design and Software Engineering Francois Bancilhon: President, O2 Technology, Leading ODBMS Expert Michael Jesse Chonoles: Chief of Methodology, Advanced Concepts Center of Lockheed Martin David Shang: OO Programming Language Designer, Motorola Labs Michael Spertus: President, Geodesic Systems, Program Automation Prof. Brain Henderson-Sellers: Director, Centre for Object Technology Applications and Research (Victoria) Ian Mitchell: Vigor Technology. Heads Rapid Prototyping Laboratory: http://osiris.sund.ac.uk/research/canopus/mitchell/rpl.html Interviews (including OCJ): January: Grady Booch February: James Rumbaugh March: Ivar Jacobson (Part I) - Get the latest on the UML June: Steve Mellor, Plus Jacobson (Part II) Soon: Sally Shlaer Newsgroup Dialog: - Monthly "Best Thread" from comp.object Robert Martin, Tim Ottinger Week in OT: Jane Grau - Late breaking news on object technology 4 times/month Feature Articles: Too many to repeat here, OCJ has presented many original features on object technology and OMO continues to present many more. Best new articles every month from SIGS Journals including Web Apps, Object Magazine, C++ Report, JOOP/ROAD, Object Expert, Smalltalk Report, X Journal, Java Report, Object Buyer's Guide, and etc. Thanks to our readership for patronage, praise, and feedback. Please keep visiting or give us a try soon. Please also feel free to inform friends and colleagues of this free new medium. From the OCJ Guidelines: Object Currents' unique hypertext media provides for advances over earlier journals - links to home pages, sites, databases and information servers, interaction, animation, graphics, code retrieval and execution, expanded pages, video, virtual reality and chat sessions. While all of these may not have appeared in these first issues, they will appear in the future. Check it out! Best Regards, Bob Hathaway Robert John Hathaway III Editor in Chief Object Magazine Online Email: bhathaway@sigs.com - Correspondence, Submissions
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody package Date: 3 Jun 1997 04:04:55 GMT Organization: Digital Fix Development Message-ID: <5n0557$rmj$1@news.digifix.com> References: <5m2u4j$2b5$3@news1.ucsd.edu> <33853BCB.595B@ix.netcom.com> <weesh-2705970810450001@user-37kbv65.dialup.mindspring.com> <33937D7E.573E@earthlink.net> In-Reply-To: <33937D7E.573E@earthlink.net> On 06/02/97, Dennis Munsie wrote: >> >> Who qualifies for this offer? >> >> ALL levels of Macintosh Developer Program and Apple Media Program members >> worldwide: >> >> - who have signed non-disclosure agreements >> - who did not already receive the OpenStep tools as part of their WWDC >> bundle >> >Okay, here's my question. What is so 'secret' about OpenStep 4 >that we need a NDA to get a hold of it? I have no real intrest >in getting a copy of an OS that runs only on Intel, but if I did, >why wouldn't something like this be available to anyone who is on >Apple's Developer Mailings? > You have to have signed the beta seed agreement. My guess is that Apple doesn't have to pay the License fees on beta shipped versions. >BTW, what exactly does the extra $100 a year get compared to the >basic mailing at $150? All that I could find was that it would >get you access to their testing labs, access to their developer >question line (with a per question fee), and hardware discounts, >something that really doesn't help the small time developers (like >me) that bought their Mac first, and then decided to develop for >it. > From what I've read, the $250 is the lowest support level at this point. But it would get you the DR release which the $150 apparently doesn't.. -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: boehring@biomed.ruhr-uni-bochum.de (Daniel Boehringer) Newsgroups: comp.sys.next.programmer Subject: Re: services call from the commandline? Date: 3 Jun 1997 10:12:57 GMT Organization: Ruhr-Universitaet Bochum, Rechenzentrum Message-ID: <5n0qn9$jf$1@sun579.rz.ruhr-uni-bochum.de> References: <5mnkfe$63r@tallowcross.uni-frankfurt.de> just write the commandline program yourself: sort of pseudocode: @implementation MiscString(PasteboardExtension) - toPasteboard:(Pasteboard *) pboard { NXAtom pasteTypes[] = {NXAsciiPboardType}; [pboard declareTypes:pasteTypes num:1 owner:NXApp]; [pboard writeType: pasteTypes[0] data:buffer length:[self length]+1]; return self; } @end void main() { MiscString *str=[[MiscString alloc] initString:argv[2]]; Pasteboard *pb=[Pasteboard new]; [str toPasteboard:pb]; NXPerformService(argv[1],pb); }
From: hhoff@ragnarok.en.eunet.de.NOSPAM (Holger Hoffstaette) Newsgroups: comp.sys.next.programmer Subject: Re: Tab to buttons? Date: 27 May 1997 23:36:18 GMT Organization: Far Out Labs, Inc. Message-ID: <5mfr5i$a9g@ragnarok.en.eunet.de> References: <5md1po$6su@chile.earthlink.net> Cc: rjnorman@earthlink.net (posted & mailed) "Robert Norman" wrote: > I'm a WWDC convert trying to get in step with OpenStep. I'm working through > the tutorials and making some progress however -- when I use IB to work on a > view, I'm having trouble controlling the tabbing. For example, in Currency > Converter with two selectable NSTextFields and a button, after I make the > nextKeyView connections between the two NSTextFields, which I confirm using > the inspector the tabbing order goes from the textFields through the button > and back to the first TextField. This tabbing order problem has cropped up > in the other tutorials as well including out of order elements and -- again > -- buttons! This is common first-time pitfall. If the window's initialFirstResponder outlet is not set, IB tries to outsmart you and ignores the nextKeyView chain, using the built-in left-to-right/top-down tabbing. Holger -- hhoff@ragnarok.en.eunet.de.NOSPAM LOAD "MACH_KERNEL",8,1
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer Subject: Beginner question... Moveable text box? Date: 3 Jun 97 10:54:00 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFB9A84B-1A2FC@141.214.134.235> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit I'm just starting out and so this may be a basic question. I've done the tutorials. I want to have a window that can have scrollable text boxes dropped onto it. However, I want the scrollable text boxes to be selectable and moveable within the window. I was wondering if anyone has extended NSScrollView to give these features (preferably with a title bar, resize handle, and button to iconize it). Is there something else I should be looking at? 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 sub=1 '\001') at testView.m:279 > Program generated(1): Memory access exception on address 0x1 (protection > failure). > 0x50069c8 in objc_msgSend () > -- Ralph Zazula Running Start, Inc. zazula@running-start.com 520/760-4890 (4891 FAX) http://www.running-start.com
Newsgroups: comp.sys.next.programmer Subject: Re: DO question: trapping client deaths Message-ID: <339443DA.7E5@running-start.com> From: Ralph Zazula <zazula@running-start.com> Date: Tue, 03 Jun 1997 09:18:34 -0700 References: <EB68yM.9I0@gateway.ali.bc.ca> Organization: Running Start, Inc. MIME-Version: 1.0 To: frank@ali.bc.ca Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi - You need to catch these conditions with an exception handler. Something like this: - updateClients { // foreach client... NX_DURING // try to update a client [aClient update:...]; NX_HANDLER // ignore failures and catch in the // invalidation notifier NX_ENDHANDLER } - senderIsInvalid:sender { List *remoteList; int i,c; // get the list of objects from the invalid connection remoteList = [sender remoteObjects]; for(i=0, c=[remoteList count]; i<c; i++) { // remove each client from our array [self removeClientObject:[remoteList objectAt:i]]; } return self; } Ralph Frank Pang wrote: > > In NS3.3: > I have a program using DO where a client can send messages for the server > to do some fetching and processing. In the meantime, there's a > possibility that the client dies before the server is able to send a reply > to the client. In this scenario, the server will try to send a message to > a dead client, causing [myNXConnection run] to exit. Is this an exception > condition? It doesn't seem to raise exceptions. What's a clean way to > handle client deaths / invalid connections? And where can I get some > sample code for this? Thanks. > -- Ralph Zazula Running Start, Inc. zazula@running-start.com 520/760-4890 (4891 FAX) http://www.running-start.com
From: "Suhail M. Ahmed" <suhails@ibm.net> Newsgroups: comp.lang.objective-c,comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Books? Date: Tue, 03 Jun 1997 14:50:02 +0200 Organization: EUnet Belgium, Leuven, Belgium Message-ID: <339412FA.EC77BE69@ibm.net> References: <AFB77F1A-9EC97B@141.214.134.235> <5mug7h$jvs$1@ironhorse.plsys.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Hi, Check out www.cbooks.com, they carry 13 books on NeXTSTEP. Cheers! Suhail
From: gary-nospam-@screaming.org (Gary W. Longsine) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Rhapsody webobjects problem Date: 3 Jun 1997 03:28:01 GMT Organization: Save the Skeet Foundation Message-ID: <5n0301$ba4$1@news.platinum.com> References: <AFB88DE5-DE6E60@141.214.134.235> Cc: comrade@umich.edu In <AFB88DE5-DE6E60@141.214.134.235> it appeared that "Robert A. Decker" wrote: > I've been trying to get WebObjects (which I received at WWDC) running. > However, it appears that some libraries aren't installed. I've reinstalled > webobjects, openstep user, and openstep developer and the libraries still > aren't present. Here's the errors in console that I get: > > Software Version OPENSTEP 4.2 for Mach (Lantern5V) > probing for DOS > dyld: /NextDeveloper/Apps/WebObjectsBuilder.app/WebObjectsBulder can't open > the library: /NextLibrary/Frameworks/EOAccess.framework/Versions/B/EOAccess > (No such file or directory. errno = 2) I have not seen this problem, but I also have not seen 4.2 PR2 (which I think is the version from WWDC). I do know that for the developer tools, the installation order is somewhat important (doh!) on some of the developer software. If you install the developer packages out-of-order, missing files might be one result (incompatible files might be another). In general, for the developer tools, the install order is: tools libs docs (docs are not really so order dependant, but that's the way it's listed on the little install guide that comes with the CD) I think the way the EOF stuff is packaged is being changed, too, so I'm not sure where you would expect these specific libraries to come from in 4.2... in the older 4.1 release, the EO stuff was part of a separate product, "Enterprise Object Framework" which came on a separate CD. The EOF CD had user & developer components. You may or may not need both for Web Objects development (i'd guess you need both). If you have an EOF CD-ROM in your kit, install that software and your problem will be fixed. If you don't, then maybe someone else with this kit will know if the EOF bits are now (4.2 release) located on the user/devel cds, or if perhaps your kit was missing the EOF cd. /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: fox@jeans.fokus.gmd.de (Oliver Fox) Newsgroups: comp.sys.next.programmer Subject: EOModeler to Database Date: 3 Jun 1997 13:31:44 GMT Organization: GMD-FOKUS Message-ID: <5n16c0$6ru@stern.fokus.gmd.de> I have EOF and a sybase database-server running on my black next so far - so good now I want to access to the sybase database-server from another next-machine in the network (for example: make a new EOModel from an existing database with EOModeler) HOW to do this ? Is there a way to type in the path in the database-login-panel off the EOModeler ? or another easy way to do that? thanks oliver
Newsgroups: comp.sys.next.programmer From: gabriel@trigger.ali.bc.ca (Gabriel Musatescu) Subject: Re: Strangeness... Message-ID: <EB7KLq.KJL@gateway.ali.bc.ca> Sender: nobody@gateway.ali.bc.ca Organization: A.L.I. Technologies, Inc. Date: Tue, 3 Jun 1997 16:32:13 GMT References: <beauvois-0306970231110001@amour.la.utk.edu> In article <beauvois-0306970231110001@amour.la.utk.edu> beauvois@usa.net writes: > > The original code, up to this point: > > - (NSMutableDictionary *)createDictionaryForView:(id)aView > hasSubviews:(BOOL)sub > { > NSMutableDictionary *dictionary2; > NSString *aString2; > NSMutableArray *newArray4; > > if (dictionary2) [dictionary2 release]; > You got what you deserve. Don't expect an unallocated object to be nil. dictionary2 holds an invalid object address at this point.
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Rhapsody webobjects problem Date: 3 Jun 97 15:26:45 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFB9E83A-805E@141.214.134.235> References: <5n0301$ba4$1@news.platinum.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit nntp://news.itd.umich.edu/comp.sys.next.advocacy I hope I'm not breaking an NDA or anything, but we don't get support for our Prelude stuff so I wanted to post what the problem was so others don't go through it. It looks like they forgot to include the EOF stuff for Mach on Intel. They included it for solaris and NT, but I really don't have a desire to use NT. They're deciding what they'll do to fix this. rob -- <mailto: "Robert A. Decker" comrade@umich.edu> Listen to my Realaudio playlist:<http://hmrl.cancer.med.umich.edu/Rob/index.ssi> Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #10: "Wow. I'm a genius too. I think. BEEP." -Chris Elliott
From: gutier@unixg.ubc.ca (Gerald Gutierrez) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Followup-To: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Date: 1 Jun 1997 05:21:58 GMT Organization: University of British Columbia, Vancouver, B.C., Canada Message-ID: <5mr0tm$opq$3@nntp.ucs.ubc.ca> References: <338b6960.76768104@news.pdnt.com> <AFB060E7-7C68@207.173.163.78> Let's see ... for someone to be able to have a look at Rapsody for "free", and thus potentially gain Apple many more Rhapsody supporters who may previously not have connections with Apple or even NeXT, that someone has to be already registered with Apple's Developer Program. Either : 1) being a member of Apple's Developer Program is free ( which I don't believe so ), or 2) the word "free" is being used incredibly loosely, or 3) this is one big huge lie. : Are you a developer registered with Apple's Developer Program? Have you : signed the Seed Agreement? If so: : >Additional information on the WWDC Prelude to Rhapsody package may be : >found at : ><http://product.info.apple.com/pr/press.releases/1997/q3/970429.pr.rel.wwdc : >..html>. To request your free OpenStep tools (Part Number R0724ZA) today, : >contact us in one of the following ways: : > : >Phone : >1-800-282-2732 Toll Free (US) : >1-800-637-0029 Toll Free (Canada) : >1-716-871-6555 International : > : >Email : >order.adc@apple.com : > : >FAX : >716-871-6511 : > : So there... it is free! : Mike -- `'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`' Gerald Gutierrez Computer Engineering Faculty of Applied Science gutier@unixg.ubc.ca University of British Columbia
From: michael@ninebits.com (Michael Balle) Newsgroups: comp.sys.next.programmer Subject: Tabview Date: Tue, 3 Jun 1997 22:20:50 +0200 Organization: Nine Bits Message-ID: <199706032220502927214@[222.223.224.4]> Hi, I have been playing a little bit with the Prelude to Rhapsody on Windows NT. In the interface builder they use a tab view to move between Object instances, Classes etc. Is this tab view available as a palette that can be used from IB in you own interface? regards Michael
From: sboker@calliope.psych.nd.edu (Steven M. Boker) Newsgroups: comp.sys.next.programmer Subject: Serial port sgtty line control? Date: 3 Jun 1997 20:33:51 GMT Organization: University of Notre Dame Message-ID: <5n1v3f$kuj@news.nd.edu> I've a problem in serial communications. I'm running OpenStep 4.0 on a Pentium 166 with an Intel ATX style motherboard. I'm using the Serial Port 4.00 and TTY Port Server 4.00 drivers. I'm trying to communicate with a slightly idiosyncratic serial device -- a Flock of Birds magnetic coil position tracker. I've gotten this device to respond by booting to DOS and using RTS/CTS hardware flow control. I can also talk to the device using OpenStep and XON/XOFF handshaking using the ttyda device. But I'm having trouble getting it to use RTS/CTS hardware flow control in OpenStep with ttydfa. The trouble is that if RTS is set while DTR is clear, the external device holds itself in a reset/standby mode. OS holds DTR clear as a default. As soon as I open the device, RTS is set and DTR is clear and the external device goes into standby. I can set DTR with ioctl(comhandle,TIOCSDTR,NULL), but the device stays in standby waiting for RTS to be cleared. Anyone know how to clear RTS? The following _almost_ works. I'm viewing the results using a breakout box to make sure I know what the signals are actually doing. This one has been driving me nuts! #define OPENPARAMS O_RDWR char * sys_com_port[2] ={"/dev/ttyfa","/dev/ttyfb"}; comport = 1; comhandle = open(sys_com_port[comport], OPENPARAMS); ioctl(comhandle,TIOCSDTR,NULL); Thanks for any helping hands! Steve -- Dr. Steven M. Boker 219-631-4941 (voice) sboker@nd.edu 219-631-8883 (fax) http://www.nd.edu/~sboker/ 219-257-2956 (home) Dept. of Psychology, University of Notre Dame, Notre Dame, IN 46556
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer Subject: Re: Tabview Date: 3 Jun 1997 20:31:09 GMT Organization: Digital Fix Development Message-ID: <5n1uud$idl$1@news.digifix.com> References: <199706032220502927214@[222.223.224.4]> In-Reply-To: <199706032220502927214@[222.223.224.4]> On 06/03/97, Michael Balle wrote: >Hi, > >I have been playing a little bit with the Prelude to Rhapsody on Windows >NT. In the interface builder they use a tab view to move between Object >instances, Classes etc. Is this tab view available as a palette that can >be used from IB in you own interface? > Currently... no. It will be in DR and Rhapsody, along with the Outline view that is used in IB (or something like it) -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.software,comp.sys.next.programmer Subject: Re: OS Enterprise for NT? Date: 3 Jun 1997 20:29:32 GMT Organization: Digital Fix Development Message-ID: <5n1urc$ib7$1@news.digifix.com> References: <5mf6uu$ghq$1@bvadm.bv.tek.com> <5mi0dn$t0i$1@falcon.inetnebr.com> <5mi0m5$kev$1@news.digifix.com> <5mugd5$2e0$1@solaris.cc.vt.edu> In-Reply-To: <5mugd5$2e0$1@solaris.cc.vt.edu> On 06/02/97, Chuck_Esterbrook@orcacomputer.com wrote: >sanguish@digifix.com (Scott Anguish) wrote: >> > >> >> Its important to note also that when Rhapsody ships the >> Runtime costs drop to ZERO. >> > >Why don't they drop the runtime costs to ZERO right now? I've got >software to ship for 4.2 NT and that would certainly help. > Presumably because they can't rework those license agreements. >Also, how are they going to handle Display PostScript? I believe >they currently pay Adobe a license fee. > >Just curious if anyone has some info about these things. > I doubt we'll ever know the details of the deal Adobe made with Apple... -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: don@globalobjects.com (Don Yacktman) Newsgroups: comp.sys.next.programmer Subject: Re: Tabview Date: 3 Jun 1997 21:21:57 GMT Organization: Global Objects Inc. Message-ID: <5n21tl$645$1@news.xmission.com> References: <199706032220502927214@[222.223.224.4]> <5n1uud$idl$1@news.digifix.com> sanguish@digifix.com (Scott Anguish) wrote: > On 06/03/97, Michael Balle wrote: > > I have been playing a little bit with the Prelude to Rhapsody > > on Windows NT. In the interface builder they use a tab view to > > move between Object instances, Classes etc. Is this tab view > > available as a palette that can be used from IB in you own > > interface? > > Currently... no. > > It will be in DR and Rhapsody, along with the Outline view > that is used in IB (or something like it) Well, the OPENSTEP MiscKit has some early cuts at it in the Temp area. I plan to have this object moved into the kit proper for the next release. The gotcha right now is that I have three(!) different versions of it, and I need to try and merge them into a superset without making the API ugly... There is a version in the NEXTSTEP MiscKit which works fine--for NEXTSTEP apps, anyway. It also has the outline view (more or less), which you can connect to a MiscTree and bingo... -- Later, -Don Yacktman don@misckit.com <a href="http://www.misckit.com/don.html">My home page</a>
From: mann@nautilus.weblab.rest.tasc.com (John Mann) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Rhapsody webobjects problem Date: 3 Jun 1997 22:17:14 GMT Organization: TASC Message-ID: <5n255a$qql@lois.Read.TASC.COM> References: <5mvimg$16s@news1-alterdial.uu.net> I'm having the same problem, except I already own a developer version of OPENSTEP with EOF. EOF for Mach was not distributed with the WWDC Prelude to Rhapsody. It is included only in the OPENSTEP Enterprise CD. If you want to use things like the report Wizard in this version of WebObjects, you'll have to buy a copy of EOF (presumably version 2.0) from Apple, or convince them that it should have been part of the handout. WWDC Prelude to Rhapsody. It is included only in the OPENSTEP Enterprise CD. If you want to use things like the report Wizard in this version of WebObjects, you'll have to buy a copy of EOF (presumably version 2.0) from Apple, or convince them that it should have been part of the handout.
From: mann@nautilus.weblab.rest.tasc.com (John Mann) Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer Subject: Re: Rhapsody webobjects problem Date: 3 Jun 1997 22:16:54 GMT Organization: TASC Message-ID: <5n254m$qqi@lois.Read.TASC.COM> References: <5mvimg$16s@news1-alterdial.uu.net> I'm having the same problem, except I already own a developer version of OPENSTEP with EOF. EOF for Mach was not distributed with the WWDC Prelude to Rhapsody. It is included only in the OPENSTEP Enterprise CD. If you want to use things like the report Wizard in this version of WebObjects, you'll have to buy a copy of EOF (presumably version 2.0) from Apple, or convince them that it should have been part of the handout.
Newsgroups: comp.sys.next.programmer From: guenther@onevision.de (Guenther Fuerthaller) Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Message-ID: <EB7L0v.3rL@onevision.de> Sender: news@onevision.de Organization: OneVision Vertriebs-GmbH, Regensburg, Germany References: <AFB88EFE-2D5E03@206.165.44.70> Date: Tue, 3 Jun 1997 16:41:19 GMT In article <AFB88EFE-2D5E03@206.165.44.70> "Lawson English" <english@primenet.com> writes: > Never said that they did. However, they require one trip per setting. GX's > retained mode design requires one trip per LINE OF TEXT (although a font > renderer is likely called more frequently -one or more times per glyph). > > > I mention xyshow because it raises an interesting point that has been > > overlooked by the Agent Provocateur. PostScript is entirely neutral on > > the topic of text direction. EVERY text handling operation accepts and/or > > > returns x AND y offsets, which may be position or negative. As far as PS > > is concerned, text may run in ANY direction, even diagonally! > > Great, but each direction-setting requires sending that offset through the > DPS channel. You obviously have no idea how the DPS client library works. Let me try to give you some pointers. First of all you finally have to realize, what has already been mentioned by others, that the possible problems with DPS communication are general problems of inter process communication in a multi tasking environment. The bad case mentioned in the documentation is a typical bad case in IPC when two processes perform a very synchronizing communication (A sends to B and waits for response). However this case is rarely given when you simply draw with DPS. Second: As long as you are drawing/redrawing a view content you send ps code to something that can be thought as a pipe (you know what a pipe is?), thus you perform something like buffered writing to a file. Therefore multiple single operator calls may be sent together/en block to the window server. Third: Additionally the Mach OS supports a feature called memory mapping. This enables the pseudo pipe from above to avoid copying of data during transmission to the window server. For example on Openstep for Mach i am able to render images of 30 MB or more using the standard image operator and sending the image data with DPSWriteData. In contrast this doesn't work on Openstep for NT. > And? GX retains that info within the shape. There's a single API call to > handle ALL the kerning and so on. The communications between the font > renderer and the internal GX cache can be made as optimal as you like > because all of the info is available on a per-string basis. Are you aware of the possibilities of DPS? The list of possible optimizations is too long and your statements too narrow. > Excuse? What about English fonts like GX Hoefler Italic (GX fonts are going > to be useable in Rhapsody, remember?) where one of the alternative feature > sets for capital letters has a cursive "Q" with a tail that goes all the > way to the end of the next letter or where the "X" in "GX" might wrap > around an touch or almost touch the "!" to the right, FROM the right? > > How does one handle the situation where you do NOT want overlapping glyphs > and are willing to sacrifice ornamentation without the need for custom > kerning? > > GX allows one to set a flag to prevent ornamental glyphs from touching each > other by reverting to a less ornmented version of the glyph, when > available. > > How would you handle this using PS? I agree that there should be an instance in Openstep that handles such issues but i strictly disagree that this should be done by DPS. Note that DPS is completely stupid about such issues and this is good. Guenther
From: "Ed Deans" <eadeans@ibm.net> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Tue, 3 Jun 1997 19:03:19 -0700 Organization: Road Runner Message-ID: <5n2ic0$f53@proxye1.san.rr.com> References: <5mo5jb$t71$1@news.digifix.com> <AFB501C4-1CCD2@206.165.44.25> <5moesg$36b$1@news.digifix.com> <5mosb1$9v6@proxye1.san.rr.com> <5ms72n$49$1@bignews.shef.ac.uk> mmalcolm crawford wrote in article <5ms72n$49$1@bignews.shef.ac.uk>... >On 05/31/97, "Ed Deans" wrote: >> >>Lawson English <english@primenet.com> said: >> >> >> >>Talk about silly questions. Rhapsody doesn't give MacOS users >> >>ANYTHING except (maybe) an upgrade path sometime in the NeXT 18-24 >> >>months to obtain these oh-so-important buzzwords that were the >> >>reason why they bought a Mac in the first place. >> >> You're pretty much right from what we know about Rhapsody. Where's the >> innovation that will push Rhapsody into the fore of the market's mind so >> Apple will survive? >> >Umm, was this rhetorical sarcasm? In case it wasn't, how about: > >(Innovation as far as MacOS goes): That's the problem: "as far as MacOS goes"... >Preemptive multitasking; multi-threading; multi-user capability... > >Unified imaging model; OO development environment; unequalled cross-platform >compatibility... Unified imaging model? It remains to be seen how important this will be in the real world. OO development environment? Pick one, there are plenty of others or varied quality including Visual Studio, Visual Age and CommonPoint. Cross-platform unequalled? That's tough to say. Yellow Box looks like it's only Rhapsody and Win95/NT right? There are development systems such as the one from Star Division (http://www.stardivision.com for English; German pages have much more info) that support more platforms. In fact, Star Division's Office suite deploys on MacOS, NT, Win95, OS/2 and Linux. I know that it isn't exactly equal to OpenStep but the issue is cross-platform compatibility. >and so on. I'm still trying to understand how Rhapsody will draw user from MacOS & Windows NT/98 and developers from MacOS & NT/98 as well. Mac people see gains but have you seen what's been talked about regarding future MacOS versions? We're hearing Yellow Box (partial), perhaps Mach or NuKernel, preemptive threads, etc. Sounds like Copland to me. No ones going to run to Rhapsody if they know that in a year or so longer they can get a virtual identical user / develop experience on MacOS. What does Rhapsody offer to NT people? Apple not even defined what this OS will be yet--end user? workstation? server? so developers like Netscape aren't signing up in droves with their end-user apps. >> > Given developers faith that Apple can and will ship a new OS >> >before the end of the Century? >> >> Even on its deathbed, the worst case for Copland was a ship date before Jan >> 1, 2000. >> >Again it's not clear what you're implying here -- in case you're suggesting >Rhapsody won't ship until then, I'm prepared to lay a bet that it will ship >to the general public prior to end of 1998 (developers will get their hands >on it by end Q3 this year) -- and this is a conservative estimate. I didn't imply that. I was simply saying that Rhapsody isn't scheduled to be any later than the last estimate for Copland. I'm sure Rhapsody will ship (if Apple's still in business) but I've still see ZERO evidence that it will be anything more than a niche system. We're only a year out from the Unified version now. It's about six months or less from the Premier edition and things are still sketchy as to what this NeXT Generation "NT-killer" will offer that will make it live up to its name. --Ed. >Best wishes, > >mmalc. > >-- > >
Date: Tue, 03 Jun 1997 21:12:02 -0600 From: Tracla@aol.com Subject: NeXT People Question Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.programmer Message-ID: <865389689.23631@dejanews.com> Organization: Deja News Usenet Posting Service To: Tracla@aol.com Hello - my name is Tracey.. I am working with AT&T in HR as a partner. We are looking for NeXT System Administrators.. But my question is, since this is a new software product for me is: Do people outside of NeXT software have these skills? Are there any dedicated websites where we can post job opportunities What are the salaries of people with this skill set? Thank you for any assistance you can provide Tracey Claybrooke 813-685-6004 Tampa, Florida -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer Subject: Re: Tabview Date: 3 Jun 1997 22:22:04 GMT Organization: Digital Fix Development Message-ID: <5n25ec$km2$1@news.digifix.com> References: <199706032220502927214@[222.223.224.4]> <5n1uud$idl$1@news.digifix.com> <5n21tl$645$1@news.xmission.com> In-Reply-To: <5n21tl$645$1@news.xmission.com> On 06/03/97, Don Yacktman wrote: >sanguish@digifix.com (Scott Anguish) wrote: >> On 06/03/97, Michael Balle wrote: >> > I have been playing a little bit with the Prelude to Rhapsody >> > on Windows NT. In the interface builder they use a tab view to >> > move between Object instances, Classes etc. Is this tab view >> > available as a palette that can be used from IB in you own >> > interface? >> >> Currently... no. >> >> It will be in DR and Rhapsody, along with the Outline view >> that is used in IB (or something like it) > >Well, the OPENSTEP MiscKit has some early cuts at it in the Temp >area. I plan to have this object moved into the kit proper for >the next release. The gotcha right now is that I have three(!) >different versions of it, and I need to try and merge them into >a superset without making the API ugly... > It sure sould be nice if someone at Apple/Next could slip us the API that they are going to be using.... ... compatibility with that would be cool.. >There is a version in the NEXTSTEP MiscKit which works fine--for >NEXTSTEP apps, anyway. It also has the outline view (more or >less), which you can connect to a MiscTree and bingo... >> -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: don@globalobjects.com (Don Yacktman) Newsgroups: comp.sys.next.programmer Subject: Re: Tabview Date: 4 Jun 1997 05:41:53 GMT Organization: Global Objects Inc. Message-ID: <5n2v71$3pl$1@news.xmission.com> References: <199706032220502927214@[222.223.224.4]> <5n1uud$idl$1@news.digifix.com> <5n21tl$645$1@news.xmission.com> <5n25ec$km2$1@news.digifix.com> sanguish@digifix.com (Scott Anguish) wrote: > On 06/03/97, Don Yacktman wrote: > >Well, the OPENSTEP MiscKit has some early cuts at it in the Temp > >area. I plan to have this object moved into the kit proper for > >the next release. The gotcha right now is that I have three(!) > >different versions of it, and I need to try and merge them into > >a superset without making the API ugly... > > > > It sure sould be nice if someone at Apple/Next could slip us > the API that they are going to be using.... > > ... compatibility with that would be cool.. I agree. Failing that, I may attempt a class-dump to see what the NS 4.2 API is and go from there. Trouble is, I know they'll change it, so... Any idea who I should talk to to get the API? I suspect I could get it and model the MiscKit object off it if I can get in touch with the right person. I'm an EAP, so I'll ask my contact tomorrow, but you never know how effective that will be...they can be pretty busy guys... :-) -- Later, -Don Yacktman don@misckit.com <a href="http://www.misckit.com/don.html">My home page</a>
From: Friedrich HAEUPL <friedrich.haeupl@banyan.siemens.at> Newsgroups: comp.sys.next.programmer Subject: Serial port under Rhapsody/Open/NextStep Date: Wed, 04 Jun 1997 13:05:27 -0700 Organization: siemens Message-ID: <3395CA87.58FD@banyan.siemens.at> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Is there a certain API for the serial interfaces under OpenStep/NextStep and Rhapsody? Is there any sourcecode available, that shows how to access the serial ports. With the powerful tools provided it show be easy to write a programm that gets data via the serial ports.
From: Friedrich HAEUPL <friedrich.haeupl@banyan.siemens.at> Newsgroups: comp.sys.next.programmer Subject: Serial interface in OpenStep NextStep Rhapsody Date: Wed, 04 Jun 1997 13:04:48 -0700 Organization: siemens Message-ID: <3395CA60.F7C@banyan.siemens.at> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Is there a certain API for the serial interfaces under OpenStep/NextStep and Rhapsody? Is there any sourcecode available, that shows how to access the serial ports. With the powerful tools provided it show be easy to write a programm that gets data via the serial ports.
From: pretzl@pobox.com (Allan Peretz) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: Wed, 04 Jun 1997 11:30:47 GMT Organization: Planet Digital Network Technologies Message-ID: <3395514a.24565841@news.pdnt.com> References: <5m2u4j$2b5$3@news1.ucsd.edu> <338b6960.76768104@news.pdnt.com> <338B8919.6351@earthlink.net> <338B8EA4.1D57@earthlink.net> <AlvinKoh-2805971604500001@fsng172mac.ops.sing.paging.mot.com> <5mhvfv$k1i$1@news.digifix.com> <33908F5C.3B0A@wootech.com> <33917b98.71663816@news.pdnt.com> <3391EFFB.552E@research.canon.com.au> On Mon, 02 Jun 1997 07:56:11 +1000, cstory@research.canon.com.au wrote: >If you're not a developer, you don't need it. The point of the >Prelude is to introduce developers to the tools that will be >included in Rhapsody DR (Developer's Release). It's to encourage >software development for Rhapsody. If you're not a developer, you >don't really fit in. Apple isn't pushing NextStep. Excuse me!? How do you know what I need? I need some time to evaluate, test, and learn this product early so I can determine how it may benefit my clients. As I said in my earlier post, it is clearly in Apple's advantage to allow me this opportunity. Allan Peretz
From: neuss@informatik.th-darmstadt.de.nospam (Christian Neuss) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Rhapsody webobjects problem Followup-To: comp.sys.next.programmer,comp.sys.next.advocacy Date: 4 Jun 1997 11:15:11 GMT Organization: Technische Hochschule Darmstadt Message-ID: <5n3inv$er0$1@news.th-darmstadt.de> References: <AFB88DE5-DE6E60@141.214.134.235> <5n0301$ba4$1@news.platinum.com> Gary W. Longsine (gary-nospam-@screaming.org) wrote: > I have not seen this problem, but I also have not seen 4.2 PR2 (which I think > is the version from WWDC). Same here.. but the problem is definitely the missing EOF libs. > I think the way the EOF stuff is packaged is being changed, too, so I'm not > sure where you would expect these specific libraries to come from in 4.2... > in the older 4.1 release, the EO stuff was part of a separate product, > "Enterprise Object Framework" which came on a separate CD. The EOF CD had > user & developer components. You may or may not need both for Web Objects > development (i'd guess you need both). For general WebObjects programming, all you need is EOF user. Of course, for EOF development you need the complete kit. Regards, Chris -- // Christian Neuss "static typing? how quaint.." // http://www.nexttoyou.de/~neuss/ // fax: (+49) 6151 16 5472
From: neuss@informatik.th-darmstadt.de.nospam (Christian Neuss) Newsgroups: comp.sys.next.programmer Subject: Re: Serial port sgtty line control? Date: 4 Jun 1997 12:14:45 GMT Organization: Technische Hochschule Darmstadt Message-ID: <5n3m7l$er0$4@news.th-darmstadt.de> References: <5n1v3f$kuj@news.nd.edu> Steven M. Boker (sboker@calliope.psych.nd.edu) wrote: > The trouble is that if RTS is set while DTR is clear, the external > device holds itself in a reset/standby mode. OS holds DTR clear > as a default. As soon as I open the device, RTS is set and DTR > is clear and the external device goes into standby. I can set DTR > with ioctl(comhandle,TIOCSDTR,NULL), but the device stays in standby > waiting for RTS to be cleared. Anyone know how to clear RTS? Sorry that I can't help with the programming, I haven't programmed the bare metal under OPENSTEP yet :-) .. but couldn't you simply wire a cable that fakes RTS? After all, the hardware handshake is somewhat redundant. Rgds, Chris -- // Christian Neuss "static typing? how quaint.." // http://www.nexttoyou.de/~neuss/ // fax: (+49) 6151 16 5472
From: Alex Blakemore <alex@genoa.com> Newsgroups: comp.sys.next.programmer Subject: Re: Strangeness... Date: 4 Jun 1997 04:31:07 GMT Organization: Genoa Software Systems Message-ID: <5n2r2b$l6@saturn.genoa.com> References: <beauvois-0306970231110001@amour.la.utk.edu> <EB7KLq.KJL@gateway.ali.bc.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: gabriel@trigger.ali.bc.ca In <EB7KLq.KJL@gateway.ali.bc.ca> Gabriel Musatescu wrote: > Don't expect an unallocated object to be nil. true enough when speaking of _automatic_ variables, (those defined local to a method or block) but just to be pedantic, it is safe to assume that instance variables are initialized to nil - that is defined by the language. so code in the "init" method that sets ivars to zero or nil is redundant, unless of course you expect init to be sent more than once to the same object. -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
From: mcgredo@crl.crl.com (Donald R. McGregor) Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.programmer Subject: Re: NeXT People Question Date: 4 Jun 1997 08:37:43 -0700 Organization: Miskatonic University Department of Classics Message-ID: <5n4247$odp@crl.crl.com> References: <865389689.23631@dejanews.com> In article <865389689.23631@dejanews.com>, <Tracla@aol.com> wrote: >Hello - my name is Tracey.. I am working with AT&T in HR as a partner. >We are looking for NeXT System Administrators.. > >But my question is, since this is a new software product for >me is: > >Do people outside of NeXT software have these skills? Since NeXT is BSD Unix under the skin, anyone with a Unix background should be able to pick it up fairly quickly. There's some flashy NeXT sysadmin tools that make life easier, but anyone who's competent should be able to pick that up quickly enough. Rhapsody might have enough flashy sysadmin tools for a non- unix type to get by, but that remains to be seen. >What are the salaries of people with this skill set? Uh, $500,000/yr. Yeah, that's the ticket. :-) -- Don McGregor | I did it for the children. mcgredo@crl.com |
From: NoSpa.Mgustilo@mail.med.upenn.edu Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.codewarrior Subject: MacOS Rumors: PowerComputing offering Developer Disounts? Date: Wed, 04 Jun 1997 11:30:06 -0500 Organization: University of Pennsylvania Message-ID: <NoSpa.Mgustilo-0406971130060001@renal8.med.upenn.edu> third hand anecdote of PowerCC offering a PowerCenter Pro 210 for US$1800 to a developer (they did check for developer status) full text at this URL: <http://rumors.netexpress.net>; article is dated 06/04/97 ----------- Bicycle Crash Test Dummy for Hire gustilo@mail.med.upenn.edu
Date: Wed, 04 Jun 1997 11:52:27 -0600 From: Tracla@aol.com Subject: FL/NPB/ NeXT Systems Administrators (3) needed Newsgroups: comp.sys.next.marketplace,comp.sys.next.programmer Message-ID: <865439271.19617@dejanews.com> Organization: Deja News Usenet Posting Service AT&T is seeking (3) NeXT Systems Programmers in North Palm Beach, Florida. PURPOSE: Ensure the availability of NeXT hardware and sofware services including configuration, installation and tuning. QUALIFICATIONS: Bachelors degree or equivalent 6 months with NeXT specific administration 1 year with UNIX administration Excellent project management skills Please send resume and salary history to: Tracey Claybrooke Tracla@aol.com 813-685-6004 phone Thank you -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: riley@dellirium.uwlax.edu (David Riley) Newsgroups: comp.sys.next.programmer Subject: AWT library port status? Date: 4 Jun 1997 17:18:26 GMT Message-ID: <5n4812$sjk@wiscnews.wiscnet.net> I understand that someone is doing a native NEXTSTEP port of the Java AWT library. Can anyone tell me the current status, or put me in touch with someone who can? Please accept my apologies if I am asking something that is common knowledge to regular readers of this newgroup. /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ David D. Riley, Chair riley@cs.uwlax.edu Computer Science Department [NeXTMail or MIME okay] University of WI - La Crosse
From: mpaque@wco.com (Mike Paquette) Newsgroups: comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 4 Jun 1997 10:56:23 -0700 Organization: Electronics Service Unit No. 16 Sender: mpaque@mpaque Distribution: world Message-ID: <5n4a87$2bi@mpaque.mpaque> References: <AFB88EFE-2D5E03@206.165.44.70> In article <AFB88EFE-2D5E03@206.165.44.70> "Lawson English" <english@primenet.com> writes: [On text layout, and round trips to the DPS engine] > Never said that they did. However, they require one trip per setting. GX's > retained mode design requires one trip per LINE OF TEXT (although a font > renderer is likely called more frequently -one or more times per glyph). At some level, any text system will require the association of position information with a glyph, even if only a default position. That is, the output device needs to know where to put the ink or pixels. It happens that the decomposition used in OPENSTEP permits this information to be exposed in the public APIs. It is also worth noting that the design of the communication channel with the DPS engine under OPENSTEP is such that NO round trips are required. All the commands generated by the text system are encoded in a compact binary format as part of a buffered stream. The stream buffers are asynchronously flushed to the DPS engine when full, or when the application returns to the main event loop, or very infrequently by an explicit flush command. The text system does not have to wait for a reply from the DPS engine. There is no round trip overhead, and there is a substantial performance advantage on multiprocessor platforms. -- Mike Paquette (mpaque@wco.com) 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: embuck@palmer.cca.rockwell.com (Erik M. Buck) Newsgroups: comp.sys.next.programmer Subject: Re: Beginner question... Moveable text box? Date: 4 Jun 1997 17:53:33 GMT Organization: Rockwell Collins Message-ID: <5n4a2t$1sh5@castor.cca.rockwell.com> References: <AFB9A84B-1A2FC@141.214.134.235> Cc: comrade@umich.edu In <AFB9A84B-1A2FC@141.214.134.235> "Robert A. Decker" wrote: > > I'm just starting out and so this may be a basic question. I've done the > tutorials. > > I want to have a window that can have scrollable text boxes dropped onto > it. However, I want the scrollable text boxes to be selectable and moveable > within the window. > I was wondering if anyone has extended NSScrollView to give these > features (preferably with a title bar, resize handle, and button to iconize > it). Is there something else I should be looking at? > > thanks, > > rob > This is a VERY bad user interface idea. I do not know of any implementations like that. (Except in obsolete Windows versions!) What you are describing is multiple windows with scolling text for content. Why limit the "extended NSScrollView ... (preferably with a title bar, resize handle, and button to iconize it)" to the inside of another window ?
From: juergen.albertsen@flensburg.netsurf.de (Juergen Albertsen) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy Subject: Re: Rhapsody webobjects problem Date: Wed, 04 Jun 1997 16:49:41 GMT Organization: Private Message-ID: <339483cc.2367213@news.allcon.net> References: <AFB88DE5-DE6E60@141.214.134.235> Hi Robert! > I understand that we don't receive support for our Rhapsody cd's. Can you >forward this to the rhapsody evangelist if you can't help me? > > I've been trying to get WebObjects (which I received at WWDC) running. >However, it appears that some libraries aren't installed. I've reinstalled >webobjects, openstep user, and openstep developer and the libraries still >aren't present. Here's the errors in console that I get: For some very strange reasons that I don't understand NeXT (or Apple) has decided to ship WebObjects and the Enterprise Objects Framework (EOF) as different products for the Mach platform. Unfortuntaley, you need EOF to run WebObjects. I don't know how the giveaways at the WWDC were packed, but maybe there's some CD at bottom of your box? If not, it seems rather pointless to me to freely distribute a product without fullfilling all requirements to run it! Jürgen --- Jürgen Albertsen juergen.albertsen@flensburg.netsurf.de Face the facts -- forget euphoria!
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc,comp.sys.next.advocacy Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Wed, 4 Jun 1997 15:38:43 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <0nZQF3_00UhWQ23iFZ@andrew.cmu.edu> References: <5mo5jb$t71$1@news.digifix.com> <AFB501C4-1CCD2@206.165.44.25> <5moesg$36b$1@news.digifix.com> <5mosb1$9v6@proxye1.san.rr.com> <5ms72n$49$1@bignews.shef.ac.uk> <5n2ic0$f53@proxye1.san.rr.com> In-Reply-To: <5n2ic0$f53@proxye1.san.rr.com> Excerpts from netnews.comp.sys.next.advocacy: 3-Jun-97 Re: GX OOP class proposal &.. by "Ed Deans"@ibm.net >> Unified imaging model; OO development environment; unequalled >> cross-platform compatibility... > > Unified imaging model? It remains to be seen how important this will > be in the real world. Just because you refuse to look doesn't mean that you can't see what the effects are in the real world. Try taking a look at how well the lack of a unified imaging model hampers the development of X applications which want to print, and compare that to NEXTSTEP, where the unified imaging model makes that headache disappear. > OO development environment? Pick one, there are plenty of > others or varied quality including Visual Studio, Visual Age and > CommonPoint. Go read DejaNews for previous commentary on exactly how CommonPoint has done in terms of providing a OO development environment as compared to NEXTSTEP. Specificly, where are the best-of-breed apps and revoluationary improvements from other OO development environments? > Cross-platform unequalled? That's tough to say. Try using NEXTSTEP sometime. Just using one machine, I can cross-compile for 4 different architectures simply by clicking on checkboxes in ProjectBuilder. Name another commercial development environment that provides better cross-compilation support....? . > Yellow Box looks like it's only Rhapsody and Win95/NT right? We're talking about OpenStep. OpenStep is available for the m68k black hardware; Intel hardware running Mach, Win 95, and NT; SPARC hardware running Mach or Solaris; and NEXTSTEP used to be available for the HPPA architecture as well. Furthermore, parts of OpenStep like PDO/EOF are available for a broader range of Unix operating systems but including all of the above configurations as well. There's also the GNUStep project which will provide a portable and free implementation of OpenStep for Linux which presumably could easily be ported to any other Unix-like or POSIX-compliant environment. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer Subject: Re: Beginner question... Moveable text box? Date: 4 Jun 97 16:13:44 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFBB44C2-184CA@141.214.134.235> References: <9706041752.AA15173@palmer.cca.rockwell.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: "Erik M. Buck" <embuck@palmer.cca.rockwell.com> On Wed, Jun 4, 1997 1:52 PM, Erik M. Buck <mailto:embuck@palmer.cca.rockwell.com> wrote: >Why limit the "extended NSScrollView ... (preferably with a title bar, >resize handle, and button to iconize it)" to the inside of another window ? It's for a survey builder application. We want to have a text box that can be moved around on the screen and that can be iconized. The text box will be linked to other text boxes to give a visual representation of the path that may be taken through the survey. The surveys are dynamic, so multiple paths can be followed based how the questions are answered (therefore some boxes can have multiple links coming off). The box will contain the Question, which will either be text, or something representing a picture, or sound clip, or movie). For now I just want to be able to move a text box around, and to be able to iconize it. Right now I just want to the most basic thing. It will build a survey that can used with our Newton survey engine, a paper survey engine we have, and the web (webobjects would be great for this. Right now we're using Intelliweb from Micromass). 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
Newsgroups: comp.sys.next.programmer From: frank@ali.bc.ca (Frank Pang) Subject: Re: DO question: trapping client deaths Message-ID: <EB9q7u.6np@gateway.ali.bc.ca> Sender: nobody@gateway.ali.bc.ca Organization: A.L.I. Technologies, Inc. Date: Wed, 4 Jun 1997 20:28:41 GMT References: <EB68yM.9I0@gateway.ali.bc.ca> In article <EB68yM.9I0@gateway.ali.bc.ca> frank@ali.bc.ca (Frank Pang) writes: Yes I am replying to my own question. After reading the replies and the manuals I got my server to trap client deaths. HOWEVER, there is also another piece of info that helped make this work: From nextanswers: * In some circumstances (to be discussed in a future article), NXConnection objects can be created for your application implicitly by the DO system as objects are vended to processes other than the one that you originally connected to. If this happens, you won't be notified automatically of port deaths for the new NXConnection instances. Your application must become the delegate of the first NXConnection instance that you create and respond to the connection:didConnect: delegate message. The new NXConnection is passed to this routine, which gives you an opportunity to register for invalidation notification with that connection. -- Frank Pang, frank@ali.bc.ca Software Developer A.L.I. Technologies Ltd., (NeXT & MIME accepted) 95-10551 Shellbridge Way, Richmond, BC, 279-5422 (ext. 366) Canada V6X 2W9
From: Chuck_Esterbrook@orcacomputer.com Newsgroups: comp.sys.next.programmer Subject: How to insert an icon into an NSTextView? Date: 4 Jun 1997 21:47:31 GMT Organization: Virginia Tech, Blacksburg, Virginia Message-ID: <5n4npj$ppo$1@solaris.cc.vt.edu> Anyone know how to insert an icon into an NSTextView? I see there is a feature for attachments, but my icons aren't attachments: they're just content/decore. A direct reply to chuck@orcacomputer.com would be appreciated. Any info or pointers are appreciated. Thanks. -Chuck
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 4 Jun 1997 22:20:56 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5n4po8$8k0$1@brachio.zrz.TU-Berlin.DE> References: <0nZQF3_00UhWQ23iFZ@andrew.cmu.edu> In article <0nZQF3_00UhWQ23iFZ@andrew.cmu.edu> Charles William Swiger <cs4w+@andrew.cmu.edu> writes: > Excerpts from netnews.comp.sys.next.advocacy: 3-Jun-97 Re: GX OOP class > proposal &.. by "Ed Deans"@ibm.net > > Cross-platform unequalled? That's tough to say. > > Try using NEXTSTEP sometime. Just using one machine, I can > cross-compile for 4 different architectures simply by clicking on > checkboxes in ProjectBuilder. Name another commercial development > environment that provides better cross-compilation support....? Or name any other cross-platform environment where a 'port' has been delivered accidentally! At least that's the rumor on 4.x Developer/HP. Marcel
From: cstory@research.canon.com.au Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: Thu, 05 Jun 1997 09:44:44 +1000 Organization: Canon Information Systems Research Australia Message-ID: <3395FDEC.6FFA@research.canon.com.au> References: <5m2u4j$2b5$3@news1.ucsd.edu> <338b6960.76768104@news.pdnt.com> <338B8919.6351@earthlink.net> <338B8EA4.1D57@earthlink.net> <AlvinKoh-2805971604500001@fsng172mac.ops.sing.paging.mot.com> <5mhvfv$k1i$1@news.digifix.com> <33908F5C.3B0A@wootech.com> <33917b98.71663816@news.pdnt.com> <3391EFFB.552E@research.canon.com.au> <3395514a.24565841@news.pdnt.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Allan Peretz wrote: > > On Mon, 02 Jun 1997 07:56:11 +1000, cstory@research.canon.com.au > wrote: > > >If you're not a developer, you don't need it. The point of the > >Prelude is to introduce developers to the tools that will be > >included in Rhapsody DR (Developer's Release). It's to encourage > >software development for Rhapsody. If you're not a developer, you > >don't really fit in. Apple isn't pushing NextStep. > > Excuse me!? How do you know what I need? I need some time to > evaluate, test, and learn this product early so I can determine how it > may benefit my clients. As I said in my earlier post, it is clearly > in Apple's advantage to allow me this opportunity. > > Allan Peretz Your clients are developers? If so, ok, I agree, but then they're probably in the developer program, right? If they aren't developers, why do you need to evaluate development tools for them? Give me a break. If you were a developer, you would know better than to expect fair treatment from Apple anyway... Aloha, Cliff
From: "Robert Norman" <rob@neurodata.com> Newsgroups: comp.sys.next.programmer Subject: Driver for Parallel Port? Date: Wed, 04 Jun 1997 15:52:08 -0700 Organization: NEURODATA Message-ID: <5n4rm6$43@chile.earthlink.net> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Hi opensteppers! I'm prototyping a data acquisition program that will use custom drivers later but for now, I'd just like to talk to the outside world through the parallel port. Are there generic drivers out there? (The 4.2 documentation states that it is not possible to develop drivers with 4.2 --you need 3.? ) Rob Norman rjnorman@earthlink.net
From: John Zachary <zachary@bit.csc.lsu.edu> Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: Wed, 04 Jun 1997 19:26:18 -0500 Organization: LSU Robotics Research Laboratory Message-ID: <339607AA.5386@bit.csc.lsu.edu> References: <5m2u4j$2b5$3@news1.ucsd.edu> <338b6960.76768104@news.pdnt.com> <338B8919.6351@earthlink.net> <338B8EA4.1D57@earthlink.net> <AlvinKoh-2805971604500001@fsng172mac.ops.sing.paging.mot.com> <5mhvfv$k1i$1@news.digifix.com> <33908F5C.3B0A@wootech.com> <33917b98.71663816@news.pdnt.com> <3391EFFB.552E@research.canon.com.au> <3395514a.24565841@news.pdnt.com> <3395FDEC.6FFA@research.canon.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit cstory@research.canon.com.au wrote: > If you were a developer, you would know better than to expect fair > treatment from Apple anyway... > This is the best comment I've read all day! -- John Zachary LSU Robotics Research Laboratory
From: mwyner@apple.com (Michelle Wyner) Newsgroups: comp.sys.next.programmer Subject: thanks for the feedback! Date: Wed, 04 Jun 1997 17:54:35 -0700 Organization: Apple Computer Message-ID: <mwyner-0406971754350001@wynemi.apple.com> was: rhapsody sample code ideas Thanks to all those who sent me feedback on what they think would be useful sample code to write for Rhapsody, and what sample code (OpenStep/NEXTSTEP) has been useful in the past. As always, if you have any more comments/ideas on what would be good sample code to write for Rhapsody, feel free to email me at mwyner@apple.com. -- Michelle Wyner Apple Developer Tech Support Rhapsody Sample Code Warrior
From: Timothy Luoma <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: quadfat 'finger' Date: Wed, 4 Jun 1997 17:49:01 -0700 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.SUN.3.96.970604174516.4085A-100000@kira> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII trying to compile finger.1.04 under ns3.3 or 4.1 I get this same message: cc -arch m68k -arch i386 -arch sparc -pipe -O -DLOGGING -DAUTHUSER -c finger.c cc -arch m68k -arch i386 -arch sparc -pipe -O -DLOGGING -DAUTHUSER -c lprint.c For architecture m68k: lprint.c: In function `show_text': lprint.c:284: `errno' undeclared (first use this function) lprint.c:284: (Each undeclared identifier is reported only once lprint.c:284: for each function it appears in.) *** Exit 1 Stop. Any clues as to how to get around this? Thanks TjL -- TjL <luomat@peak.org> / http://www.peak.org/~luomat/next/ "The best things in life are made into inferior versions and bundled with the latest Microsoft systems" Bookmarks: http://www.peak.org/~luomat/next/bookmarks.html
From: rob@neurodata.com (Robert Norman) Newsgroups: comp.sys.next.programmer Subject: Parallel port driver use? Date: Wed, 04 Jun 1997 19:50:51 -0700 Organization: NEURODATA Message-ID: <rob-0406971950520001@pool059-max1.pasadena.ca.us.dialup.earthlink.net> OK... I previously asked about the availability of a driver for the parallel port under OpenStep 4.2 (WWDC). Now I see in Configure.app that there seems to be a driver installed "On-Board ParallelPort (v4.00)" but how is it used? Where do I find the documentation? I'm sure it's in there somewhere but.... Hope you-all can help. -- Rob Norman rjnorman@earthlink.net osition and all the other rows adjust. Is NSTableView an appropriate place to do this and where would I start. Thanks for the support. This is just a great system! -- Rob Norman rjnorman@earthlink.net
From: dcoyle@ctp.com (David A. Coyle) Newsgroups: comp.sys.next.programmer Subject: Looking for Frameworks in all the wrong places... Date: 5 Jun 1997 08:44:21 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5n5u95$5a8$1@concorde.ctp.com> Hi all: I'm having a problem that, while not serious, is proving a major productivity killer. Yesterday, I had changed the Install In: target for two frameworks I'm working on to be some directory (/LocalDeveloper/Frameworks from $(HOME)/Library/Frameworks). Cleaned 'em, made install, etc. And then recompiled an app that uses the frameworks. No luck: although I changed all the framework search orders, even removed and re-added the frameworks (in PB) at the new locations, the app would compile fine, but when I ran it, it would exit, saying it was unable to load the dynamic library at the *old location*. grepping through the project directory yielded absolutely no reference to that old location, however. (!) So, end of the day, I bit the bullet and moved the frameworks back to $(HOME)Library/Frameworks. Ditto for changing the search orders, etc. Scrubbed all references to /LocalDeveloper/Frameworks Everything builds fine (looks in the right places, too) So what happens when I try to _run_ the app now? It looks in /LocalDeveloper/Frameworks! I spent half of yesterday trying to get it to look there, without avail, and now it does, just when I no longer want it to?!?!?! What the heck is going on???? 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: michael@ninebits.com (Michael Balle) Newsgroups: comp.sys.next.programmer Subject: Vertical SplitView Date: Thu, 5 Jun 1997 12:44:31 +0200 Organization: Nine Bits Message-ID: <19970605124431743973@[222.223.224.4]> Hi, the release notes that comes with the "Prelude to Rhapsody" states that the NSSplitView has been enhanced to support vertical splitting. How do you create a vertical splitter from IB?. I was quite easy to do the horizontal split view, but the NSSplitView does not have a property indicating whether it is splitting vertical or horizontally. regards Michael
From: wendling@next.univ-rennes1.fr (Fabrice Wendling) Newsgroups: comp.sys.next.programmer Subject: DISPLAYING ON MULTIPLE COMPUTERS Date: 5 Jun 1997 12:07:32 GMT Organization: Universite de Rennes 1, France Message-ID: <5n6a64$6ba$1@news.univ-rennes1.fr> Keywords: DISPLAY Hello, I'm looking for an application that would be able to broadcast the display of one NeXTSTEP computer to one or more other computers on the same network. I found an application called ScreenCast, by Otherwise Co. However, it was a demo version. I tried to join Otherwise by mail, without success. Does anybody have some hints about an application or a easy way to display a computer screen on other computers ? Thanks,
From: Keith Jacob <postmaster@127.0.0.1> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX OOP class proposal & contest Date: Thu, 05 Jun 1997 17:50:27 +2910 Organization: The University of Adelaide Message-ID: <349676CB.A6D@127.0.0.1> References: <5mo55p$t5j$1@news.digifix.com> <AFB4FD83-CD19@206.165.44.25> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lawson English wrote: > > Scott Anguish <sanguish@digifix.com> said: > Lawson English <english@primenet.com> said: > > > > > >Unless, of course, I actually suceed in making most of the GX API > > >accessable and useable to HyperCard developers... > > > > > > And even still it has almost NO relevance to this. > > > > GX hasn't had adopters largely due to its not being wanted by > > most people. Not to mention they haven't heard about it! Bad marketing? > > > > I hardly think your little stack interface programming deal > > will save GX. > > It's all in how something is marketed, you know... No, he probably doesn't know...almost every contributor [I have read] to these newsgroups conveniently forgets this point. Maybe it is ignorance, but more likely it is because they are idealists. Good marketing could have saved Apple from their current problems, even WITH the "Resting on Laurels". Bad marketing and terrible lapses in quality control COMBINED with a lack of foresight have almost sunk them. BTW, I AM an Apple supporter...I just don't see how ignoring reality helps anyone...except idealists! Keith -- Spammers will need to remove their worth from my address. Everyone else, remove none.
From: "Jurriaan van der Lingen" <jurriaan@fygir.nl> Newsgroups: comp.sys.next.programmer Subject: Loading bundles in Windows NT Date: 5 Jun 1997 15:37:36 GMT Organization: XS4ALL, networking for the masses Sender: fygir@194.229.196.119 Message-ID: <01bc71c6$5bd39ef0$77c4e5c2@tatra> Does anyone have experience with dynamic bundles in windows NT. I have bundles included as "bproj" in the project. This runs fine with dynamic bundles on mach (OPENSTEP 4.0) Also it runs fine on Windows NT when I use the very same code in "subproj" instead of "bproj" However, when I use bundles in Windows NT (OPENSTEP 4.1), it crashes: 10000000:<MyApp>\RESOURCES\<MyBundle>.BUNDLE\<MyBundle>.DLL Program received signal SIGSEGV, Segmentation fault. 0x31008a4c in objc_msgSend () Help appreciated, Thanks!
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer Subject: Window redrawing? Beginner question. Date: 5 Jun 97 12:18:25 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFBC5F14-43E030@141.214.134.235> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit I have a slider that controls a text box. As I slide the slider the text box grows and rotates (stop laughing - this is my very first Openstep program). However, the background window doesn't redraw itself. I end up with a white spirograph like thing on a gray window. This is even when I try to force the redraw of the window (but from reading the classes it sounds like this should be automatic). Here's my code that does everything. As you can see I'm just screwing aronud: - (void)scroller:(id)sender { int val; float fval; NSRect textViewBounds; NSSize sizeOfTextField; NSWindow *theMainWindow; val = [sender intValue]; fval = [sender floatValue]; textViewBounds = [textBox bounds]; sizeOfTextField.width = fval/10; sizeOfTextField.height = fval/10; [textBox setIntValue:val]; [textBox rotateByAngle:fval]; [textBox setBoundsRotation:fval]; [textBox setFrameRotation:fval]; [textBox setFrameSize:sizeOfTextField]; theMainWindow = [NSApp mainWindow]; [theMainWindow setAutodisplay:YES]; [theMainWindow flushWindow]; [theMainWindow update]; [NSApp updateWindows]; } 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
Newsgroups: comp.sys.next.programmer Subject: Re: Driver for Parallel Port? Message-ID: <1997Jun5.094807.98962@cc.usu.edu> From: edx@cc.usu.edu Date: 5 Jun 97 09:48:06 MDT References: <5n4rm6$43@chile.earthlink.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: rob@neurodata.com In <5n4rm6$43@chile.earthlink.net> "Robert Norman" wrote: > Hi opensteppers! > > I'm prototyping a data acquisition program that will use custom drivers > later but for now, I'd just like to talk to the outside world through the > parallel port. Are there generic drivers out there? (The 4.2 documentation > states that it is not possible to develop drivers with 4.2 --you need 3.? ) > The version 1.0 release of NXCam which I recently put on Peanuts and Peak comes with a parallel port driver which is customized for developers needing to program the parallel port. It allows you to access all port registers from user-level code. And it's free. If you need the developer docs with sample test code, give me a shout at edx at cc dot usu dot edu
From: rob@neurodata.com (Robert Norman) Newsgroups: comp.sys.next.programmer Subject: Expand view bounds in scroller? Date: Thu, 05 Jun 1997 09:54:02 -0700 Organization: NEURODATA Message-ID: <rob-0506970954030001@pool065-max5.pasadena-ca-us.dialup.earthlink.net> Hi again... I'm trying to put together a demo for a prospective client and need to get one more feature locked down... I have a waveform view -- say it's a sound file -- and I'd like to expand the view in horizontal direction to expand the wave. I've tried expanding my view's bound.size.width to no effect. I've tried various kinds messaging to the views clipview and scrollview using "reflectScrolledClipView" etc. My view does not draw longer nor does the scroller reflect the change in size by changing the scroll control size. I've examined the some of the other examples but can't find out how to change my views length and have it reflected in the scroller. Help please! -- Rob Norman rjnorman@earthlink.net
From: sd;ljfs;@lassdjfsl.com Newsgroups: comp.sys.next.programmer Subject: Use What the Pros Use: Mailloop Date: 5 Jun 1997 20:12:44 GMT Organization: Sprint Internet Passport Message-ID: <5n76js$v@newsfep3.sprintmail.com> Mailloop v3.0 is an industrial strength bulk mailer. Mailloop v3.0 is a bulk mailing program for the internet. It can take a single message and broadcast it to either a listing of e-mail addresses or to a listing of newsgroups. Creates custom newsgroup lists by filtering NNTP servers. Creates custom e-mail lists by extracting them from newsgroups. Can also extract e-mail addresses or newsgroup names from other protocols. Built-in NNTP, SMTP, POP, FTP, HTTP, EXEC, CGI, WHOIS and FINGER clients. Anti-cancelbot feature. Automatically processes remove requests. Includes extensive on-line help. The most powerful bulk mailer available. Requires Windows 95 or NT 3.51+ and internet access. The "Program" Mailloop is only available in English. For more info goto: http://205.199.4.219 http://205.199.4.219 Using it is easy: 1) Create a message ( http://205.199.4.219/editor.htm ) ...by using the pull down menus or ...by using any editor 2) Create a newsgroups list ( http://205.199.4.219/news3.htm ) ...by filtering an NNTP server or ...by importing from any text file or ...by manually creating with any editor or ...by extracting from any FTP file or ...by extracting from any HTTP file 3) Create an e-mail list ( http://205.199.4.219/em3.htm ) ...by extracting from newsgroups or ...by importing from any text file or ...by manually creating with any editor or ...by extracting from a WHOIS response or ...by extracting from a fingering response or ...by extracting from a UNIX response or ...by extracting from any FTP file or ...by extracting from any HTTP file or 4) Broadcast the message ...to the e-mail list or ( http://205.199.4.219/embc.htm ) ...to the newsgroup list ( http://205.199.4.219/ngbc.htm ) 5) Then process the remove requests ...by using the mailbox processor ( http://205.199.4.219/pop.htm ) 6) If you want you can use the Newsletter Sever ...The Newsletter Sever will allow you to have an topic-specific newsletter that other can subscribe and unsubscribe to. ...Customizing this server response files ( http://205.199.4.219/response.htm ) ...Creating a new newsletter ( http://205.199.4.219/create.htm ) ...Creating and Updating a newsletter the actual newsletter ( http://205.199.4.219/update.htm ) For more info visit http://205.199.4.219 http://205.199.4.219
From: Michael.Gentry@m-c-i.com (Michael Gentry) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 5 Jun 1997 21:38:19 GMT Organization: Internet MCI Message-ID: <5n7bkb$190$1@news.internetmci.com> References: <5mo5jb$t71$1@news.digifix.com> <AFB501C4-1CCD2@206.165.44.25> <5moesg$36b$1@news.digifix.com> <5mosb1$9v6@proxye1.san.rr.com> In-Reply-To: <5mosb1$9v6@proxye1.san.rr.com> On 05/31/97, "Ed Deans" wrote: > Scott Anguish wrote in article <5moesg$36b$1@news.digifix.com>... >>On 05/30/97, "Lawson English" wrote: >> Given developers faith that Apple can and will ship a new OS >>before the end of the Century? > >Even on its deathbed, the worst case for Copland was a ship date before Jan >1, 2000. I thought the end of the century was December 31, 2000? - mrg -- "Java's fame is due to the massive public relations campaign, and not to any eminent technical merits." - Niklaus Wirth
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5n76js$v@newsfep3.sprintmail.com> Date: 5 Jun 1997 20:57:59 GMT Control: cancel <5n76js$v@newsfep3.sprintmail.com> Message-ID: <cancel.5n76js$v@newsfep3.sprintmail.com> Sender: sd;ljfs;@lassdjfsl.com Spam cancelled by sef@kithrup.com
From: jq@papoose.quick.com (James E. Quick) Newsgroups: comp.sys.next.programmer Subject: Re: Physically contiguous memory in DriverKit drivers Date: 5 Jun 1997 15:21:30 -0400 Organization: Quick and Associates Message-ID: <5n73jq$ede@papoose.quick.com> References: <5ls7gu$ga5@news.sns-felb.debis.de> In article <5ls7gu$ga5@news.sns-felb.debis.de>, Martin Bvckle <bueckle@schelling.dbag.ulm.DaimlerBenz.COM> wrote: >Hi to all DriverKit gurus, > >Does anybody know how to alloc large pieces (for instance 16MB) >of physically contiguous memory within a driver? In priciple, >IOMallocLow can be used for the job. But this function only >works in the low 16MB of the memory, therefore it is not >possible to alloc whole 16MB areas. 16MB, that's pretty steep. Make sure you really need this much memory. If you are sure that the allocation will not occur during interrupt handling you could use kalloc(). Note that this returns memory which is wired down (cannot be paged out). Is this a driver for a real device, or is it a pseudo driver? Will it recieve any hardware interrupts? If it won't you might want to unwire much of the memory you allocate to ensure that it doesn't cause other performance problems. -- ___ ___ | 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: mwyner@apple.com (Michelle Wyner) Newsgroups: comp.sys.next.programmer Subject: Re: Vertical SplitView Date: Thu, 05 Jun 1997 16:27:14 -0700 Organization: Apple Computer Message-ID: <mwyner-0506971627140001@wynemi.apple.com> References: <19970605124431743973@[222.223.224.4]> In article <19970605124431743973@[222.223.224.4]>, michael@ninebits.com (Michael Balle) wrote: > Hi, > > the release notes that comes with the "Prelude to Rhapsody" states that > the NSSplitView has been enhanced to support vertical splitting. How do > you create a vertical splitter from IB?. I was quite easy to do the > horizontal split view, but the NSSplitView does not have a property > indicating whether it is splitting vertical or horizontally. > > regards > Michael Michael, It's not something you do from Interface Builder. If you read a little bit farther down in the Release Notes, it mentions 2 new methods for the NSSplitView class: - (BOOL)isVertical; - (void)setVertical:(BOOL)flag; So if you pass in YES to "setVertical", it'll flip the split view to a vertical one. -- Michelle Wyner Apple Developer Tech Support Rhapsody Sample Code Warrior
From: Matt Watson <mgw@pacbell.net> Newsgroups: comp.sys.next.programmer Subject: Re: Looking for Frameworks in all the wrong places... Date: Thu, 05 Jun 1997 17:33:40 -0700 Organization: Thaumaturge, Inc. Message-ID: <33975ADB.3C4D@pacbell.net> References: <5n5u95$5a8$1@concorde.ctp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit When you link against a framework, which has its "install in" location embedded in it, this location gets embedded in the executable as well. If you do an "otool -L" on the framework and binary, you'll see the install name of the framework, and the location at which the binary expects to find the framework. These locations are somewhat separate from the framework search paths you can enter in the ProjectBuilder inspector. Hope this helps, matt.
From: stephlise@telco.com (Steph&Lise.) Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <998b11f9.7055665@hiroken.hiroken.or.jp> Control: cancel <998b11f9.7055665@hiroken.hiroken.or.jp> Message-ID: <despam.998b11f9.7055665@hiroken.hiroken.or.jp> Date: Tue, 3 Jun 1997 01:00:00 -700 Cancel Spam.
From: jabi@acsu.buffalo.edu Newsgroups: comp.sys.next.programmer Subject: Outline Object Date: Thu, 05 Jun 1997 22:02:41 -0400 Organization: University At Buffalo Message-ID: <33976FC1.35D3@arch.buffalo.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-User: jabi Hello: Does anyone have a Tree or an Outline object for OPENSTEP? Thanks. - Wassim Jabi
Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.programmer From: edew@netcom.com (Eric Dew) Subject: Re: NeXT People Question Message-ID: <edewEBC19q.H4D@netcom.com> Sender: edew@netcom2.netcom.com Organization: Netcom Online Communications Services (408-241-9760 login: guest) References: <865389689.23631@dejanews.com> <5n4247$odp@crl.crl.com> Date: Fri, 6 Jun 1997 02:22:37 GMT In article <5n4247$odp@crl.crl.com> mcgredo@nospam.crl.com writes: >In article <865389689.23631@dejanews.com>, <Tracla@aol.com> wrote: >>Hello - my name is Tracey.. I am working with AT&T in HR as a partner. >>We are looking for NeXT System Administrators.. >> >>But my question is, since this is a new software product for >>me is: >> >>Do people outside of NeXT software have these skills? > >Since NeXT is BSD Unix under the skin, anyone with a Unix >background should be able to pick it up fairly quickly. There's >some flashy NeXT sysadmin tools that make life easier, but >anyone who's competent should be able to pick that up quickly >enough. > And NeXTs are easier to administer than PCs (running NT, for example) or Macs or Solarises or HP-UXes. You can get two NeXT sys admins to manage up to 200 machines with little problems. >Rhapsody might have enough flashy sysadmin tools for a non- >unix type to get by, but that remains to be seen. > >>What are the salaries of people with this skill set? > >Uh, $500,000/yr. Yeah, that's the ticket. :-) > It depends on the region. In the Silicon Valley, $500K is a little on the low side :-) EDEW
From: com@com.com (com@com.com) Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.programmer Subject: Re: NeXT People Question Date: Fri, 06 Jun 1997 05:15:26 GMT Organization: com@com.com Message-ID: <33979c83.1360723@news.voyageronline.net> References: <865389689.23631@dejanews.com> <5n4247$odp@crl.crl.com> <edewEBC19q.H4D@netcom.com> On Fri, 6 Jun 1997 02:22:37 GMT, edew@netcom.com (Eric Dew) wrote: >>>What are the salaries of people with this skill set? >> >>Uh, $500,000/yr. Yeah, that's the ticket. :-) >> >It depends on the region. In the Silicon Valley, $500K is a little on the >low side :-) Eric, I would be curious (and other lurkers) to know what a NextStep programmer makes? In particular, those many & several programmers I see that post here. The ones that do work "off location" (i.e. their home or office) (vs. i.e. working for a traditional "company").
From: don@globalobjects.com (Don Yacktman) Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.programmer Subject: Re: NeXT People Question Date: 6 Jun 1997 06:01:18 GMT Organization: Global Objects Inc. Message-ID: <5n893e$obh$3@news.xmission.com> References: <865389689.23631@dejanews.com> <5n4247$odp@crl.crl.com> <edewEBC19q.H4D@netcom.com> <33979c83.1360723@news.voyageronline.net> com@com.com (com@com.com) wrote: > Eric, I would be curious (and other lurkers) to know what a NextStep > programmer makes? In particular, those many & several programmers > I see that post here. The ones that do work "off location" (i.e. their > home or office) (vs. i.e. working for a traditional "company"). Well, to inject some tiny hint of reality into this, the consultants charge what they can get, and that will vary greatly from one project to the next for most of them. If they are popular and well known, like Omni, they can get $250/hour on some contracts. If they don't have those sorts of credentials, they may be as cheap as, say a student who scrapes out code by night and goes for $15/hour. And there's a whole spectrum in between. So you could say that the NeXT developers are making something between $30k to $500k per year...though I _strongly_ suspect that most are nearer to the 60-100k range (depending upon skill) than either extreme. This is somewhat conjecture, somewhat based on what I know of others in the same field, but should be pretty close to the truth. I won't tell you what I make or charge, though, in any terms more precise than the above. :-) -- Later, -Don Yacktman don@misckit.com <a href="http://www.misckit.com/don.html">My home page</a>
From: Rene Berber <_rberber@spin.com.mx_> Newsgroups: comp.sys.next.programmer Subject: Re: Outline Object Date: 6 Jun 1997 05:44:06 GMT Organization: SPIN-Internet Message-ID: <5n8836$ffi$1@news.spin.com.mx> References: <33976FC1.35D3@arch.buffalo.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Cc: jabi@acsu.buffalo.edu In <33976FC1.35D3@arch.buffalo.edu> jabi@acsu.buffalo.edu wrote: > Hello: > Does anyone have a Tree or an Outline object for OPENSTEP? > Thanks. > The MiscKit 1.9 does have an outline view, look into the MiscTreeBrowser class and in the examples for MiscTreeBrowserTest, perhaps is what you are looking for. -- Rene Berber rberber@spin.com.mx (MIME/NeXTMail welcomed)
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.programmer,comp.sys.next.software Subject: Re: NeXT People Question Date: Fri, 6 Jun 1997 11:31:19 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <Mna2p7O00UhBA1mvM2@andrew.cmu.edu> References: <865389689.23631@dejanews.com> <5n4247$odp@crl.crl.com> <edewEBC19q.H4D@netcom.com> <33979c83.1360723@news.voyageronline.net> Excerpts from netnews.comp.sys.next.advocacy: 6-Jun-97 Re: NeXT People Question by com@com.com@com.com > Eric, I would be curious (and other lurkers) to know what a NextStep > programmer makes? In particular, those many & several programmers > I see that post here. The ones that do work "off location" (i.e. their > home or office) (vs. i.e. working for a traditional "company"). It depends on whether you choose to be a salaried employee or whether you want to play the consulting game. Consultants can make anywhere up to $250 an hour, but somewhere in the region of $125-$150 is going to be a lot more typical. Of course, a consultant has to find new work on a regular basis, and they can end up not working for a while if the market slows down. The low-end for salaried employees is about $30,000 - $35,000, and really good NEXTSTEP programmers can see over $100K, with $50K being a reasonable guestimate for the average salary for an experienced programmer. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: "Ross K. Leonard" <ross@everyware.com> Newsgroups: comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.next.software,comp.sys.mac.advocacy Subject: Re: Prelude WebObjects serial number? Date: 4 Jun 1997 21:26:46 GMT Organization: EDC Message-ID: <01bc712e$787a5a50$475fbace@ross> References: <AFB4C48B-1201E2@141.214.134.235> Robert A. Decker <comrade@umich.edu> wrote in article <AFB4C48B-1201E2@141.214.134.235>... > > I just finished installing the Prelude WebObjects stuff (on Mach) and > it's now asking me for a serial number. It's making reference to a > registration card. I've looked through the 'Getting Started on WebObjects' > book and the 'Installation Guide' booklet, but I can't find a reference to > the serial number. I also looked at the CD. > This seems to be everything I received at WWDC that has anything to do > with WebObjects. Has anyone else gotten around this? > > > rob I'm just working through the same install, and I believe the serial # is on the sticker on the Tyvek CD sleeve. However, when I use that serial number, I get an error while running serial.app (can't write to license.table). I don't know if they are related or not. -Ross-
Newsgroups: comp.sys.next.programmer From: Fabien_Roy@no.spam.free.fdn.fr Subject: Re: EOModeler to Database Message-ID: <EBDHxD.1qL@free.fdn.fr> Sender: news@free.fdn.fr Organization: Fabien Roy Consultant. References: <5n16c0$6ru@stern.fokus.gmd.de> Date: Fri, 6 Jun 1997 21:20:01 GMT In article <5n16c0$6ru@stern.fokus.gmd.de>, you wrote: > > I have EOF and a sybase database-server running on my black next > so far - so good > > now I want to access to the sybase database-server from another > next-machine in the network (for example: make a new EOModel from an > existing database with EOModeler) > HOW to do this ? > > Is there a way to type in the path in the database-login-panel off the > EOModeler ? > or another easy way to do that? > > thanks > > oliver 0) Shutdown the SQL server. 1) Edit the file /usr/sybase/interfaces 2) Replace al instances of "localhost" by the name of your host. 3) Copy the modified file in the client (/usr/sybase/interfaces) 4) Restart server. 5) Test from client Hope that helps. --- Fabien Roy --------------------------------------------------------------------- Fabien_Roy@free.fdn.org (NextMail/MIME accepted) Fabien Roy Consultant NEXTSTEP/OPENSTEP/EOF Consultant, SYBASE DBA 10 rue de la DEFENSE 93100 MONTREUIL, France Tel: 33 (0)1 45 28 32 23 Fax: 33 (0)1 48 55 09 90 GSM: 33 (0)6 60 46 36 83 advance for any insights, Lloyd Goldwasser goldwass@lifesci.lscf.ucsb.edu A side problem: NAN seems no longer to be working, either. I get the compiler warning ediv invalid operation error although all of my NAN code worked fine before, and I'm #including <math.h> and everything else that I was #including before. It's not that it doesn't recognize NAN, it's that it suddenly doesn't like it.
From: doyle@wrq.com Newsgroups: comp.sys.next.programmer Subject: Prelude WO won't launch - serial number required? Date: Thu, 05 Jun 1997 21:42:06 -0700 Organization: WRQ Message-ID: <3397951D.B32738B1@wrq.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 After installing all three Prelude WebObjects packages under Intel Mach (apparently successfully in the installer's opinion), I can't get WebObjectsBuilder to launch. On double-click, the app icon appears (dimmed) in the lower right corner of the screen for about a second, then just silently goes away. The http server is running and responding; virgin install of the whole Prelude package on a 1GB partition on a plain old 166 Pentium from WebObjects customer Dell Computer seems otherwise to work fine. I had no serial number to give the installer, which said I would be missing "significant functionality" without a serial number. This would certainly qualify as "significant!" Did others get a demo serial number? Did I lose a card or something? Any clue what would cause this behavior? I'd kinda like to try this stuff, but I'm flummoxed. dm
From: thompson@filoli.com (Peter Thompson) Newsgroups: comp.sys.next.programmer Subject: Looking for a good game framework... Date: 6 Jun 1997 23:41:28 GMT Organization: Relax, don't worry! Have a homebrew! Message-ID: <5na778$tlt$1@news.filoli.com> Hi folks, I'm looking for a framework that I can write a game or two around. Is there such a beast for OpenStep? I ran across an old copy of GameKit for NeXT, but I'm not sure I'm up to the porting task (it doesn't really work with NeXT3.3 tools either). Thanks for any leads, Peter Thompson. -- ************************************************************************** Peter Thompson ---- thompson@filoli.com ---- Filoli Information Systems. **************************************************************************
Newsgroups: comp.sys.next.programmer,comp.sys.next.misc,comp.lang.java.setup From: Fabien_Roy@no.spam.free.fdn.fr Subject: Re: Problems w/ Kaffe 0.9 [next-m68k] (undef'd syms _java_util_zip_*) Message-ID: <EBDJ8p.1uq@free.fdn.fr> Sender: news@free.fdn.fr Organization: Fabien Roy Consultant. References: <5m2ms1$t35$1@morgoth.sfu.ca> Date: Fri, 6 Jun 1997 21:48:24 GMT try -all_load option of cc excerpt from ReleaseNotes/CompilerTools.rtf * The new -all_load flag provides a way to link in all the members of the specified libraries. When this flag is present all library members for the specified libraries are always loaded from the library. This solves problems with respect to the use of rld(3), objc_loadModules(3) and NXBundles where the application wants to make available all of the library routines to the code it dynamically loads. This provides a more general solution that which is provided with the "-u libsys_s" like flags with respect to the NeXT supplied shared libraries. This solution works for all types of libraries including those not supplied by NeXT. -- Fabien Roy --------------------------------------------------------------------- Fabien_Roy@free.fdn.org (NextMail/MIME accepted) Fabien Roy Consultant NEXTSTEP/OPENSTEP/EOF Consultant, SYBASE DBA 10 rue de la DEFENSE 93100 MONTREUIL, France Tel: 33 (0)1 45 28 32 23 Fax: 33 (0)1 48 55 09 90 GSM: 33 (0)6 60 46 36 83
From: mpaque@wco.com (Mike Paquette) Newsgroups: comp.sys.next.programmer Subject: Re: Problems with compiling after OpenStep conversion Date: 6 Jun 1997 18:38:25 -0700 Organization: Electronics Service Unit No. 16 Sender: mpaque@mpaque Distribution: world Message-ID: <5nae2h$a1@mpaque.mpaque> References: <5n7cjb$osr@ucsbuxb.ucsb.edu> In article <5n7cjb$osr@ucsbuxb.ucsb.edu> goldwass@lifesci.lscf.ucsb.edu (Lloyd Goldwasser) writes: > All of my files now compile with nothing worse than warnings, and > then the compiler moves on to this file: > > /NextLibrary/Frameworks/Foundation.framework/Headers/NSObjCRuntime.h > > which gives me several parse errors ("before '@'", "before '*'"). > Then I find that the above file has been automatically added to the > project as a Non Project File (it wasn't there before). That's perfectly normal. It's not in your project, but now Project Builder can display the 'errant' lines in it's editor. > The error > message also says > > possible Objective-C token in C input. Use -ObjC > > which I take to be a suggestion to add it as an argument among the > build options. When I do so, /bin/gnumake complains that it's an > invalid option and stops. -ObjC is an option to be passed to the compiler, not as a build option (which passes it to gnumake, which complains...). You can add this to the OTHER_CFLAGS line in your project's Makefile.preamble: # Flags passed to compiler (in addition to -g, -O, etc) OTHER_CFLAGS = -ObjC The problem is almost certainly a header file containing Obj-C constructs, such as NSObjCRuntime.h, being included in a file ending with a .c suffix. The compiler front end sees that .c suffix and decides that it is looking at a vanilla C file, rather than an Obj-C file, and uses the vanilla C back end. The vanilla C back end sees the Obj-C constructs and complains. -- Mike Paquette (mpaque@wco.com) 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: stephane@lysis.ch (Stephane Corthesy) Newsgroups: comp.sys.next.programmer Subject: Re: Loading bundles in Windows NT Date: 6 Jun 97 10:12:57 GMT Organization: Lysis S.A. Message-ID: <3397e2a9.0@news.planet.ch> References: <01bc71c6$5bd39ef0$77c4e5c2@tatra> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: jurriaan@fygir.nl In <01bc71c6$5bd39ef0$77c4e5c2@tatra> "Jurriaan van der Lingen" wrote: > Does anyone have experience with dynamic bundles in windows NT. > I have bundles included as "bproj" in the project. > This runs fine with dynamic bundles on mach (OPENSTEP 4.0) > Also it runs fine on Windows NT when I use the very same code in "subproj" > instead of "bproj" > > However, when I use bundles in Windows NT (OPENSTEP 4.1), it crashes: > > 10000000:<MyApp>\RESOURCES\<MyBundle>.BUNDLE\<MyBundle>.DLL > Program received signal SIGSEGV, Segmentation fault. > 0x31008a4c in objc_msgSend () > > > Help appreciated, Thanks! > In NeXTanswer ??? (I don't remember which one...), you can find the answer to this problem with NT: you need to add the frameworks you use into the bundles too, not only into the main project. You wouldn't need it if you don't use ANY functions or external variables from the frameworks, like NSApp, or NSLog(). StÝphane -- "L'ordinateur n'a pas l'intelligence qu'on lui prete, mais celle qu'on lui donne." Stephane Corthesy Lysis S.A. Rue des Cotes de Montbenon 8 CH-1003 Lausanne Switzerland Tel. +41.21.312.91.91 Fax +41.21.312.93.43 E-mail: stephane@lysis.ch (NeXTMail welcome)
From: schack@skyler.arc.ab.ca (Brian Schack) Newsgroups: comp.sys.next.programmer Subject: Re: Loading bundles in Windows NT Date: 06 Jun 1997 08:39:17 -0600 Organization: Alberta Research Council, Calgary Alberta, Canada Sender: schack@skyler.arc.ab.ca Message-ID: <vb4tbbyewa.fsf@skyler.arc.ab.ca> References: <01bc71c6$5bd39ef0$77c4e5c2@tatra> In-reply-to: "Jurriaan van der Lingen"'s message of 5 Jun 1997 15:37:36 GMT >>>>> "Jurriaan" == "Jurriaan van der Lingen" <jurriaan@fygir.nl> writes: Jurriaan> Does anyone have experience with dynamic bundles in Jurriaan> windows NT. I have bundles included as "bproj" in the Jurriaan> project. This runs fine with dynamic bundles on mach Jurriaan> (OPENSTEP 4.0) Also it runs fine on Windows NT when I Jurriaan> use the very same code in "subproj" instead of "bproj" Jurriaan> However, when I use bundles in Windows NT (OPENSTEP Jurriaan> 4.1), it crashes: Jurriaan> 10000000:<MyApp>\RESOURCES\<MyBundle>.BUNDLE\<MyBundle>.DLL Jurriaan> Program received signal SIGSEGV, Segmentation fault. Jurriaan> 0x31008a4c in objc_msgSend () Does the bundle try to reference a variable or function declared in the main program? If so, then you'll have to jump through some hoops (courtesy of the WindowsNT linker). Basically, all variables and functions that are need to be global to the bundles and the main program should be put into a framework (read "DLL"). I can give you details if this is in fact the problem. -- ---------------------------------------------------------------------------- Brian Schack |mailto:schack@arc.ab.ca | "I don't want to achieve Alberta Research Council |http://www.arc.ab.ca | immortality through my 6815 8th St NE | | work ... I want to achieve Calgary, Alberta |ph: (403) 297-7564 | it through not dying." Canada T2E 7H7 |fax: (403) 297-2339 | - Woody Allen ----------------------------------------------------------------------------
From: com@com.com (com@com.com) Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.programmer Subject: Re: NeXT People Question Date: Sat, 07 Jun 1997 05:30:21 GMT Organization: com@com.com Message-ID: <3398eef1.3101355@news.voyageronline.net> References: <865389689.23631@dejanews.com> <5n4247$odp@crl.crl.com> <edewEBC19q.H4D@netcom.com> <33979c83.1360723@news.voyageronline.net> <5n893e$obh$3@news.xmission.com> On 6 Jun 1997 06:01:18 GMT, don@globalobjects.com (Don Yacktman) wrote: >com@com.com (com@com.com) wrote: >> Eric, I would be curious (and other lurkers) to know what a NextStep >> programmer makes? In particular, those many & several programmers >> I see that post here. The ones that do work "off location" (i.e. their >> home or office) (vs. i.e. working for a traditional "company"). > >Well, to inject some tiny hint of reality into this, the would the above, working off 'location', be in line with reality, in your opinion? >consultants charge what they can get, and that will vary >greatly from one project to the next for most of them. If >they are popular and well known, like Omni, they can get >$250/hour on some contracts. If they don't have those sorts I guess I was talking about what the individual takes home. I seem to remember reading somewhere. .that the "company" will bill (i.e.) $200./hour the programmer will get about 1/2 this. . which is still about 50% above what someone on salary gets (with the trade off, being. . salary=job protection, more benefits, more stable life, etc.) >of credentials, they may be as cheap as, say a student who >scrapes out code by night and goes for $15/hour. And there's I would be curious to know if such animals do exist? (in NextStep that is). >a whole spectrum in between. So you could say that the NeXT >developers are making something between $30k to $500k per >year...though I _strongly_ suspect that most are nearer to the >60-100k range (depending upon skill) Skill. . . . and location. . . a person making a $100k/year in the People's Republic of California. . or the Worker's Paradise of New York City, would probably not be living as well as someone that has left those type locales. .and relocated to the backwoods of wherever. Hence the question about working "off location". Even the most remote parts of the nation are connected now. >This is somewhat conjecture, somewhat based on what I know of >others in the same field, but should be pretty close to the truth. >I won't tell you what I make or charge, though, in any terms more >precise than the above. :-) And I wouldn't want you too! I'm looking more for "general" terms. I guess you have the 'low' and the 'high'---and most people probably fall into that bell curve in the middle. On the other hand, how many people that use NextStep are mediocre enough to be described as average? hmmm... Thanks for the input. > >-- >Later, > >-Don Yacktman >don@misckit.com ><a href="http://www.misckit.com/don.html">My home page</a> >
From: com@com.com (com@com.com) Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.programmer,comp.sys.next.software Subject: Re: NeXT People Question Date: Sat, 07 Jun 1997 05:31:22 GMT Organization: com@com.com Message-ID: <3398ef04.3120517@news.voyageronline.net> References: <865389689.23631@dejanews.com> <5n4247$odp@crl.crl.com> <edewEBC19q.H4D@netcom.com> <33979c83.1360723@news.voyageronline.net> <Mna2p7O00UhBA1mvM2@andrew.cmu.edu> <5n9qga$cnu$4@news2.digex.net> On 6 Jun 1997 20:04:26 GMT, John Kheit <jkheit@cnj.digex.net> wrote: >I think it being a highly demanded, but short supplied market has >a lot to do with the higher rates.... All that will change in the >upcoming years (so live it up while you can :). This is a most interesting statement! Would you care to guess as to the figures? I know that NextStep was never a "wild" success in terms of lots of people using it, but WAS a success in terms of programmers making money. Are you talking 20% off the figures above? ---or something more drastic, like "80%"? Also, can you give a time frame? (lets assume Rhapsody ships on time etc.). When would the above figure(s) start to influence the market?
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.programmer,comp.sys.next.software Subject: Re: NeXT People Question Date: 7 Jun 1997 05:42:29 GMT Organization: monoChrome, Inc., NJ, USA Message-ID: <5nasc5$fva$1@news2.digex.net> References: <865389689.23631@dejanews.com> <5n4247$odp@crl.crl.com> <edewEBC19q.H4D@netcom.com> <33979c83.1360723@news.voyageronline.net> <Mna2p7O00UhBA1mvM2@andrew.cmu.edu> <5n9qga$cnu$4@news2.digex.net> <3398ef04.3120517@news.voyageronline.net> com@com.com (com@com.com) wrote: > On 6 Jun 1997 20:04:26 GMT, John Kheit <jkheit@cnj.digex.net> > wrote: > This is a most interesting statement! Would you care to guess > as to the figures? I know that NextStep was never a "wild" success > in terms of lots of people using it, but WAS a success in terms > of programmers making money. Are you talking 20% off the figures > above? ---or something more drastic, like "80%"? > Also, can you give a time frame? (lets assume Rhapsody ships on > time etc.). When would the above figure(s) start to influence > the market? Well, my best guestimate would be for those figures to asymtotically approach the industry average...The rate of change would depend upon how quickly and in what numbers people will go to the environment. I image w/in 3 years that OPENSTEP salaries get inline with the rest of the industry... Then again, it's pretty much speculation on my part :) -- 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: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer Subject: Why is setFrameOrigin: in NSView appear to be off 9 pixels? Date: 7 Jun 97 01:55:30 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFBE7016-34530E@141.214.134.235> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit (This is kind of embarrassing. The first time I sent this out I accidently sent it to comp.lang.java.programmer) I have an object subclassed from NSBox. I can drag the box object around the window and resize it, but things seem to be off 9 pixels. Here's the relevant stuff in my header: @interface MyNSBox : NSBox { NSPoint boxClickLocation; BOOL resizeOnDrag; } - (void) mouseDown:(NSEvent *) theEvent; - (void) mouseDragged:(NSEvent *) theEvent; - (BOOL) isPointInResizeArea:(NSPoint) thePoint; @end and here's the relative functions in my code: - (void) mouseDown:(NSEvent *) theEvent { NSPoint windowClickLocation = [theEvent locationInWindow]; boxClickLocation = [self convertPoint:windowClickLocation fromView:nil]; if ([self isPointInResizeArea:boxClickLocation]) { resizeOnDrag = YES; } else { resizeOnDrag = NO; } } - (void) mouseDragged:(NSEvent *) theEvent { NSPoint mouseLocation = [theEvent locationInWindow]; if (resizeOnDrag) { // bunch of stuff } else { NSPoint finalLocation; finalLocation.x = (mouseLocation.x - delta.x); finalLocation.y = (mouseLocation.y - delta.y - 9); [self setFrameOrigin:finalLocation] } [self setNeedsDisplay:YES]; [[self superview] setNeedsDisplay:YES]; } In the mouseDragged function you can see that when I set finalLocation.y variable I have to subtract 9 from the y value of the mouse. If I don't the box jumps up 9 pixels when I start to drag it. This also happens when I resize the box (except in this case the resize is 9 pixels higher than where the mouse is). I can do this little hack to fix it, but I'd like to know why it's happening. 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: atze@aspohr.dart.de (Alexander Spohr) Newsgroups: comp.sys.next.programmer Subject: WO: You backtracked too far Date: 6 Jun 1997 11:34:23 GMT Organization: InterShop Communications Message-ID: <5n8sjv$a9e@linux1.netconx.de> ideas and help needed! we are developing a WO-application which uses frames. at the top there is a frame with a menu-bar which is build at the start of the session. if the user clicks around in the lower frames for a while and then uses the menu-bar we just get "You backtracked too far". to avoid this backtrack-buffer overrun in the lower frames we started to recycle pages by using them again and again (subclassed pageWithName:). but the problem will still be there if we have more than X different pages or some we can't recycle this way. so the question: how can we avoid this without updating the menu-bar every time? has anyone solved this problem? Atze ------------------------ | menu-bar | ------------------------ | | | | | | | frame1 | frame2 | | | | | | | ------------------------
From: michael@ninebits.com (Michael Balle) Newsgroups: comp.sys.next.programmer Subject: Re: Vertical SplitView Date: Fri, 6 Jun 1997 09:59:52 +0200 Organization: Nine Bits Message-ID: <19970606095952134237@[222.223.224.4]> References: <19970605124431743973@[222.223.224.4]> <mwyner-0506971627140001@wynemi.apple.com> Michelle Wyner <mwyner@apple.com> wrote: Michelle, thanks for the answer. Are you going to add support to IB to control this behavior?. Michael > It's not something you do from Interface Builder. If you read a little > bit farther down in the Release Notes, it mentions 2 new methods for the > NSSplitView class: > > - (BOOL)isVertical; > - (void)setVertical:(BOOL)flag; > > So if you pass in YES to "setVertical", it'll flip the split view to a > vertical one.
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.programmer,comp.sys.next.software Subject: Re: NeXT People Question Date: 6 Jun 1997 20:04:26 GMT Organization: monoChrome, Inc., NJ, USA Message-ID: <5n9qga$cnu$4@news2.digex.net> References: <865389689.23631@dejanews.com> <5n4247$odp@crl.crl.com> <edewEBC19q.H4D@netcom.com> <33979c83.1360723@news.voyageronline.net> <Mna2p7O00UhBA1mvM2@andrew.cmu.edu> Charles William Swiger <cs4w+@andrew.cmu.edu> wrote: > It depends on whether you choose to be a salaried employee or > whether you want to play the consulting game. Consultants can > make anywhere up to $250 an hour, but somewhere in the region of > $125-$150 is going to be a lot more typical. Of course, a > consultant has to find new work on a regular basis, and they can > end up not working for a while if the market slows down. > The low-end for salaried employees is about $30,000 - $35,000, > and really good NEXTSTEP programmers can see over $100K, with > $50K being a reasonable guestimate for the average salary for an > experienced programmer. Boy, I think it depends on region, b/c here in NY the salaries I've heard of are a LOT higher. I've heard consultants getting round $350/hr (sheesh as bad as lawyers :) but on average $175-225, and full time work for a decently experienced NeXT programmer (3+ years) is in the 80-100k on average, and 130k on up for the super duper guys... I think it being a highly demanded, but short supplied market has a lot to do with the higher rates.... All that will change in the upcoming years (so live it up while you can :). -- 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: penrose@sfc.keio.ac.jp (Penrose Christopher) Newsgroups: comp.sys.next.programmer Subject: NSScrollView won't scroll NSMatrix Date: 07 Jun 1997 16:52:48 GMT Organization: Keio University Shonan Fujisawa Campus, Kanagawa Japan Distribution: world Message-ID: <PENROSE.97Jun8015248@ccs01.sfc.keio.ac.jp> Hi Folks! I have created an NSMatrix inside of an NSScrollview: [self setHasVerticalScroller:YES]; [self setScrollsDynamically:YES]; [self setAutoresizesSubviews:YES]; [self setAutoresizingMask:NSViewWidthSizable|NSViewHeightSizable]; tempRect.size = [NSScrollView contentSizeForFrameSize:theFrame.size hasHorizontalScroller:YES hasVerticalScroller:NO borderType:borderType]; matrix = [[NSMatrix alloc] initWithFrame:tempRect]; [self addSubview:matrix]; [self setDocumentView:matrix]; [matrix setAutoresizingMask:NSViewWidthSizable|NSViewHeightSizable]; [matrix setAutoscroll:YES]; /* YES, please scroll damnit!! */ [[matrix superview] setAutoresizesSubviews:YES]; In the application, the NSScrollView never produces scroll handles even though the NSMatrix document view expands beyond the frame of the NSScrollView's content view. I use: [theMatrix addRow]; [theMatrix setNeedsDisplay:YES]; to add new rows to the NSMatrix. I first tried to save time and I initially used interface builder. I achieved identical results: the NSScrollView never updates. Any help would be greatly appreciated. Chris Penrose penrose@sfc.keio.ac.jp penrose@music.princeton.edu
From: don@globalobjects.com (Don Yacktman) Newsgroups: comp.sys.next.programmer Subject: Re: Looking for a good game framework... Date: 7 Jun 1997 17:28:54 GMT Organization: Global Objects Inc. Message-ID: <5nc5om$h51$1@news.xmission.com> References: <5na778$tlt$1@news.filoli.com> thompson@filoli.com (Peter Thompson) wrote: > Hi folks, > > I'm looking for a framework that I can write a game or two around. Is > there such a beast for OpenStep? I ran across an old copy of GameKit > for NeXT, but I'm not sure I'm up to the porting task (it doesn't > really work with NeXT3.3 tools either). I'm the original author of that kit. I will be doing the porting and, actually, a complete overhaul of the kit. I don't know *when* it will be ready, but announcements will be made as soon as I begin to release parts of it. The plan is to release it as various parts become available, rather than waiting for the whole kit to be finished so that people can start using at least those parts. -- Later, -Don Yacktman don@misckit.com <a href="http://www.misckit.com/don.html">My home page</a>
Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.programmer From: brianw@sounds.wa.com (Brian Willoughby) Subject: Re: NeXT People Question Message-ID: <EBF342.5v7.0.scream@sounds.wa.com> Organization: Sound Consulting, Bellevue, WA, USA References: <865389689.23631@dejanews.com> <5n9qga$cnu$4@news2.digex.net> <3398ef04.3120517@news.voyageronline.net> <5nasc5$fva$1@news2.digex.net> Date: Sat, 7 Jun 1997 17:55:14 GMT In article <5nasc5$fva$1@news2.digex.net>, John Kheit <jkheit@cnj.digex.net> wrote: >Well, my best guestimate would be for those figures to asymtotically >approach the industry average...The rate of change would depend >upon how quickly and in what numbers people will go to the environment. >I image w/in 3 years that OPENSTEP salaries get inline with the >rest of the industry... Then again, it's pretty much speculation >on my part :) I would be curious to know what the industry average is for programmers in the markets that are most similar to OPENSTEP now. I have a couple of figures for the Seattle area, but I don't know if they're reliable or more wishful thinking. In other words, what are Object Oriented software developers making on contract in various parts of the US, or the world. My impression is that talented OO programmers command reasonably high rates which wouldn't upset a long-term NEXTSTEP/OPENSTEP professional. -- Brian Willoughby NEXTSTEP, OpenStep, Rhapsody Software Design Sound Consulting Bellevue, WA, U.S.A. Registered NeXT/Apple Enterprise Alliance Partner BrianW@SoundS.WA.com NeXTmail welcome
From: mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> Newsgroups: comp.sys.next.programmer Subject: Re: Looking for a good game framework... Date: 7 Jun 1997 15:37:02 GMT Organization: P&L Systems Message-ID: <5nbv6u$2pu$1@ironhorse.plsys.co.uk> References: <5na778$tlt$1@news.filoli.com> In-Reply-To: <5na778$tlt$1@news.filoli.com> On 06/07/97, Peter Thompson wrote: >Hi folks, > >I'm looking for a framework that I can write a game or two around. Is >there such a beast for OpenStep? I ran across an old copy of GameKit >for NeXT, but I'm not sure I'm up to the porting task (it doesn't >really work with NeXT3.3 tools either). > I believe Don Yacktman is porting this as part of the OpenStep MiscKit. Best wishes, mmalc. --
From: don@globalobjects.com (Don Yacktman) Newsgroups: comp.sys.next.programmer Subject: Re: Looking for a good game framework... Date: 7 Jun 1997 19:13:54 GMT Organization: Global Objects Inc. Message-ID: <5ncbti$h51$4@news.xmission.com> References: <5na778$tlt$1@news.filoli.com> <5nbv6u$2pu$1@ironhorse.plsys.co.uk> mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> wrote: > On 06/07/97, Peter Thompson wrote: > >[... GameKit ...] > I believe Don Yacktman is porting this as part of the OpenStep MiscKit. As noted in my other post, I am porting it. It will NOT be part of the MiscKit, however. It will be a seperate framework. It _will_ depend upon the MiscKit however. Several MiscKit objects are generalizations of objects that originated in the GameKit...which also means that you should be looking at the MiscKit, too. By the way, I've not yet decided what licensing terms will be for the GameKit. You can pretty much count on something different than the terms of the old kit, but I don't know what the new terms will be. What I'm fighting against here is that (a) if people were paying for it, I could devote more time to making it great, which I would really like to do but (b) I suspect few will be willing to pay for it. I could open it up for collaboration like the MiscKit, and that may or may not solve the problem. I'd also like to (c) make sure that if any license fees are reequired that the cost of entry to students and shareware programmers is very low. (Ie, charge fees, say, for commercial users but not for people writing freeware?) Another possibility is to make it free, but allow people to buy support or even just send in voluntary donations. All support and donated funds would then be fed into future development of the kit in a non-profit sort of way. This is sort of what I'm leaning towards, but I don't know if it will be good enough to support the kind of development time I'd like to devote to the project... Past experience tells me the kit will be most successful as a MiscKit-like scheme, but the market is about to change radically, so that may or may not be the best model for the future. Any comments on this--here or in private email--would be welcome, of course. :^) -- Later, -Don Yacktman don@misckit.com <a href="http://www.misckit.com/don.html">My home page</a>
From: embuck@palmer.cca.rockwell.com (Erik M. Buck) Newsgroups: comp.sys.next.programmer Subject: Re: Expand view bounds in scroller? Date: 7 Jun 1997 18:29:59 GMT Organization: Rockwell Collins Message-ID: <5nc9b7$5fo2@castor.cca.rockwell.com> References: <rob-0506970954030001@pool065-max5.pasadena-ca-us.dialup.earthlink.net> Cc: rob@neurodata.com In <rob-0506970954030001@pool065-max5.pasadena-ca-us.dialup.earthlink.net> Robert Norman wrote: > Hi again... > I have a waveform view -- say it's a sound file -- and I'd like to expand > the view in horizontal direction to expand the wave. I've tried expanding > my view's bound.size.width to no effect. I've tried various kinds > messaging to the views clipview and scrollview using > "reflectScrolledClipView" etc. > > My view does not draw longer nor does the scroller reflect the change in > size by changing the scroll control size. > The -setBounds: method will do what you want and it even sends a notification to the scroller if any to reflect the new bounds. (This works under 3.3... I have not tried with 4.2)
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5nc9jt$d7n$1022@usenet85.supernews.com> Date: 7 Jun 1997 20:02:10 GMT Control: cancel <5nc9jt$d7n$1022@usenet85.supernews.com> Message-ID: <cancel.5nc9jt$d7n$1022@usenet85.supernews.com> Sender: jasonben@aol.com Spam cancelled by sef@kithrup.com
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5nbtf8$2q5@inout.beachnet.com> Date: 7 Jun 1997 20:31:10 GMT Control: cancel <5nbtf8$2q5@inout.beachnet.com> Message-ID: <cancel.5nbtf8$2q5@inout.beachnet.com> Sender: sales@gclounge.com Spam cancelled by sef@kithrup.com
From: John Kheit <jkheit@cnj.digex.net> Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.programmer Subject: Re: NeXT People Question Date: 7 Jun 1997 20:31:56 GMT Organization: monoChrome, Inc., NJ, USA Message-ID: <5ncgfs$fnp$1@news2.digex.net> References: <865389689.23631@dejanews.com> <5n9qga$cnu$4@news2.digex.net> <3398ef04.3120517@news.voyageronline.net> <5nasc5$fva$1@news2.digex.net> <EBF342.5v7.0.scream@sounds.wa.com> brianw@sounds.wa.com (Brian Willoughby) wrote: > In article <5nasc5$fva$1@news2.digex.net>, John Kheit > <jkheit@cnj.digex.net> wrote: > >Well, my best guestimate would be for those figures to asymtotically > >approach the industry average...The rate of change would depend > >upon how quickly and in what numbers people will go to the > >environment. I image w/in 3 years that OPENSTEP salaries get > >inline with the rest of the industry... Then again, it's pretty > >much speculation on my part :) > I would be curious to know what the industry average is for > programmers in the markets that are most similar to OPENSTEP now. > I have a couple of figures for the Seattle area, but I don't know > if they're reliable or more wishful thinking. In other words, > what are Object Oriented software developers making on contract > in various parts of the US, or the world. My impression is that > talented OO programmers command reasonably high rates which > wouldn't upset a long-term NEXTSTEP/OPENSTEP professional. Well, I don't know the national figures. And the NYC figures I have are a combination of hearsay, and anecdotal experience. And of course it depends on experience... But if graduating student can expect a 30k/yr salary for doing basic C type coding, I'd say you would be in the 40k/yr doing OO work of import and significance (i.e. where the C++ project relied on the ++ in some significant way)... I would think that salary boost would apply throughout, but don't know for sure... And a graduating student with a fair bit of NeXT experience (say 1 yr) can usually land in at 50-60k/yr... -- 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: embuck@palmer.cca.rockwell.com (Erik M. Buck) Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.next.programmer Subject: Re: NeXT People Question Date: 7 Jun 1997 18:47:58 GMT Organization: Rockwell Collins Message-ID: <5ncacu$5fo3@castor.cca.rockwell.com> References: <865389689.23631@dejanews.com> <5n4247$odp@crl.crl.com> <edewEBC19q.H4D@netcom.com> <33979c83.1360723@news.voyageronline.net> Cc: com@com.com In <33979c83.1360723@news.voyageronline.net> com@com.com wrote: EXPERT nextstep programmers who ALSO have some knowledge of one of the many application fields using NeXTstep earn low 6 figures from remote offices and mid 6 figures for on-site work. Add 25% for New York City and some other expensive areas. NOVICE nextstep programmers earn about standard fees 50-75k / year One reason fees are high (apart from supply and demand) is that expert nextstep programmers are capable of creating apps for vertical or horizontal markets very rapidly and inexpensively. Employers are pay a premium for the "opportunity costs" of distracting the programmers from building and selling their own apps. For instance, my company earns 7 figures for selling apps but only six figures for contract work. With nextstep, the risks of building apps without a paying customer are low. It is an extremely good business to be in right now.
From: mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> Newsgroups: comp.sys.next.programmer Subject: Re: Plotting points in a view Date: 8 Jun 1997 13:18:15 GMT Organization: P&L Systems Message-ID: <5neben$7om$1@ironhorse.plsys.co.uk> References: <5nd81d$hsp@sps1.phys.vt.edu> In-Reply-To: <5nd81d$hsp@sps1.phys.vt.edu> On 06/08/97, Nathan Urban wrote: >I'm trying to create a view subclass that can plot colored points. The >workhorse method is -plotX:(int)x andY:(int)y withColor:(NSColor *)aColor. >The main intent is to have a view where every pixel has a color, and to >be able to change a pixel's color. How should I implement this class? >I've been looking at making it a subclass of NSImageView, and plotting >points by modifying the data held by an NSBitmapImageRep and then >compositing the NSImage. Is this the best way to do it? It seems >unnecessarily complicated, what with having to convert the NSColor to >bit planes and then writing into the data array (I'm not even sure how >to do that), etc. I bet there's some DPS operator for plotting points >that I can use in an NSView's -drawRect:. What are the tradeoffs between >doing that and using an NSImageView? > Speed; there is no way plotting individual points will be as quick as an NSBitmapImageRep. Is this View going to be a constant size, or could the user resize it? >I guess in the former case, I'd have to subclass NSView and maintain my >own array of colors, which is basically what NSImageView is doing anyway.. >(Or maybe not.. I only need to plot.. so maybe I could just plot to a >view, and it would buffer it for me so I wouldn't have to know what I've >plotted in order to redraw.) > Well, you should really create a Model of what you want to draw, with its own data and plotting routine, and have this display in a View, its action mediated by a Controller object... If you just draw the new points into a buffered View and composite that onto screen it should be fairly quick *when you're adding new points*; for redraws of the whole thing, though, especially if you print it, it may get a bit unwieldy. If you do decide to use PostScript, however, it is better to use a single NSRectFillList call (which takes an array of rectangles and draws them all at once) rather than using many PSrectfill calls. >If I do go the NSImageView route, I guess I'd have to >modify the NSBitmapImageRep's bitmap data and then invoke >-compositeToPoint:operation: on its image in my -plotX:... method. > You could subclass NSBitmapImageRep and define your own colourPoint:(NSPoint) p withColour:(NSColor) colour method... This would access the NSBitmapImageRep's data directly -- you'd have a relatively simple method to work out which bits to twiddle based on the width and height of the image... You would probably use the drawAtPoint method rather than composite... >(Or instead of using an NSBitmapImageRep, should I subclass NSImageRep >or NSCustomImageRep and use a simple array of static NSColor instances >and override its -draw method to plot them all to a view? How would I >do that efficiently?) > I don't think you would! :-) >I suppose the disadvantage of using an >NSImageView is that I could only use it as a grid of colored pixels.. >that _is_ the main intent, but I might also want to draw text into it >later or something.. then again, I might put another view on top of it >with alpha transparency and draw into that.. > Yes, that would probably be better. For really fast work with bitmaps, note the observations Don Yacktman made a while back... Subject: Re: drawing bitmap to screen quickly From: don@globalobjects.com (Don Yacktman) Date: 1997/02/03 Message-Id: <5d5d4r$t2l@news.xmission.com> Newsgroups: comp.sys.next.programmer For the best speed, read the DPS release notes and performance notes that come with the NEXTSTEP documentation. What you want to do is to *exactly* match the buffer's layout to the screen RAM's layout (meshed vs. planar, bits per pixel, etc.) and then, on slabs, if your buffer is the right *width* (there's a complex formula in one of the performance notes) you'll get an extra boost. This is all very hardware dependent, but it is the same thing you'd have to do with Interceptor. (So you may want to make different drawing routines to deal with various buffer geometries to match different hardware platforms, for example.) In fact, because of the width thing (DPS notices an "optimal" width and does a special machine instruction for a wider copy when it sees it) it is harder to beat DPS when using Interceptor on color slabs without dropping into asm code yourself... I hope this helps, Best wishes, mmalc. --
From: theisen@akaMail.com (Dirk Theisen) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.codewarrior Subject: Re: Rhapsody DR to also support 9500 and 9600 PowerMacs Date: Sun, 8 Jun 1997 02:04:11 +0200 Organization: Dept. of Computer Science, University of Bonn, Germany Message-ID: <1997060802041120916@rhrz-isdn2-p8.rhrz.uni-bonn.de> References: <NoSpa.Mgustilo-3105971757190001@ts6-13.upenn.edu> Hi! <NoSpa.Mgustilo@mail.med.upenn.edu> wrote: > full text at this URL: <http://www.powermacintosh.com/pmr/rhapsodynews.shtml> I really cannot understand why is should delay the shipment of DR1 when they would qualify PM7600 systems. These have essentially the same motherboard as the 8500's and are only missing the video-out circuits (which will probably not supported). Can anyone explain this, PLEASE!? Dirk -- Computer Science, University of Bonn, Germany http://titan.informatik.uni-bonn.de/~theisen/
From: mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> Newsgroups: comp.sys.next.programmer Subject: Re: OOE client library and docs? Date: 8 Jun 1997 12:35:18 GMT Organization: P&L Systems Message-ID: <5ne8u6$730$1@ironhorse.plsys.co.uk> References: <SHAFFER.97May26231355@durer.phyast.pitt.edu> In-Reply-To: <SHAFFER.97May26231355@durer.phyast.pitt.edu> On 05/27/97, C. David Shaffer wrote: > >I'd like to build an OOE client (a document which can contain OOE >documents) but everything I've read says that the OOE client libs and >docs must be licensed from Xanthus (LightHouse?). Is this true? >Maybe I'll have to live with Object Links. > I think (unless anyone cares to contradict?) we can regard OOE as being effectively obselete; certainly if you're looking to the future I can't see it working on Rhapsody, so unless you have a compelling reason to use it for an app which will run "as is" for several years, I'd choose another method. Best wishes, mmalc. --
From: kc@ignem.omnigroup.com (Ken Case) Newsgroups: comp.sys.next.programmer Subject: Re: Project Header in 4.x PB? Date: 8 Jun 1997 19:38:55 GMT Organization: Omni Development, Inc. Message-ID: <5nf1of$cs7$1@gaea.omnigroup.com> References: <5ndk1t$p6l$1@vader.wolfware.ipc.net> Christopher Wolf (cwolf@wolfware.com) wrote: : Has anyone ever figured out what the "Project Header" check-box in : OpenStep 4.x Project Builder File Attributes inspector does? I've : never seen it documented anywhere. That publishes the header to all subprojects within the project (so that other subprojects can import it with just the name, not a path), but not to other projects (as "Public Header" does). -- Ken Case kc@omnigroup.com Omni Development, Inc. http://www.omnigroup.com
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: <26791865137630@digifix.com> Date: 8 Jun 1997 03:57:17 GMT Organization: Digital Fix Development Message-ID: <14012865742421@digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - ISV company pages - ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news: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. news: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. news: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. news: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. news: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. news: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!!! news: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. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news: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. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news: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 The ftp sites ============= ftp://ftp.next.peak.org - The main site for North American submissions formerly ftp.cs.orst.edu ftp://ftp.informatik.uni-muenchen.de: - (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next - Peanuts mirror in the US ftp://terra.stack.urc.tue.nl - (Dutch NEXTSTEP User Group) 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: nurban@sps1.phys.vt.edu (Nathan Urban) Newsgroups: comp.sys.next.programmer Subject: Plotting points in a view Date: 7 Jun 1997 23:13:49 -0400 Organization: Virginia Polytechnic Institute and State University Message-ID: <5nd81d$hsp@sps1.phys.vt.edu> I'm trying to create a view subclass that can plot colored points. The workhorse method is -plotX:(int)x andY:(int)y withColor:(NSColor *)aColor. The main intent is to have a view where every pixel has a color, and to be able to change a pixel's color. How should I implement this class? I've been looking at making it a subclass of NSImageView, and plotting points by modifying the data held by an NSBitmapImageRep and then compositing the NSImage. Is this the best way to do it? It seems unnecessarily complicated, what with having to convert the NSColor to bit planes and then writing into the data array (I'm not even sure how to do that), etc. I bet there's some DPS operator for plotting points that I can use in an NSView's -drawRect:. What are the tradeoffs between doing that and using an NSImageView? I guess in the former case, I'd have to subclass NSView and maintain my own array of colors, which is basically what NSImageView is doing anyway.. (Or maybe not.. I only need to plot.. so maybe I could just plot to a view, and it would buffer it for me so I wouldn't have to know what I've plotted in order to redraw.) If I do go the NSImageView route, I guess I'd have to modify the NSBitmapImageRep's bitmap data and then invoke -compositeToPoint:operation: on its image in my -plotX:... method. (Or instead of using an NSBitmapImageRep, should I subclass NSImageRep or NSCustomImageRep and use a simple array of static NSColor instances and override its -draw method to plot them all to a view? How would I do that efficiently?) I suppose the disadvantage of using an NSImageView is that I could only use it as a grid of colored pixels.. that _is_ the main intent, but I might also want to draw text into it later or something.. then again, I might put another view on top of it with alpha transparency and draw into that.. I'm also not quite sure how the NSImage caching works. If I change the NSImageRep's data, do I need to inform the NSImage so it can update its cache, or what? As you can tell, I really have no idea what I'm doing here.. :)
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Project Header in 4.x PB? Date: 8 Jun 1997 06:38:53 GMT Organization: WolfWare Message-ID: <5ndk1t$p6l$1@vader.wolfware.ipc.net> Has anyone ever figured out what the "Project Header" check-box in OpenStep 4.x Project Builder File Attributes inspector does? I've never seen it documented anywhere. - Chris -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: "Robert Norman" <rob@neurodata.com> Newsgroups: comp.sys.next.programmer Subject: Re: Expand view bounds in scroller? Date: Sun, 08 Jun 1997 14:08:51 -0700 Organization: NEURODATA Message-ID: <5nf74c$lev@argentina.earthlink.net> References: <rob-0506970954030001@pool065-max5.pasadena-ca-us.dialup.earthl ink.net> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit I previously had problems implementing what should have been -- and is -- a simple operation. Change the scale of a view expanding or contracting the horizontal scale with the scroll view properly reflecting the change. My mistake was in focusing on the views bounds rather than the frame. I was thinking of the frame as the space occupied on the screen, but that is the job of the scrolled views superviews . Changing the bounds seems to expand the postscript scale so that, for example, vertical lines got thicker as I expanded the bounds - that can be useful where you're zooming a view but not what I wanted. The solution is simple: -(void)changeSpeed: (id) sender { theFrame = [self frame]; if([sender tag]) // I'm using two buttons to expand and contract the view. theFrame.size.width *= 2; else theFrame.size.width /=2; [self setFrame: newFrame]; [self setNeedsDisplay:YES]; } and voila -- "it just works" . The view posts notification by default. Thanks to Tom Hageman for pointing me in the right direction. Hope this helps, (Perhaps there should be a channel for us WWDC newbies so that this channel is not tied up with mundane programming problems? ) Rob Norman Wave Master
From: Chuck_Esterbrook@orcacomputer.com Newsgroups: comp.sys.next.programmer Subject: StabsToCodeView Date: 9 Jun 1997 12:58:25 GMT Organization: Virginia Tech, Blacksburg, Virginia Message-ID: <5ngulh$iu9$1@solaris.cc.vt.edu> Does anyone know how to use StabsToCodeView.exe which comes with OPENSTEP 4.x for NT? I couldn't find any docs under NT, Mach or NeXTanswers. Thanks, Chuck@OrcaComputer.com
From: emclean@slip.net (Emmett McLean) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: 8 Jun 1997 00:49:44 -0700 Organization: Slip.Net Message-ID: <5ndo6o$d4n@slip.net> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> In article <338DB457.2579@abacus.com>, Jim Gagnon <jimg@abacus.com> wrote: > >Interesting. The long-time NeXT guys tell me that NeXTStep has real >issues as a server OS. I don't believe it. > Apparently, if you put more than twelve-or-so >users on it at a time, it really bogs down. Depends on the machine, an 030 running NS 3.0 with 16 Megs of RAM will slow up once there are about 15+ users. An 040/33 machine running NS 3.3 with 64 Megs of RAM will handle 25+ users nicely. That's my experience based upon my programming expericnce with NS 3.0 and NS 3.1 at San Francisco State. > Also, there's a limit of 200 processes in NeXTStep, which is >actually pretty low for even a client-side OS. The limit is more like 400. I know this because in the graduate Operating Systems each year there is a project on creating threads and almost every year someone forgets to put a time out wrapper on the threads he creates and brings down the entire system by creating hundreds of new processes. We've been able to watch the number of processes as the machine slows to death. 400 based upon this experience is the limit. After about 300 the the machine is effectively useless. > To top it off, a NeXTStep system really needs the >protection of a firewall -- security is definitely behind the levels >defined by Solaris and AIX. Well, I'd take NS over NT anyday. Are you comparing a sparc to a cube? Not a fair comparison since a sparc costs about 20 times more. Or are you refering to NS running under Solaris? > Apple may have cancelled updates to AIX (and who can blame them), but >they've got some work before Rhapsody is up to snuff as a standalone >Server. They could have done a lot worse than NS. I don't expect Rhapsody compete with Sun in the 100,000$+ market. But expect Rhapsody will be terrific, much better than it is now, in its areas of strength. Like DTP and video/audio stuff. NS is pretty good in these areas now - even without Apple. Emmett
From: jroepcke@compusmart.ab.ca Newsgroups: comp.sys.next.programmer Subject: Hair Pulling with my first stab at IB! (please help :-)) Date: 8 Jun 1997 18:54:53 GMT Organization: Roepcke Computing Solutions Message-ID: <5nev5t$bqg$1@news.sas.ab.ca> Hi Folks, First off, I'm a real OpenStep developer newbie... but I'm very pumped about shedding that status ASAP! I'm also a bit stubborn... here's an example: In the book "Discovering OpenStep: A Developer Tutorial", on page 27, it says: "Select the Images display of the nib file window. Drag the NSReturnSign image to the main window and drop it over the button. If you check the attributes of the button in the Inspector panel, you'll notice two things have been added: NSReturnSign is now listed as [...]"... Well friends, I don't have a NSReturnSign icon in the images display! (OpenStep 4.1 Intel, User+Dev+EOF 2.0) The picture shows: NSHighlighted NSHighlighted... NSRadioButton NSReturnSign NSSwitch I've got: NSHighlighted NSHighlighted... NSRadioButton NSSwitch This sucks! ;-) Where is it? I looked everywhere, and I can't find it... if I load the project /NextDeveloper/Examples/AppKit/CurrencyConverter/PB.project (or whatever it's called) and open the nib file, the button in the window shows the NSReturnSign icon, and the button's inspector lists it, but the images display doesn't have the icon in it! (I've tried resizing and scrolling the window -- it's not there...) Does anyone know what the problem is? Installing OpenStep Developer went fine, I think.... So, I'm stubborn because I don't want to continue until I know what's up with IB and this NSReturnSign. Now, I'm pretty sure that the book is at OpenStep 4.0 level, but I still want to know why I don't have that icon. My thinking is, "if that's missing, what else might be?". Can people please check their IB's image displays and tell me if they have that (on OS 4.1). And if possible, explain what's going on, and ensure me that I can/should continue on :-) I'd really appreciate it. Thanks a bunch, Jim Roepcke PS: If you could cc: your reply to my e-mail address, I'd really appreciate it... with all the e-mail I read nowadays (mailing lists), I rarely get time to read usenet :-( This is the first time in a long time, unfortunately.
From: penrose@sfc.keio.ac.jp (Penrose Christopher) Newsgroups: comp.sys.next.programmer Subject: NSSplitView AddSubview: gives lockFocus assertion error Date: 08 Jun 1997 12:04:08 GMT Organization: Keio University Shonan Fujisawa Campus, Kanagawa Japan Distribution: world Message-ID: <PENROSE.97Jun8210408@ccs00.sfc.keio.ac.jp> Hi Folks! I am trying to add more than one SoundView to an NSSplitView. The code: aSoundView = [[SoundView alloc] initWithFrame:newFrame]; [aSoundView setSound:aSound]; [aSplitView addSubview:aSoundView]; The above code gives this error when a second view is added: Jun 8 20:34:31 Mr.MixUp[1671] *** Assertion failure in -[MrSoundView lockFocus], NSView.m:1849 Jun 8 20:34:31 Mr.MixUp[1671] lockFocus sent to a view which is not in a window lockFocus complains that the view isn't in a window, but addSubview should associate the SoundView with a window. Any help would be greatly appreciated. Christopher Penrose penrose@sfc.keio.ac.jp
From: penrose@sfc.keio.ac.jp (Penrose Christopher) Newsgroups: comp.sys.next.programmer Subject: Detail: NSSplitView addSubview: lockFocus assertion failure Date: 08 Jun 1997 12:58:03 GMT Organization: Keio University Shonan Fujisawa Campus, Kanagawa Japan Distribution: world Message-ID: <PENROSE.97Jun8215803@ccs01.sfc.keio.ac.jp> Greetings NeXTpatriots: I must indicate that this scenario is occuring in Openstep 4.1. In my last post, I indicated that I was receiving lockFocus assertion failures when I attempted to add more than one SoundView to an NSSplitView: aSoundView = [[SoundView alloc] initWithFrame:newFrame]; [aSoundView setSound:aSound]; [aSplitView addSubview:aSoundView]; Jun 8 20:34:31 Mr.MixUp[1671] *** Assertion failure in -[MrSoundView lockFocus], NSView.m:1849 Jun 8 20:34:31 Mr.MixUp[1671] lockFocus sent to a view which is not in a window Using identical code, I created NSTextViews inside the NSSplitView with no problem at all. I can make as many as I like. Is this another never to be fixed SoundView bug? Does any one have any information regarding a possible workaround? I had this identical problem with another application I just wrote, my fix was to resignMain from the current window while I attached a new SoundView. This allows me to add the first SoundView (I only use one per window in my other application), but not subsequent SoundViews in my current application. Again, if anyone has any information regarding a possible work around (or better yet, source code for SoundView) please don't hesitate to contact me. Email preferred: Chris Penrose penrose@sfc.keio.ac.jp penrose@music.princeton.edu
From: j-ochs@nwu.edu (Joshua Ochs) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: Sun, 08 Jun 1997 20:44:48 -0600 Organization: Northwestern University, Evanston, IL, US Message-ID: <j-ochs-0806972044490001@wil123068.res-hall.nwu.edu> References: <338b6960.76768104@news.pdnt.com> <AFB060E7-7C68@207.173.163.78> <5mr0tm$opq$3@nntp.ucs.ubc.ca> It's in the details. When you order, you need your customer number, which you get as part of the developer program. The program is worth paying the $250 for - not only do you get teh developer mailing, you'll get MacOS 8, betas of 9, Rhapsody Dev release, Premeir release, and betas of Unified. The reason I only list betas of unified and Mac OS 9 is that your membership would run out in June 98 - just before they ship. It's worth it. On another note, for those that are in Apple Dev programs and ordered the OpenStep tools, did anyone get them? I ordered before they announced it to developers (thank you EvangeList), but never received a confirmation or anything. I am desperate to get started with this. - Joshua Ochs Diamond Software In article <5mr0tm$opq$3@nntp.ucs.ubc.ca>, gutier@unixg.ubc.ca (Gerald Gutierrez) wrote: >Let's see ... for someone to be able to have a look at Rapsody for "free", >and thus potentially gain Apple many more Rhapsody supporters who may >previously not have connections with Apple or even NeXT, that someone has >to be already registered with Apple's Developer Program. > >Either : > >1) being a member of Apple's Developer Program is free ( which I don't >believe so ), or > >2) the word "free" is being used incredibly loosely, or > >3) this is one big huge lie. > >: Are you a developer registered with Apple's Developer Program? Have you >: signed the Seed Agreement? If so: > >: >Additional information on the WWDC Prelude to Rhapsody package may be >: >found at > >: ><http://product.info.apple.com/pr/press.releases/1997/q3/970429.pr.rel.wwdc >: >..html>. To request your free OpenStep tools (Part Number R0724ZA) today, >: >contact us in one of the following ways: >: > >: >Phone >: >1-800-282-2732 Toll Free (US) >: >1-800-637-0029 Toll Free (Canada) >: >1-716-871-6555 International >: > >: >Email >: >order.adc@apple.com >: > >: >FAX >: >716-871-6511 >: > > >: So there... it is free! > >: Mike > > > >-- > > `'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`' > > Gerald Gutierrez Computer Engineering > Faculty of Applied Science > gutier@unixg.ubc.ca University of British Columbia
Message-ID: <33942496.52A0@gcomm.com> Date: Tue, 03 Jun 1997 16:05:10 +0200 From: ablasco@gcomm.com MIME-Version: 1.0 Subject: Re: gcc compile under OS4.1? References: <Pine.NXT.3.95.970526113529.4291D-100000@pollux.jura.uni-bonn.de> <5mcsdb$4su$1@gaea.omnigroup.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Newsgroups: comp.sys.next.programmer Timothy J. Wood wrote: > > There appears to be some interaction between the shared library support > added for 4.0 and gcc 2.7.2.x. One alternative would be to use the > GNUSource.pkg available from OS4.2. This contains a modified version of > 2.7.2.1. You may be able to apply the patches between 2.7.2.1 and 2.7.2.2 > to this version and get something that will build on OS 4.1. > > You shouldn't have to buy OS4.2 in order to get the GNUSource.pkg, > of course. I don't know if Apple makes it available on their web site or > if you'll just need to find someone with a copy that you can ftp. > > Konstantin Wiesel <kwiesel@mailhost.jura.uni-bonn.de> writes: > >I tried to compile Gcc 2.7.2.2 under OS4.1. It compile the stage1 compiler > >but compiling stage 2 with stage 1 fails. The getattr programm bus errors > >and make stops the compiling process. Does anybody had mor luck with this > >one? > > >Regards > >Konstantin Wiesel > >Email:kwiesel@mailhost.jura.uni-bonn.de > > -tim Sorry for this question, but why do you need this version of the compiler? What advantages has over the compiler that comes with my 3.3 dev version? Sorry for this stupid question, but I am virtually lost in the Next/Openstep world. P.S Where can I get that version of the compiler and Perl? TIA Amando Blasco
From: msb@plexare.com (Michael S. Barthelemy) Newsgroups: comp.sys.next.programmer Subject: Finding intersection point of two paths in PostScript Date: 9 Jun 1997 17:53:42 GMT Organization: Filoli Information Systems Message-ID: <5nhfv6$bgq$1@news.filoli.com> I need to be able to find where two paths intersect in PostScript. If the paths intersect multiple times, I need to be able to figure out which point is the first intersection given a point on one of the paths which will be considered the starting point of that path. Any pointers (no pun intended) on how to accomplish this would be greatly appreciated. Thanks, Mike Barthelemy
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5ndqol$nm$1161@news.internetmci.com> Date: 8 Jun 1997 16:51:32 GMT Control: cancel <5ndqol$nm$1161@news.internetmci.com> Message-ID: <cancel.5ndqol$nm$1161@news.internetmci.com> Sender: Photorep45@ibm.net Spam cancelled by sef@kithrup.com
From: mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> Newsgroups: comp.sys.next.programmer Subject: Re: Beginner question... Moveable text box? Date: 8 Jun 1997 12:28:34 GMT Organization: P&L Systems Message-ID: <5ne8hi$6tr$1@ironhorse.plsys.co.uk> References: <AFB9A84B-1A2FC@141.214.134.235> <5n4a2t$1sh5@castor.cca.rockwell.com> In-Reply-To: <5n4a2t$1sh5@castor.cca.rockwell.com> On 06/04/97, Erik M. Buck wrote: > This is a VERY bad user interface idea. I do not know of any > implementations like that. (Except in obsolete Windows versions!) > >What you are describing is multiple windows with scolling text for content. >Why limit the "extended NSScrollView ... (preferably with a title bar, >resize handle, and button to iconize it)" to the inside of another window ? > I agree that it sounds poor in UI terms.. it would be good if you could come up with a metaphor analagous to the (NeXT) Window UI rather than *exactly* the same, since this would give an inconsistent interface (Windows within Windows don't feature anywhere else); furthermore you may face a problem if you port to another version of OpenStep (say you want to deploy on Rhapsody/NT)... what UI do you use? Assuming you sort this out, though, rather than "extending" the NSScrollView, I'd have thought the better approach would be to create a new compound Object, based around a View; it will comprise a number of elements -- subviews -- including a title bar (which itself will have an iconise button), resize Buttons, and the NSScrollView itself..? Best wishes, mmalc. --
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer Subject: Impressed Date: 8 Jun 97 17:01:59 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFC0960B-4A57D@141.214.134.235> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit I've been programming using the Prelude for Rhapsody stuff for about two weeks now (actually less than that because a lot of time was spent just getting the stuff installed correctly). I went through the three sample programs that were in the books that came with the cds. This weekend I decided I would start working on something of my own. In two days I've created a subclass of NSBox that can be dragged around a window when grabbed on its edges, can be resized if grabbed in the lower right corner, and contains an NSTextView with a horizontal ruler and the attributes menu and scrollbars. I'm mirroring on NeXT a project we're doing in Java. In just a few hours work on the weekend I've done more than we've done in a month of Java work. (Hell, in Java 1.0.2 not even something as basic as scrolling is free!). 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: bozack@zobak.org (Daft.) Newsgroups: comp.sys.next.programmer Subject: compiling gcc 2.7.2.2 under openstep 4.1 Date: 9 Jun 1997 17:17:41 GMT Organization: The Black Box, Houston, Tx (713) 480-2686 Message-ID: <5nhdrl$dcj@news.blkbox.com> NNTP-Posting-User: bozack I've compiled and installed gcc 2.7.2.2 under openstep 4.1, but execution of any binary compiled with gcc results in a bus error. Is there anything special I should be doing when compiling/installing gcc? I get not so much as a warning during compilation. Thanks, Dan tform medical signal processing application -- this based on reading the next docs and absorbing the general hype about openstep development. When I returned home, my potential client supplied me with a PC that did not have SCSI etc, and used some non-standard sound boards. I purchased my own SCSI controller and disk, got the stuff installed with a minimum of fuss, and then started out on the Tutorials... again just a few glitches that were solved in this newsgroup. Now I've produced a demonstration package for my client that builds a signal (sin wave, square wave) dyamically in one view. You move sliders to change the amplitude, frequency, noise content, and phase of the waveform and the wave updates as you drag the sliders. This was to test the graphics and the results were impressive. The waveforms update very fast with no disturbing screen update garbage. I'm just using PSxxxx calls to do the graphics. I then connected a duplicate of that view to an FFT buffer and connected the waveform to the FFT controller. The second waveform view displays the FFT analysis also dynamically. I'm impressed with the speed of the code, the speed of coding, and the fact that during all the development I never -- never-- crashed the system even though I'm using pointers to data arrays! There is still a lot to learn but this seems to be a great system both from the point of view of the programmer, but also and more important from the user perspective. I love the display postscript screen envirionment. Looking forward to working with this package --and to the necessary support of this group. Rob Robert Norman Wave Master Software
From: matt@colorpar.com (Matt Gieselman) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: Mon, 09 Jun 1997 13:42:32 -0800 Organization: The Color Partnership Message-ID: <matt-0906971342320001@204.253.173.28> References: <338b6960.76768104@news.pdnt.com> <AFB060E7-7C68@207.173.163.78> <5mr0tm$opq$3@nntp.ucs.ubc.ca> <j-ochs-0806972044490001@wil123068.res-hall.nwu.edu> In article <j-ochs-0806972044490001@wil123068.res-hall.nwu.edu>, j-ochs@nwu.edu (Joshua Ochs) wrote: > On another note, for those that are in Apple Dev programs and ordered the > OpenStep tools, did anyone get them? I ordered before they announced it to > developers (thank you EvangeList), but never received a confirmation or > anything. I am desperate to get started with this. Joshua, We just got the bundle today, we did send in a request for it and we are on the Technology Seeding program. I don't know which got us the bundle. Regards, Matt -------------------------------------------------------------- Matt Gieselman Software Engineer and Juggler The Color Partnership matt@colorpar.com www.colorpar.com -------------------------------------------------------------
From: jon@clarke.exnext.com (Jonathan Hendry) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Followup-To: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Date: 9 Jun 1997 21:39:31 GMT Organization: Internet Access Cincinnati 513-887-8877 Message-ID: <5nht6j$6he$1@ocoee.iac.net> References: <338b6960.76768104@news.pdnt.com> <AFB060E7-7C68@207.173.163.78> <5mr0tm$opq$3@nntp.ucs.ubc.ca> <j-ochs-0806972044490001@wil123068.res-hall.nwu.edu> Joshua Ochs (j-ochs@nwu.edu) wrote: : It's in the details. When you order, you need your customer number, which : you get as part of the developer program. The program is worth paying the : $250 for - not only do you get teh developer mailing, you'll get MacOS 8, : betas of 9, Rhapsody Dev release, Premeir release, and betas of Unified. : The reason I only list betas of unified and Mac OS 9 is that your : membership would run out in June 98 - just before they ship. It's worth : it. Speaking of the developer program, I joined, but I've no idea what my identification number is. I received my big brown box o'stuff. No sign of an id #, as far as I could see. Any ideas? Thanks, Jon -- Jonathan W. Hendry jon@exnext.com
From: "Robert A. Decker" <comrade@umich.edu> Newsgroups: comp.sys.next.programmer Subject: Re: Beginner question... Moveable text box? Date: 8 Jun 97 22:45:12 -0400 Organization: University of Michigan ITD News Server Message-ID: <AFC0E682-F71C7@141.214.134.235> References: <5ne8hi$6tr$1@ironhorse.plsys.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: "mmalcolm crawford" <Malcolm_Crawford@plsys.co.uk> On Sun, Jun 8, 1997 8:28 AM, mmalcolm crawford <mailto:Malcolm_Crawford@plsys.co.uk> wrote: >I'd have thought the better approach would be to create a new >compound Object, based around a View; it will comprise a number of elements >-- subviews -- including a title bar (which itself will have an iconise >button), resize Buttons, and the NSScrollView itself..? So far I've come up with something simpler, but I love how it turned out. It's a subclass of NSBox, with the contentview set to an NSScrollview which contains a text view. I'm able to grab the box on any of its four edges and drag it around. I'm also able to grab it at the lower right corner to resize it (thereby resizing all the views it contains). The content of the box has scrollers, and one of those ruler dealies. It's excellent! (I still have to come up with the way I'll iconize it though). rob -- <mailto: "Robert A. Decker" comrade@umich.edu> Listen to my Realaudio playlist:<http://hmrl.cancer.med.umich.edu/Rob/index.ssi> Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #10: "Wow. I'm a genius too. I think. BEEP." -Chris Elliott
From: David Young <daver@jacobs.Geeks.ORG> Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: 9 Jun 1997 23:52:42 GMT Organization: Geeks Organizations Message-ID: <5ni50a$bju$2@darla.visi.com> References: <338b6960.76768104@news.pdnt.com> <AFB060E7-7C68@207.173.163.78> <5mr0tm$opq$3@nntp.ucs.ubc.ca> <j-ochs-0806972044490001@wil123068.res-hall.nwu.edu> <5nht6j$6he$1@ocoee.iac.net> NNTP-Posting-Date: 09 Jun 1997 18:52:42 CDT In comp.sys.next.programmer Jonathan Hendry <jon@clarke.exnext.com> wrote: > Speaking of the developer program, I joined, but I've no idea what my > identification number is. I received my big brown box o'stuff. No sign > of an id #, as far as I could see. Any ideas? How long did it take? I haven't gotten my stuff yet. -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 :: ut only to those who have signed the NDA. Plenty of time to get that done before it ships, though. -- Cliff ============= Maui no ka oi =============
From: eric@skatter.USask.Ca Newsgroups: comp.sys.next.programmer Subject: Appending to text object Date: 9 Jun 1997 19:42:38 GMT Organization: University of Saskatchewan Message-ID: <5nhmbe$42v@tribune.usask.ca> When debugging software I find it handy to have a scrollable text display to show diagnostic messages. Here's the method I'm now using: - (void)displayf:(NSString *)format, ... { va_list argList; NSString *text; va_start (argList, format); text = [[NSString alloc] initWithFormat:format arguments:argList]; [displayText replaceCharactersInRange:(NSRange){[[displayText string] length],0} withString:text]; [displayText scrollRangeToVisible:(NSRange){[[displayText string] length],0}]; [text release]; va_end (format); } This seems like an awful lot of effort, though. I'm also a litle concerned about the amount of work that has to get done by the two [[displayText string] length] calls, too. So, are there any Appkit/Foundatation kit experts out there that can suggest a better way of handling this? -- Eric Norum eric@skatter.usask.ca Saskatchewan Accelerator Laboratory Phone: (306) 966-6308 University of Saskatchewan FAX: (306) 966-6058 Saskatoon, Canada. NeXTMail accepted.
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: Re: Loading bundles in Windows NT Date: Mon, 09 Jun 1997 06:24:49 -0600 Organization: ErgoTech Message-ID: <339BF611.18FF@ergotech.com> References: <01bc71c6$5bd39ef0$77c4e5c2@tatra> <vb4tbbyewa.fsf@skyler.arc.ab.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Brian Schack wrote: > Does the bundle try to reference a variable or function declared in > the main program? If so, then you'll have to jump through some hoops > (courtesy of the WindowsNT linker). Basically, all variables and > functions that are need to be global to the bundles and the main > program should be put into a framework (read "DLL"). There's actually a linker option, which is memory serves correctly is something like "-ignore undefined" that will allow the bundle to be built and reference some things in the main program. I haven't exhaustively tested this, in my case the only reference I needed was the superclass of the object in the bundle. You _do_ still need to link with any frameworks. One side effect is that the bundle actually seems to load. On NT both frameworks and bundles are .dll files, which, I believe, are always shared. Does anyone know if bundles under Mach 4.0 are shared? Frameworks are almost useless on NT unless you know the classes in the frameworks before you start to run the application (i.e. you could have coded the whole thing in C++ without a real runtime). Since bundles are shared they seem an acceptable alternative. On Mach 4.0 frameworks dynamically load correctly, that is, you can get the principle class etc. etc. Bundles would be acceptable, if they are shared. I'd hate to have to build different stuctures on different operating systems. I still don't have a solution to the problem of DO not decoding and object passed by-copy when that object is linked to the application in a framework. There seems to be no mechanism to cause the framework to load other than hard coding a reference to something in the framework within the main application. Anyone have any more insight into this. Sharing a framework between a client and a server so that objects can be copied from one to the other would seem to be a fairly common design. Choosing the framework at run-time is a logical extension. Finally, the real strength of NS3.3 for ISV's was a great component architecture. The ability to load, and unload components _easily_ was a feature used by us and others in building applications that were sufficiently far ahead of the curve to stay in business. Proving that NeXT/Apple entirely missed this point and really hurt some of the few remaining ISV's on NS, the unloading mechanism is TOTALLY broken on 4.x. Some of this _may_ return in Rhaposody, eventually. At WWDC the demise of Lighthouse was lamented not because their products were necessarily better than the alternatives, but because they were open, component architecture. Many people had added features to Lighthouse applications that were unique to their business. It is Lighthouse's task to recreate this same functionality in Java, and that would appear to be, in part, is why Sun bought them. At the same time that Sun is busy promoting JavaBeans and an open component architecture, NeXT/Apple is busy breaking the architecture that they had. Is anyone else frustrated by this and the time waste that surrounds it? Jim
From: John Zachary <zachary@bit.csc.lsu.edu> Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: Mon, 09 Jun 1997 16:04:35 -0500 Organization: LSU Robotics Research Laboratory Message-ID: <339C6FE3.28D6@bit.csc.lsu.edu> References: <338b6960.76768104@news.pdnt.com> <AFB060E7-7C68@207.173.163.78> <5mr0tm$opq$3@nntp.ucs.ubc.ca> <j-ochs-0806972044490001@wil123068.res-hall.nwu.edu> <matt-0906971342320001@204.253.173.28> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Matt Gieselman wrote: > > In article <j-ochs-0806972044490001@wil123068.res-hall.nwu.edu>, > j-ochs@nwu.edu (Joshua Ochs) wrote: > > > > On another note, for those that are in Apple Dev programs and ordered the > > OpenStep tools, did anyone get them? I ordered before they announced it to > > developers (thank you EvangeList), but never received a confirmation or > > anything. I am desperate to get started with this. > > Joshua, > > We just got the bundle today, we did send in a request for it and > we are on the Technology Seeding program. I don't know which got > us the bundle. > > Regards, > Matt > If this is OpenStep 4.2 you are looking for, I ordered an edu copy last week (W). The Apple rep told me that OS4.2 was made available that day. Perhaps this is the reason for your delay. -- John Zachary LSU Robotics Research Laboratory
From: Susan Hovde <hovde@al.noaa.gov> Newsgroups: comp.sys.next.programmer Subject: kernel loader inactive - Rhapsody Date: Mon, 09 Jun 1997 14:21:58 -0600 Organization: NOAA Aeronomy Lab Message-ID: <339C65D0.32F9@al.noaa.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit After installing the first round of Prelude to Rhapsody installation, the instructions on the screen said to make sure there was no diskette in the floppy drive and press RETURN to continue. I ejected the drivers diskette and pressed RETURN. The system started to reboot. At the boot: prompt I typed -v and pressed RETURN. The screen now shows a window with the title "NeXT Mach Operating System" and a lot of lines referring to registering devices, checking disks, and using default tables. The last line is: kernel loader inactive, pausing What should I do now? This seems to indicate an error. This is the third time I have tried to install this system. The first time, I got it installed all the way, but when using the screen control panel, I somehow caused the screen to go blank and I could not figure out how to get it back, so I decided to reinstall. Then the system hung while copying files from the CD. This time, the copying went ok but now I'm stuck with this inactive kernel loader! ;-) Thanks for any help. Susan
From: jon@clarke.exnext.com (Jonathan Hendry) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Followup-To: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Date: 10 Jun 1997 00:05:49 GMT Organization: Internet Access Cincinnati 513-887-8877 Message-ID: <5ni5ot$81v$1@ocoee.iac.net> References: <338b6960.76768104@news.pdnt.com> <AFB060E7-7C68@207.173.163.78> <5mr0tm$opq$3@nntp.ucs.ubc.ca> <j-ochs-0806972044490001@wil123068.res-hall.nwu.edu> <5nht6j$6he$1@ocoee.iac.net> <5ni50a$bju$2@darla.visi.com> David Young (daver@jacobs.Geeks.ORG) wrote: : In comp.sys.next.programmer Jonathan Hendry <jon@clarke.exnext.com> wrote: : > Speaking of the developer program, I joined, but I've no idea what my : > identification number is. I received my big brown box o'stuff. No sign : > of an id #, as far as I could see. Any ideas? : How long did it take? I haven't gotten my stuff yet. The box came quite soon, sooner than I'd expected. I believe I mailed my application on May 12. (The check is dated May 11, which was a Sunday, so I assume I mailed the app. on Monday.) I got it a couple of weeks later. -- Jonathan W. Hendry jon@exnext.com
From: "Jurriaan van der Lingen" <jurriaan@fygir.nl> Newsgroups: comp.sys.next.programmer Subject: Re: Loading bundles in Windows NT Date: 10 Jun 1997 09:17:56 GMT Organization: XS4ALL, networking for the masses Sender: fygir@194.229.196.119 Message-ID: <01bc756e$67836d10$77c4e5c2@tatra> References: <01bc71c6$5bd39ef0$77c4e5c2@tatra> Jurriaan van der Lingen (me) <jurriaan@fygir.nl> wrote in article > Does anyone have experience with dynamic bundles in windows NT. > I have bundles included as "bproj" in the project. > This runs fine with dynamic bundles on mach (OPENSTEP 4.0) > Also it runs fine on Windows NT when I use the very same code in "subproj" > instead of "bproj" > > Help appreciated, Thanks! > Stephane Corthesy wrote: > In NeXTanswer ??? (I don't remember which one...), you can find the answer > to this problem with NT: you need to add the frameworks you use into the > bundles too, not only into the main project. This was NeXTanswer 2462, including these was necessary but not sufficient. Brian Scheck wrote: > Does the bundle try to reference a variable or function declared in > the main program? If so, then you'll have to jump through some hoops > (courtesy of the WindowsNT linker). Basically, all variables and > functions that are need to be global to the bundles and the main > program should be put into a framework (read "DLL"). That turned out to be the problem. However, instead of going through the bother of converting my app into a framework, I decided to wrap these functions and global variables into a class: The NT linker / DLL loader has no problem knitting classes together. For other people it may be interesting to know how to find out exactly which cross-calls you make in your bproj before you get the crash in some remote corner of the bundle that only end users seem to find. Jim Redman wrote: > There's actually a linker option, which is memory serves correctly is > something like "-ignore undefined" that will allow the bundle to be > built and reference some things in the main program. I haven't > exhaustively tested this, in my case the only reference I needed was the > superclass of the object in the bundle. You _do_ still need to link > with any frameworks. One side effect is that the bundle actually seems > to load. I looked up the option, which is "-undefined error" or "-undefined warning". The problem is that with this options the unresolved ObjC references prevent linking, while they are resolved correctly in run time. The output of the default linker option gives you clues. It says like: MyFile.o:error LNK2001: unresolved external symbol __MyGlobalFunction MyFile.o:error LNK2001: unresolved external symbol .objc_class_name_MyClass The class references, are no problem. The function references and global variables will crash. So with the linker output, I tracked down these globals, wrapped them in a class and my problem was solved. Thanks everybody! Jurriaan van der Lingen
Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc From: cdouty@netcom.com (Chris Douty) Subject: Re: Rhapsody features? Message-ID: <cdoutyEBJpJs.64@netcom.com> Organization: Netcom On-Line Services References: <5lvn4l$tje@src-news.pa.dec.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> Date: Tue, 10 Jun 1997 05:50:16 GMT Sender: cdouty@netcom6.netcom.com In article <5ndo6o$d4n@slip.net>, Emmett McLean <emclean@slip.net> wrote: >In article <338DB457.2579@abacus.com>, Jim Gagnon <jimg@abacus.com> wrote: >> >>Interesting. The long-time NeXT guys tell me that NeXTStep has real >>issues as a server OS. > > I don't believe it. Believe it. I love my cube (twin ND's) and intel box, but MachOS is not quite up to snuff as an application server. I don't know about hosting multiple users and shell activity, etc. but there are some definite bugs and limitations in the OS and object layer wrt network applications. There is a hard limit on the number of active TCP sockets, a limit run into by people serving http. It also affects DO traffic quite severely. DO uses two TCP sockets per object connection. (Well, at least two per machine are used. nmserver may multiplex its connections for all programs on a given machine.) The network and other I/O drivers are quite slow compared to Solaris or linux or BSD. I support a true DO application which brings down a random machine at least once a day. Now, much of this instability is probably due to poor program design, but a user level program does manage to panic a PPro 200 w/ 256MB of RAM running either NS 3.3 or OS 4.1. Yeah, I'd say there are some issues for NeXTstep as a server OS _today_. Rhapsody can be better and very likely will be. > Well, I'd take NS over NT anyday. Amen to that. -Chris -- Christopher Douty - Rogue Engineer trapped in a land of software cdouty@netcom.com "Frequently the messages have meaning; that is they refer to or are correlated according to some system with physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem." -Shannon
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: Appending to text object Date: 10 Jun 1997 05:34:08 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5nip0h$kgm$1@brachio.zrz.TU-Berlin.DE> References: <5nhmbe$42v@tribune.usask.ca> In article <5nhmbe$42v@tribune.usask.ca> eric@skatter.USask.Ca writes: > When debugging software I find it handy to have a scrollable text display to show > diagnostic messages. Here's the method I'm now using: What's wrong with 'fprintf( stderr , ... )'? You can either start the app manually and have the output go to the local terminal window, or view it in another terminal window with 'tail -f /tmp/console.log'. (The workspace console is too slow). Marcel
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: Re: Loading bundles in Windows NT Date: Tue, 10 Jun 1997 06:11:06 -0600 Organization: ErgoTech Message-ID: <339D445A.2F6D@ergotech.com> References: <01bc71c6$5bd39ef0$77c4e5c2@tatra> <01bc756e$67836d10$77c4e5c2@tatra> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Jurriaan van der Lingen wrote: > > Jim Redman wrote: > > There's actually a linker option, which is memory serves correctly is > > something like "-ignore undefined" that will allow the bundle to be > > built and reference some things in the main program. I haven't > > exhaustively tested this, in my case the only reference I needed was the > > superclass of the object in the bundle. You _do_ still need to link > > with any frameworks. One side effect is that the bundle actually seems > > to load. > > I looked up the option, which is "-undefined error" or "-undefined > warning". The problem is that with this options the unresolved ObjC > references prevent linking, while they are resolved correctly in run time. > > The output of the default linker option gives you clues. It says like: > MyFile.o:error LNK2001: unresolved external symbol __MyGlobalFunction > MyFile.o:error LNK2001: unresolved external symbol .objc_class_name_MyClass > > The class references, are no problem. The function references and global > variables will crash. So with the linker output, I tracked down these > globals, wrapped them in a class and my problem was solved. > My apologies, I should do my homework before posting. The option I've been using is "-undefined suppress". This allows you to build the executable with the superclass of the class (or presumably classes) of the objects in the bundle undefined and a part of the main application, not a framework. I agree with you that this solution is limited to this one specific case. I've just tested variables and functions with "supress" and one or the other, or both doesn't work, that is the application will crash. You can still reference variables and, I assume, functions loaded in Frameworks so long as you link against the framework. Jim
From: Photorep45@ibm.net Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5ndqol$nm$1161@news.internetmci.com> Control: cancel <5ndqol$nm$1161@news.internetmci.com> Message-ID: <despam.5ndqol$nm$1161@news.internetmci.com> Date: Sun, 8 Jun 1997 06:00:00 -700 Cancel Spam.
From: Greg_Anderson@afs.com (Gregory H. Anderson) Newsgroups: comp.sys.next.programmer Subject: Re: Impressed Date: 10 Jun 1997 15:07:19 GMT Organization: Anderson Financial Systems Inc. Message-ID: <5njqj7$jqi@shelob.afs.com> References: <5nhfrv$c22@argentina.earthlink.net> "Robert Norman" <rob@neurodata.com> writes > > Now I've produced a demonstration package for my client that builds a > signal (sin wave, square wave) dyamically in one view. You move sliders > to change the amplitude, frequency, noise content, and phase of the > waveform and the wave updates as you drag the sliders. This was to test > the graphics and the results were impressive. The waveforms update very > fast with no disturbing screen update garbage. I'm just using PSxxxx > calls to do the graphics. Yes, but can you draw 3D perspective text along the waves in six different transfer modes and three different languages with different orientations? -- Gregory H. Anderson | "We're in the land of the blind, Visionary Ophthalmologist | selling working eyeballs, and they Anderson Financial Systems | balk at the choice of color." -- Tony greg@afs.com (NeXTmail OK) | Lovell, on Mac user reactions to NeXT PS: Ain't I a stinker? 8^) 8^) 8^)
From: "Travis Griggs" <tgriggs@keyww.com> Newsgroups: comp.sys.next.programmer Subject: Physical Memory Driver Date: 9 Jun 1997 17:37:46 GMT Organization: All USENET -- http://www.Supernews.com Message-ID: <01bc74fb$ac4de290$74c728c0@tgriggs4> Does NeXT/Rhapsody support/have a physical memory driver? In days gone by, we could (using HP-UX) lock down a chunk of kernel memory and read directly from it. This makes for quick transfers from external devices (e.g. machine vision cameras :) ). We have been *sorely* dissapointed with NT in this regard because it will not allow this. It does a buffered copy (i.e. read, write, read) from kernel space to user space, unless one wants to fork over huge dough for a HAL development kit. Given such, we have had to compromise by only getting a sampling of data. Thanks... -- Travis Griggs Key Technology
From: ians@cam-ani.co.uk (Ian Stephenson) Newsgroups: comp.sys.next.programmer Subject: Re: Physical Memory Driver Date: Tue, 10 Jun 1997 12:25:33 GMT Organization: Cambridge Animation Systems Ltd Sender: news@cam-ani.co.uk Message-ID: <EBK7uM.4pn@cam-ani.co.uk> References: <01bc74fb$ac4de290$74c728c0@tgriggs4> In article <01bc74fb$ac4de290$74c728c0@tgriggs4> "Travis Griggs" <tgriggs@keyww.com> writes: > Does NeXT/Rhapsody support/have a physical memory driver? In days gone by, > we could (using HP-UX) lock down a chunk of kernel memory and read directly > from it. This makes for quick transfers from external devices (e.g. machine > vision cameras :) ). We have been *sorely* dissapointed with NT in this > regard because it will not allow this. It does a buffered copy There are pretty good facilities for doing this... Basically (to loop over the buzzwords):You would need to write a suitable driver for your hardware using Driver kit. This uses MiG to generate a user api. Mach messaging is used for the transfer of data, such that (if you're carefull) you get a copy on write version of the data - ie there's no overhead to the copy, but you can't change the kernel version (best of both worlds). One problem I have had doing this is that there are problems allocating LARGE blocks of memory inside the kernel. kalloc CAN fail (depite what the manual tells you) - you just need to push it hard enough (about 30Meg is enough). Alocating such big chunks sounds silly, but if you've got a fast device delivering Hires images it's about the only option! (workaround is to wire the memory at boot time! The machines have 256Meg anyway, so they're not going to miss it) $an
From: "Mathias M. Lang" <bebeto@wg.saar.de> Newsgroups: comp.sys.next.programmer Subject: bootproblems Date: Tue, 10 Jun 1997 14:40:39 +0200 Organization: Yoyodyne Posting Systems, INN Lab. Message-ID: <339D4B47.41C6@wg.saar.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, my cube want not boot, and gives me the messagees: localhost: loginwindow: could not find WindowServer port localhost: kern_loader: server fbw running up waiting for executable /mach localhost: kern_loader: server midi running up waiting for executable /mach and so on. can anybody helps me bebeto
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: 10 Jun 1997 17:55:01 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5nk4dl$q7d$1@owl> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> In comp.sys.next.programmer Jim Gagnon <jimg@abacus.com> wrote: > John Christie wrote: > Interesting. The long-time NeXT guys tell me that NeXTStep has real > issues as a server OS. Apparently, if you put more than twelve-or-so > users on it at a time, it really bogs down. Also, there's a limit of > 200 processes in NeXTStep, which is actually pretty low for even a > client-side OS. To top it off, a NeXTStep system really needs the > protection of a firewall -- security is definitely behind the levels > defined by Solaris and AIX. I think your info may be based on 6-8 year old NeXT black hardware running an equally old version of the OS. Steve
From: logi27@imaginet.fr (Logi27 Development Team) Newsgroups: comp.sys.next.programmer Subject: Openstep and Sofware installation (installer) Date: Tue, 10 Jun 1997 18:00:54 +0200 Organization: LOGI 27 Message-ID: <logi27-1006971800540001@cyber52.montpellier.imaginet.fr> Hello everybody, I am new to OpenSetp world (comming from mac one ! suprising isn't it ?) My Concern is about software installation. Is there any software installer in the Openstep tool chest ? And if yes how it is ? Does it provide a tool to make its configuration easier ? Can it be used for our sofware installation ? What about the insatllation above Windows NT/95 ? (in the same mind) thanks a lot, a Macintosh Developper a+ Paul Plaquette -- LOGi 27 a Small Macintosh First Deveopment Compagny email : LOGi27@Imaginet.fr
From: ians@cam-ani.co.uk (Ian Stephenson) Newsgroups: comp.sys.next.programmer Subject: Re: Openstep and Sofware installation (installer) Date: Tue, 10 Jun 1997 17:06:12 GMT Organization: Cambridge Animation Systems Ltd Sender: news@cam-ani.co.uk Message-ID: <EBKKuE.5oL@cam-ani.co.uk> References: <logi27-1006971800540001@cyber52.montpellier.imaginet.fr> In article <logi27-1006971800540001@cyber52.montpellier.imaginet.fr> logi27@imaginet.fr (Logi27 Development Team) writes: > Is there any software installer in the Openstep tool chest ? /NextAdmin/Installer.app. > And if yes how it is ? It's used universally by all commercial software, and has been around for five years in a stable from. > Does it provide a tool to make its configuration easier ? > Can it be used for our sofware installation ? Yes, though how you set it up is not widly known. I did it a few yeras back, but don't remember the detail. There are commandline tools hidden inside the app to generate packages. Interestingly I also discovered, it has a mode to install via ftp. No idea how to trigger it though. > What about the insatllation above Windows NT/95 ? (in the same mind) Same app. Haven't used it though. $an
From: msoori@*genetics.bio-rad.com (msoori) Newsgroups: comp.sys.next.hardware,comp.sys.next.programmer,comp.sys.next.bugs Subject: Monitor problems with NeXT Date: Tue, 10 Jun 1997 09:56:50 +0100 Organization: Bio-Rad Laboratories Message-ID: <msoori-1006970956500001@ms.genetics.bio-rad.com> I am new to OpenStep from the Mac... I have installed Prelude to Rapsody on an intel box with and using a generic s3 video card. It works fine on my 21" Apple multi sync monitor and another pc monitor, but it dosent work on the radius Two Page Display/21gs gray scale monitor (which is fixed resolution at 1152x870 at 75hz) I cant find any display modes on the Next that displays at this resolution. The closest one listed is 1152x864. I'd like to use the color monitor for work and keep the gray scale one hooked to the next box so I can learn about OpenStep. What gives? Please respond by e-mail. Thanks, Mahesh. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ No spam for me mam! Remove * from e-mail address to reply. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ Mahesh P. Sooriarachchi. ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ Work: msoori@*genetics.bio-rad.com | ~ ~ Personal: mahesh@*value.net | This space for rent! ~ ~ Home Page: http://value.net/~mahesh/work.html | ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
From: *johnnyc*@or.psychology.dal.ca (John Christie) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: Tue, 10 Jun 1997 21:30:31 -0300 Organization: ISINet, Nova Scotia Message-ID: <*johnnyc*-1006972130320001@192.0.2.1> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5nk4dl$q7d$1@owl> In article <5nk4dl$q7d$1@owl>, Steve Dekorte <dekorte@slip.net> wrote: > In comp.sys.next.programmer Jim Gagnon <jimg@abacus.com> wrote: > > John Christie wrote: > > Interesting. The long-time NeXT guys tell me that NeXTStep has real > > issues as a server OS. Apparently, if you put more than twelve-or-so > > users on it at a time, it really bogs down. Also, there's a limit of > > 200 processes in NeXTStep, which is actually pretty low for even a > > client-side OS. To top it off, a NeXTStep system really needs the > > protection of a firewall -- security is definitely behind the levels > > defined by Solaris and AIX. > > I think your info may be based on 6-8 year old NeXT black hardware running > an equally old version of the OS. > > Steve I never wrote any of this. Please quote correctly. thankyou -- You aren't free if you CAN choose - only if you DO choose. All you are is the decisions you make. Remove "*" and ohnny (i.e. jc@) to reply via email
From: REMOVE.TO.REPLY.drg@biomath.mdacc.tmc.edu (David Gutierrez) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: Tue, 10 Jun 1997 23:43:57 -0500 Organization: Univ. Texas M.D. Anderson Cancer Center Message-ID: <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> References: <5lvn4l$tje@src-news.pa.dec.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> Mime-Version: 1.0 Content-Type: text/plain;charset=US-ASCII Content-Transfer-Encoding: 7bit In article <cdoutyEBJpJs.64@netcom.com>, cdouty@netcom.com (Chris Douty) wrote: > Yeah, I'd say there are some issues for NeXTstep as a server OS _today_. > Rhapsody can be better and very likely will be. They're certainly going to try. Plans are to eventually port a server version of Rhapsody to the Apple Network Server 500 and 700 (the big AIX boxes), as well as offering server configurations in the cases that succeed the 9600/8600. -- David Gutierrez REMOVE.TO.REPLY.drg@biomath.mdacc.tmc.edu Remove everything before the "drg" to send mail to me. Some junk e-mailers scan .sig files, as well as From: lines, to get addresses. "Only fools are positive." - Moe Howard
From: Photoagent@ibm.net Newsgroups: comp.sys.next.programmer Subject: Attention All Photographers Date: Wed, 11 Jun 1997 02:37:55 Organization: Internet MCI Message-ID: <5nlh06$eqc$1104@news.internetmci.com> PHOTO PHOENIX INTERNATIONAL We are an international New York based agency representing professional and amateur photographers in the following areas: Galleries, Business and Organizations, Book Publishers, Paper Products, Consumer Publication, Newspapers & Newsletters Publication, Special Interest Publication, and Trade Publication. The images we are seeking are some of the following: Animals/Architecture/Art/Automobiles/Business/Celebrities/Documentary/ Entertainment/Erotica-Nudity/Fashion/Food/Health/History/Hobbies/ Outdoors/People/Political/Portraits/Still Lifes/Religious/Science/ Sports/Travel. Whether your work portrays conservative, experimental, stylish, or innovative themes, your project proposal should be well thought out before submitting to us. For first contact, please submit a query letter, and samples of your work. Please include SASE. Do not send entire portfolio unless we ask for it. If we are not interested, you may not hear from us because of time limitations, so please submit what does not need to be returned. If we think your work is sellable we will respond as quickly as possible. Send to: Photo Phoenix International 33-29 58 Street Woodside, New York 11377 Tele: Florida Branch-(941) 642-660
From: michael@ninebits.com (Michael Balle) Newsgroups: comp.sys.next.programmer Subject: DO NT/MACH Date: Wed, 11 Jun 1997 09:37:08 +0200 Organization: Nine Bits Message-ID: <199706110937083521842@[222.223.224.4]> Hi, can distributed objects communicate across platform, ie. can objects in an OPENSTEP application running on WinNT communicate with and OPENSTEP application running on a MACH kernel?. The reason I ask is that the documentation mentions that MACH kernel messages are used for PDO communication. regards Michael
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5nlh06$eqc$1104@news.internetmci.com> Date: 11 Jun 1997 07:00:11 GMT Control: cancel <5nlh06$eqc$1104@news.internetmci.com> Message-ID: <cancel.5nlh06$eqc$1104@news.internetmci.com> Sender: Photoagent@ibm.net Spam cancelled by sef@kithrup.com
Control: cancel <339D445A.2F6D@ergotech.com> Newsgroups: comp.sys.next.programmer From: jerry@icgned.nl (Jerry Martin van der Duim) Subject: cancel Message-ID: <EBLnFK.Eu8@icgned.nl> Sender: news@icgned.nl Organization: IC Group Date: Wed, 11 Jun 1997 06:59:44 GMT cancel
From: Christian Neuss <neuss@informatik.th-darmstadt.de.NOSPAM> Newsgroups: comp.sys.next.programmer Subject: Re: Impressed Date: 11 Jun 1997 08:56:20 GMT Organization: Technische Hochschule Darmstadt Message-ID: <5nlp7k$9ci$1@news.th-darmstadt.de> References: <5nhfrv$c22@argentina.earthlink.net> <5njqj7$jqi@shelob.afs.com> Greg_Anderson@afs.com (Gregory H. Anderson) wrote: >Yes, but can you draw 3D perspective text along the waves in six different >transfer modes and three different languages with different orientations? ROTFL :-) Chris -- // Christian Neuss "static typing? how quaint.." // http://www.nexttoyou.de/~neuss/ // fax: (+49) 6151 16 5472
From: j-norstad@nwu.edu (John Norstad) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: Wed, 11 Jun 1997 08:17:05 -0600 Organization: Northwestern University Message-ID: <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> References: <5lvn4l$tje@src-news.pa.dec.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> In article <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu>, REMOVE.TO.REPLY.drg@biomath.mdacc.tmc.edu (David Gutierrez) wrote: > In article <cdoutyEBJpJs.64@netcom.com>, cdouty@netcom.com (Chris Douty) wrote: > > > Yeah, I'd say there are some issues for NeXTstep as a server OS _today_. > > Rhapsody can be better and very likely will be. > > They're certainly going to try. Plans are to eventually port a server > version of Rhapsody to the Apple Network Server 500 and 700 (the big AIX > boxes), as well as offering server configurations in the cases that > succeed the 9600/8600. I understand that Apple has a server engineering team devoted to solving the problems in NeXTStep for server use. There will be separate server and workstation versions of Rhapsody, like Windows NT. Servers are definitely one of their target markets for the initial versions of Rhapsody. For example, the server version will support spanning, striping, and mirroring file systems, and the core OS will be tuned for server use. -- John Norstad <mailto:j-norstad@nwu.edu> <http://charlotte.acns.nwu.edu/jln/>
From: Patrick Schulz <pschulz@symphony.amd.com> Newsgroups: comp.sys.next.programmer Subject: Re: DO NT/MACH Date: Wed, 11 Jun 1997 08:42:26 -0500 Organization: AMD Fab30 Message-ID: <339EAB42.1664@symphony.amd.com> References: <199706110937083521842@[222.223.224.4]> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi Michael, Michael Balle wrote: > can distributed objects communicate across platform, ie. can objects in > an OPENSTEP application running on WinNT communicate with and OPENSTEP > application running on a MACH kernel?. The answer is: YES. All OpenStep compliant DO implementations can communicate to each other. That includes OPENSTEP on Mach, NT, PDO for Solaris/HP-sUX AND even Sun's OpenStep Implementation (but not the GNUstep DO system [not OS compliant yet]). The Mach messages system is emulated on non Mach platforms. So it works the same way, whatever operating system you use. I recall some articles in german magazines about DO/PDO but they are in german ;-) hope that helps, Patrick. -- Patrick Schulz; 1704 Nelms Dr. #2025; Austin, TX 78744 email: pschulz@symphony.amd.com (MIME welcome) - Microsoft is not the answer, Microsoft is the question and the answer is: NO.
From: embuck@palmer.cca.rockwell.com (Erik M. Buck) Newsgroups: comp.sys.next.programmer Subject: Re: Hair Pulling with my first stab at IB! (please help :-)) Date: 11 Jun 1997 14:41:41 GMT Organization: Rockwell Collins Message-ID: <5nmdf5$b4a2@castor.cca.rockwell.com> References: <5nev5t$bqg$1@news.sas.ab.ca> Cc: jroepcke@compusmart.ab.ca In an vain attempt to play to the masses, NeXT removed the return sign paradigm and replaced it with the windows like "black bordered" default selection indication. Just proceed without the return sign.
From: "Timothy J. Luoma" <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: compiling new syslog Date: Wed, 11 Jun 1997 11:11:11 -0400 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.NXT.3.96.970611110704.19651A-100000@cc344191-a> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: Jacques Distler <distler@golem.ph.utexas.edu> I got the new syslog from: http://golem.ph.utexas.edu/~distler/progs/newlog-1.0.4.tar.gz and tried to make it (it is supposedly setup for NeXT already) however when I tried to compile, I got this: cc -g -O -I. -DBad_float_h -arch m68k -arch hppa -arch i386 -arch sparc -c strtod.c cc -g -O -I. -DBad_float_h -arch m68k -arch hppa -arch i386 -arch sparc -c vfprintf.c libtool -o libnewlog.a snprintf.o vsnprintf.o syslog.o fvwrite.o strerror.o strtod.o vfprintf.o ranlib libnewlog.a || true -o syslogerr syslogerr.c Make: Cannot load -o. Stop. *** Exit 1 Stop. Now I think that line that begins ' -o' is supposed to have something there before it, but by some mistake it is an empty variable. Can anyone explain this or how to fix? TjL BTW: isn't it true that -arch m68k should be the first of the -archs, for some backwards compatibility reason I can't quite recall (probably outdated, I think it was something to do with 3.0 or 3.1....) -- TjL <luomat@peak.org> / http://www.peak.org/~luomat/next/ NeXT bookmarks: http://www.peak.org/~luomat/next/bookmarks.html sed 's/End\ of\ sig/pithy\ quotation/g'
From: "Timothy J. Luoma" <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: NS3.3: gcc 2.7.2.2 error?? Date: Wed, 11 Jun 1997 12:01:42 -0400 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.NXT.3.96.970611115847.24622A-100000@cc344191-a> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I just installed gcc yesterday, so I am trying to compile a lot of software today.... During said efforts, I came across this In file included from inet.c:58: /usr/local/lib/gcc-lib/i386-next-nextstep3/2.7.2.2/include/bsd/memory.h:22: unterminated `#else' conditional (when making libpcap-0.2.1) This looks like a generally bad error, one that might effect more than just this compilation attempt.... Clues/explanations appreciated... TjL -- TjL <luomat@peak.org> / http://www.peak.org/~luomat/next/ NeXT bookmarks: http://www.peak.org/~luomat/next/bookmarks.html sed 's/End\ of\ sig/pithy\ quotation/g'
From: tph@longhorn.uucp (Tom Harrington) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Followup-To: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Date: 11 Jun 1997 15:18:21 GMT Organization: Mechanist Industries Message-ID: <5nmfjt$9423@eccws1.dearborn.ford.com> References: <5lvn4l$tje@src-news.pa.dec.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> John Norstad (j-norstad@nwu.edu) wrote: : example, the server version will support spanning, striping, and mirroring You know you've been on Usenet too long when you misread the above as "...will support spamming...". -- Tom Harrington ------- tph@rmii.com ------- http://rainbow.rmii.com/~tph "Somebody shoot me!" -Animaniacs -> Fractal Kit: http://rainbow.rmii.com/~tph/fractalkit/fractal.html <-
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: DO NT/MACH Date: Wed, 11 Jun 1997 13:23:59 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <cnbhwjm00iV_03V_A0@andrew.cmu.edu> References: <199706110937083521842@[222.223.224.4]> Excerpts from netnews.comp.sys.next.programmer: 11-Jun-97 DO NT/MACH by Michael Balle@ninebits.c > can distributed objects communicate across platform, ie. can objects in > an OPENSTEP application running on WinNT communicate with and OPENSTEP > application running on a MACH kernel? Yes. (P)DO and OpenStep distributed object messaging works across platforms including non-Mach systems because those other operating systems run a process named 'machd' (or some similar name). 'Machd' handles Mach messages for such operating systems in lieu of the Mach kernel and 'nmserver', which is the network message server, which is "responsible for forwarding Mach IPC over the network". -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
Date: Wed, 11 Jun 1997 13:36:21 -0600 From: beauvois@usa.net Subject: nib/InterfaceBuilder.h Newsgroups: comp.sys.next.programmer Message-ID: <866053275.20456@dejanews.com> Organization: Deja News Usenet Posting Service I'm trying to port some code from NS 2.0/1 to 3.3 but can't figure out what to replace the old <nib/InterfaceBuilder.h> header file with. Anyone know ? -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: schwarz@physik.tu-berlin.de (Georg Schwarz) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: Wed, 11 Jun 1997 21:23:35 +0200 Organization: Institut f. Theoretische Physik, TU Berlin Message-ID: <19970611212335461585@marconi.physik.tu-berlin.de> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit John Norstad <j-norstad@nwu.edu> wrote: > I understand that Apple has a server engineering team devoted to solving > the problems in NeXTStep for server use. There will be separate server and > workstation versions of Rhapsody, like Windows NT. Servers are definitely > one of their target markets for the initial versions of Rhapsody. For > example, the server version will support spanning, striping, and mirroring > file systems, and the core OS will be tuned for server use. I suscept that the "core OS" will the identical for both the "server" and "desktop" version though, won't it? -- Georg Schwarz schwarz@physik.tu-berlin.de, kuroi@cs.tu-berlin.de Institut für Theoretische Physik +49 30 314-24254, FAX -21130 Technische Universität Berlin http://home.pages.de/~schwarz/
From: rbarris@quicksilver.com (Rob Barris) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.programmer Subject: Prelude2Rhapsody on black hardware! Date: Wed, 11 Jun 1997 11:49:27 -0700 Organization: Quicksilver Software, Inc. Message-ID: <rbarris-ya023280001106971149270001@news.intelenet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit I finally got around to trying the Prelude OpenStep 4.2 CDROM on my 68040 NextStation Turbo. After mounting the disc, running Upgrader.app and then clearing off some disk space by its advice, the install worked! After having heard from any number of people "nahhh, it's an Intel only release" it was a pleasant surprise indeed to see that the discs from WWDC were actually fat builds for both Intel and "black" NeXT hardware. (So far I've only installed the User side FWIW) However I only have a 240MB HD in that machine, and now it's full. Looks like I need to swap in a 1GB drive or something before I continue with installing the "Developer" disc. Thanks guys! Rob Barris Quicksilver Software Inc. rbarris@quicksilver.com * Opinions expressed not necessarily those of my employer *
From: goldwass@lifesci.lscf.ucsb.edu (Lloyd Goldwasser) Newsgroups: comp.sys.next.programmer Subject: Re: Problems compiling after OpenStep conversion Date: 11 Jun 1997 00:27:56 GMT Organization: University of California, Santa Barbara Message-ID: <5nkred$llc@ucsbuxb.ucsb.edu> References: <5n7cjb$osr@ucsbuxb.ucsb.edu> <5nae2h$a1@mpaque.mpaque> mpaque@wco.com (Mike Paquette) wrote: > (Lloyd Goldwasser) writes: > > [problems] > [solutions] Thanks, Mike. For a duffer like me, it's an honor to have a problem that you help solve... Now I'm pulling my hair out over two other problems that have arisen since converting. First is yet another header problem. When I try to compile, I get link errors due to undefined symbols: /bin/ld: Undefined symbols: .objc_class_name_NSArray .objc_class_name_NSBundle ... although, for instance, every file that has NSArrays definitely #imports <Foundation/NSArray.h>. The list of undefined symbols encompasses even things like _SNDAlloc, although I never go near anything to do with sound. (1) Why should it have trouble with Arrays, which I think I'm treating properly; and (2) Why is it having trouble with stuff that I'm not touching? I suppose that I'm #importing something that #imports something that #imports this stuff, but I'm not doing anything that I didn't do before converting to 4.1. Second, the NAN problem that I mentioned still persists. NAN used to work fine with 3.2; now the compiler gives me ediv invalid operation error although I've #imported <math.h> (and tried whatever variants I can think of). The #define in math.h _looks_ the same; why should it be behaving differently? (I notice also that the sparc header file contains a comment that this #define needs to be fixed. I presume that it shouldn't be a problem for me: I'm compiling for black and sparc, but I'm doing so on black. Or _do_ I need to worry?) Thanks, Lloyd Goldwasser goldwass@lifesci.lscf.ucsb.edu
From: Dru Nelson <dnelson@slip.net> Newsgroups: comp.sys.next.programmer Subject: NSDictionary grammar? Date: 11 Jun 1997 20:52:04 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5nn35k$hmq$1@owl> What is the grammar for an OpenStep NSDictionary object? (or where is the documentation) Are the initFromFile:/writeToFile:atomically: methods the best way of storing little configurations in an ascii format or is there a better serializer/persistence methodology? I think Project Builder uses the same method in it's PB.project file. Dru Nelson - dnelson@slip.net - Redwood City, California
From: decoy_id@no_junk_on_the.net (L e e Altenberg) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: 12 Jun 1997 03:46:36 GMT Organization: MHPCC Message-ID: <5nnres$390$2@kaopala.mhpcc.edu> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> <19970611212335461585@marconi.physik.tu-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: schwarz@physik.tu-berlin.de In <19970611212335461585@marconi.physik.tu-berlin.de> Georg Schwarz wrote: > John Norstad <j-norstad@nwu.edu> wrote: > > > > I understand that Apple has a server engineering team devoted to solving > > the problems in NeXTStep for server use. There will be separate server and > > workstation versions of Rhapsody, like Windows NT. Servers are definitely > > one of their target markets for the initial versions of Rhapsody. For > > example, the server version will support spanning, striping, and mirroring > > file systems, and the core OS will be tuned for server use. > > I suscept that the "core OS" will the identical for both the "server" > and "desktop" version though, won't it? > Could anyone explain why an OS optimized for use as a server would not also be optimal for a workstation? -- ======================================================================= Lee Altenberg, Ph.D. Research Affiliate, University of Hawai`i at Manoa Office: Maui High Performance Computing Center 550 Lipoa Parkway, Suite 100, Kihei, Maui HI 96753 Phone: (808) 879-5077 x 296 (work), (808) 879-5018 (fax) E-mail: altenXber@mhpXcc.edu <Delete the "X"s; done to stop junk e-mail> Web: http://pueo.mhpcc.edu/~Xaltenber/ <Delete the "X"> =======================================================================
Newsgroups: comp.sys.next.programmer From: fgalot@x-lan.alienor.fr Subject: [Q]Using devices Message-ID: <EBMHKE.B00@x-lan.alienor.fr> Sender: news@x-lan.alienor.fr Organization: x&lan Date: Wed, 11 Jun 1997 17:50:38 GMT -Which device (/dev/????) should I use to communicate whith my serial port? -How configuring a device? -How configuring a port for lauching an executable automatically. (example : a user connect himself on a port via a modem et man launch a BBS program). Thanks for help. -- --------------------------------------- ® ® | ® O_O ® ® | O_O -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- Fred Galot fgalot@x-lan.alienor.fr
From: "Quang Ngo" <quang@calwest.net> Newsgroups: comp.sys.next.misc,comp.sys.next.programmer Subject: Install NeXTSTEP 3.3 on a Pentium 90Mhz w/o Floppy Date: 12 Jun 1997 04:42:28 GMT Organization: Call America Internet Services +1 (800) 563-3271 Message-ID: <01bc76ea$c1a56c80$f7663cd1@jaguar> Something went wrong with my motherboard - everything works fine except the floppy drive. The floppy drive and controller card work fine on a different system. Anyway, is it possible to install NeXTSTEP 3.3 without a floppy drive? The image files (boot and drivers) are on www.next.com. But looks like they're no help without a floppy drive. Any ideas? -Quang
Newsgroups: comp.sys.next.programmer From: dfevans@bbcr.uwaterloo.ca (David Evans) Subject: Re: Openstep and Sofware installation (installer) Sender: news@novice.uwaterloo.ca (Mr. News) Message-ID: <EBMCvw.FMH@novice.uwaterloo.ca> Date: Wed, 11 Jun 1997 16:09:32 GMT References: <logi27-1006971800540001@cyber52.montpellier.imaginet.fr> <EBKKuE.5oL@cam-ani.co.uk> Organization: University of Waterloo In article <EBKKuE.5oL@cam-ani.co.uk>, Ian Stephenson <ians@cam-ani.co.uk> wrote: >In article <logi27-1006971800540001@cyber52.montpellier.imaginet.fr> >logi27@imaginet.fr (Logi27 Development Team) writes: > >> Does it provide a tool to make its configuration easier ? >> Can it be used for our sofware installation ? >Yes, though how you set it up is not widly known. I did it a few yeras >back, but don't remember the detail. There are commandline tools hidden >inside the app to generate packages. > There's something called "PackageBuilder" that should be on Peak, Peanuts, etc. It was for the 3.x Installer but it might work now, too. At least you could contact the author and see about getting the source. >Interestingly I also discovered, it has a mode to install via ftp. No idea >how to trigger it though. > I think the PackageBuilder gizmo can handle this. -- David Evans (NeXTMail/MIME OK) dfevans@bbcr.uwaterloo.ca Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/ University of Waterloo "Default is the value selected by the composer Ontario, Canada overridden by your command." - Roland TR-707 Manual
From: dirk@object-factory.com (Dirk Olmes) Newsgroups: comp.sys.next.programmer Subject: Re: Hair Pulling with my first stab at IB! (please help :-)) Date: 9 Jun 1997 06:03:10 GMT Organization: Object Factory GmbH (Germany) Message-ID: <5ng6au$acj$1@leonie.object-factory.com> References: <5nev5t$bqg$1@news.sas.ab.ca> jroepcke@compusmart.ab.ca wrote: > Hi Folks, Hi, > Well friends, I don't have a NSReturnSign icon in the > images display! (OpenStep 4.1 Intel, User+Dev+EOF 2.0) ... well, this is a known "bug" in the new developer releases. > This sucks! ;-) Where is it? I looked everywhere, and > I can't find it... if I load the project > /NextDeveloper/Examples/AppKit/CurrencyConverter/PB.project > (or whatever it's called) and open the nib file, the > button in the window shows the NSReturnSign icon, and the > button's inspector lists it, but the images display > doesn't have the icon in it! (I've tried resizing and > scrolling the window -- it's not there...) No, it is definitely missing. Well, not really. Here's what I found some months ago: ------------------------ schnipp --------------------------------------- Hi, the NSReturnSign seems to have been lost in OpenStep for Mach 4.1 (It was in fact lost in the 4.0 beta but I assumed it would find it's way home in the sharp release). I have been able to locate the image in the file /usr/lib/NextStep/Resources/SharedGray.tiff, but the problem lies within InterfaceBuilder, namely the file InterfaceBuilder.app/Resources/images.table where the NSReturnSign entry is missing. Just add the entry and everything works as it should. Just our two cents, Urban and Malte --- Urban Nilsson, Oops Art: urban@oops.se, d7urban@dtek.chalmers.se, un@cd.chalmers.se, d7urban@mdstud.chalmers.se Hiroshima 45, Tjernobyl 86, Windows 95 ------------------------ schnapp --------------------------------------- So you just have to edit /NextDeveloper/Apps/InterfaceBuilder.app/Resources/images.table and add NSReturnSign. Restart IB and everything should be fine. BTW: The image has nothing to to with the functionality: you can still wire an NSTextField with an NSButton and select the -performClick: method to get the desired effect. Hope it helps, -dirk --- ______________________________________________________________________ Dirk Olmes OBJECT FACTORY Gesellschaft fuer Informatik und Datenverarbeitung mbH Otto-Hahn Str. 18, 44227 Dortmund, Germany Telephon +49 (0) 231 975 137 0 Telefax +49 (0) 231 975 137 99 dirk@object-factory.com http://www.object-factory.com/
From: snyers@yosemite.enst-bretagne.fr (Dominique SNYERS) Newsgroups: comp.sys.next.programmer Subject: unarchiving NSmutableArray Date: 12 Jun 1997 08:43:44 GMT Organization: ENST de Bretagne, Brest FRANCE Message-ID: <5nocs0$s6i@alfali.enst-bretagne.fr> Keywords: unarchiving, Coder I am having some problems with unarchiving a list of objects. Unarchiving is done in the following way: displayList = [NSUnarchiver unarchiveObjectWithFile:aFile]; where dispayList is a NSMutableArray. The unarchiving process go through all the initWithCoder: with the correct instantiation of variables for all the objects from archived displayList but when I want to use displaylist afterward its count value shows the correct number of objects but the objects are all nil pointers. I am puzzled. Can someone send me an example of archiving and unarchiving NSMutableArray? Thanks for the help. Dom Snyers -- -------------------------------------------------------------------------- Laboratoire d'Intelligence Artificielle et Sciences Cognitives (LIASC) Ecole Nationale Superieure des Telecommunications de Bretagne BP 832 29285 BREST CEDEX FRANCE Tel: (33) 2 98 00 14 31 email:Dominique.Snyers@enst-bretagne.fr http://liasc.enst-bretagne.fr/~snyers
From: poeiru@eqqs.org Newsgroups: comp.sys.next.programmer Subject: Look what I found Date: 12 Jun 1997 07:10:09 GMT Organization: DEC Inc. Message-ID: <5no7ch$olp$71@nw001.infi.net> Hey, just thought i'd share with everyone, I found a site with loads of nude CHEERLEADERs. The address is: http://www.mid-night.com/cheer.htm --Jason-- (Sorry for the intrusion, everyone needs some short skirts in their life) P.S. They also have a few thounsand celebrities but im not into that.
From: michael@ninebits.com (Michael Balle) Newsgroups: comp.sys.next.programmer Subject: Re: DO NT/MACH Date: Thu, 12 Jun 1997 09:12:55 +0200 Organization: Nine Bits Message-ID: <19970612091255288345@[222.223.224.4]> References: <199706110937083521842@[222.223.224.4]> <339EAB42.1664@symphony.amd.com> Patrick Schulz <pschulz@symphony.amd.com> wrote: > The answer is: YES. All OpenStep compliant DO implementations can > communicate > to each other. That includes OPENSTEP on Mach, NT, PDO for > Solaris/HP-sUX > AND even Sun's OpenStep Implementation (but not the GNUstep DO system > [not OS compliant yet]). > > The Mach messages system is emulated on non Mach platforms. So it works > the same way, whatever operating system you use. I recall some articles > in > german magazines about DO/PDO but they are in german ;-) > > hope that helps, Well, this OpenStep thing just sounds better and better. You answer open up another question. I'm new to OpenStep, though I have been reading some magazines and a book about NextStep in the past years, so I'm a little pussled about what parts of OpenStep are supported on Solaris/HP-UX. I guess they have Obj-C compilers, the OpenStep Obj-C runtime environment and the foundation classes, but I'm not quite sure about the latter. Would it be possible to take some existing data-model, "engine", written entirely in C++, wrap it with some PDO, and get a userinterface-less server process, accessible using PDO from any OpenStep application (Mach/NT)? Michael
From: poeiru@eqqs.org Newsgroups: comp.sys.next.programmer Date: 12 Jun 1997 07:10:09 GMT Message-ID: <cancel.5no7ch$olp$71@nw001.infi.net> Subject: cmsg cancel <5no7ch$olp$71@nw001.infi.net> Control: cancel <5no7ch$olp$71@nw001.infi.net> Organization: Usenet Canal Historique ECP/EMP aka SPAM or pyramidal scheme (MMF) cancelled by bofh@keltia.freenix.fr. It may also be an image too small for newsbot to be activated. See report in news.admin.net-abuse.bulletins. Date: Thu Jun 12 13:34:14 1997 Original subject was: Look what I found
From: "Peter Geissler" <peter.geissler@iwr.uni-heidelberg.de> Newsgroups: comp.sys.next.programmer Subject: Question about NXConvertRGBtoHSB Date: Thu, 12 Jun 1997 16:13:25 +0200 Organization: InterNetNews at News.BelWue.DE (Stuttgart, Germany) Message-ID: <5np066$a22$2@news.belwue.de> Hi everybody, I am in the need to implement and RGB to HSB conversion on a non NextStep system and would very much like to have the same conversion conventions as used by the NXConvertRGBtoNSB etc. functions. I have implemented several conversion schemes found in literature, but none of these give the same results as the Nextstep built-in conversion does. Does anybody knows which conversion scheme is used by these functions ? Thanks in advance, Peter Geissler -------------------------------------------------------------------------- Peter Geissler Interdisciplinary Center for Scientific Computing University of Heidelberg peter.geissler@iwr.uni-heidelberg.de ---------------------------------------------------------------------------- -
From: Evan M Benoit <benoit+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.misc Subject: Network processes when logged out Date: Thu, 12 Jun 1997 10:34:46 -0400 Organization: Sophomore, Social & Decision Sciences, Carnegie Mellon, Pittsburgh, PA Message-ID: <cnc0Y6K00iVC02mMw0@andrew.cmu.edu> Hey Everyone, I've run into a snag trying to get this network monitoring application to work on a NeXT box. The application is called Big Brother, and what it does is check the disk space and processes running on the machine it runs on (a NeXT box) and send the results over the network to the server (some unix box). It sends updated information every 5 minutes. My problem is, when I'm logged into the NeXT box it works fine, but when I log out, it stops updating its stats on the server. The process is still running on the client, though. When I log back into the NeXT box, it resumes updating, like nothing had happened. It's weird, and it hasn't happened on any other OS's I've ran this thing on. Can anyone out there give me a hand? Is there some interuption of network functions when the user logs out? Any way to work around that? Any wild guesses? Thanks a lot! -ev ---EVAN BENOIT--- loved by good feared by evil
From: "Timothy J. Luoma" <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: wu-ftpd: link statically 'ls' etc ? Date: Thu, 12 Jun 1997 11:24:47 -0400 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.NXT.3.96.970612112105.6213D-100000@cc344191-a> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII wu-ftpd suggests statically linked binaries Can this be done under NS/OS (3.3/4.1)? (I was planning on using the GNU versions, since I have source code for them ;-) If so, how? Thank ye kindly TjL -- TjL <luomat@peak.org> / http://www.peak.org/~luomat/next/ NeXT bookmarks: http://www.peak.org/~luomat/next/bookmarks.html sed 's/End\ of\ sig/pithy\ quotation/g'
From: maury@softarc.com (Maury Markowitz) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: Thu, 12 Jun 1997 11:21:44 -0400 Organization: SoftArc Inc. Message-ID: <maury-1206971121440001@199.166.204.230> References: <5lvn4l$tje@src-news.pa.dec.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> <5nmfjt$9423@eccws1.dearborn.ford.com> In article <5nmfjt$9423@eccws1.dearborn.ford.com>, tph@rmi.net wrote: > John Norstad (j-norstad@nwu.edu) wrote: > : example, the server version will support spanning, striping, and mirroring This is interesting (sorry about the reply to the reply guys). Does anyone know how this is being offered? AIX's drive management system perhaps (it's very good), Veritas maybe? Maury
From: Susan Hovde <hovde@al.noaa.gov> Newsgroups: comp.sys.next.programmer Subject: segmentation fault P2R Date: Thu, 12 Jun 1997 09:36:44 -0600 Organization: NOAA Aeronomy Lab Message-ID: <33A01784.9A9@al.noaa.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I finally got Prelude To Rhapsody running on my Pentium 120 MHz with 32 MB RAM and it seems to be working well. However, when I try to run ProjectBuilder or InterfaceBuilder I get a segmentation fault. Ideas anyone? TIA, Susan Hovde NOAA Aeronomy Lab Boulder, CO
From: paulmeyers@worldnet.att.net (Paul Meyers) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: Thu, 12 Jun 1997 13:01:10 -0400 Organization: University of Florida Message-ID: <paulmeyers-1206971301110001@177.orlando-011.fl.dial-access.att.net> References: <338b6960.76768104@news.pdnt.com> <AFB060E7-7C68@207.173.163.78> <5mr0tm$opq$3@nntp.ucs.ubc.ca> <j-ochs-0806972044490001@wil123068.res-hall.nwu.edu> <5nht6j$6he$1@ocoee.iac.net> <5ni50a$bju$2@darla.visi.com> <5ni5ot$81v$1@ocoee.iac.net> I understand the initial Rhapsody developer's release will only work on 8500s, 8600s, 9500s, and 9600s. Does anybody know what kind of time frame they have in mind for expanding support to other machines, such as PowerTower Pros or PowerCenter Pros? Paul Meyers
Newsgroups: comp.sys.next.programmer Subject: keyUp Question (simple) From: apl@kcbbs.gen.nz (Andrew Lindesay) Date: 12 Jun 97 07:20:27 GMT Message-ID: <17497162.26427.11828@kcbbs.gen.nz> Organization: Kappa Crucis Unix BBS, Auckland, New Zealand I have written a 'keyUp' into a View I'm writing and it's beeping :-( Basically I want my View to run keyUp in my View; which isn't doing very nicely, but it's then going on (or before?) to produce a 'beep'. Here is my code: - keyUp:(NXEvent *)theEvent { switch(theEvent->data.key.charCode) { case 0x7f:// DELETE KEY [self cut:self]; return self; break; default:// don't have a handler. [nextResponder keyDown:theEvent]; return nil; break; } } Can anybody suggest a solution for this little problem? Thanks! ____ Andrew Lindesay (apl@kcbbs.gen.nz) NeXTmail ok. Computers, like coffee are best taken black.
From: Patrick Schulz <pschulz@symphony.amd.com> Newsgroups: comp.sys.next.programmer Subject: Re: DO NT/MACH Date: Thu, 12 Jun 1997 11:52:56 -0500 Organization: AMD Fab30 Message-ID: <33A02968.BFC@symphony.amd.com> References: <199706110937083521842@[222.223.224.4]> <339EAB42.1664@symphony.amd.com> <19970612091255288345@[222.223.224.4]> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Michael Balle wrote: > so I'm a little pussled about what parts of OpenStep are supported on > Solaris/HP-UX. > > I guess they have Obj-C compilers, the OpenStep Obj-C runtime > environment and the foundation classes, but I'm not quite sure about the > latter. AFAIK the following products are out or announced: PDO4.x for HP-sUX and Solaris 2.5 from NeXT(Apple) (contains gcc for Obj-C, Objc++, C, C++; gdb; Objc-Runtime; all Foundation classes) EOF for PDO from NeXT(Apple) (contains all non GUI elements of EOF2.x) OpenStep for Solaris 1.0 (is 1.1 released? 2/97 it was in beta) from Sun User - Runtime System and Applications (like NeXTSTEP User 3.2) Workshop - IB, ProjectBuilder, Sun ObjC++ compiler, all STANDARD (!) OpenStep classes (none of NeXT's extensions, only what's in the '94 Spec) EOF for OpenStep Solaris from NeXT(Apple) There's an announcement which is not very clear formulated. But since there's an EOF for PDO out since mid last year this EOF is for use with Sun's OpenStep implementation. I've never seen more than this announcement. So, if it exists it should look like EOF for OPENSTEP-Mach. Here's the link: http://www.next.com/AboutNeXT/PressKit/PressReleases/1996/Solarisv2.html Some comments on Sun's OpenStep: I don't know what strategy Sun has regarding OpenStep for Solaris I'd assume that they don't have one at all. There are no activities visible from the outside, the last version I tested needs some major improvements (speed!), there should be a unification of the different OpenStep implementations (you can't deploy a program wich uses NeXT's nice extensions [text system] to Sun's implementation :-( OTOH the UI of OpenStep (workpace manager) is by far the best user environment for a workstation (IMHO), Sun delivers CDE... So folks at Sun, what's the deal with Solaris OpenStep? > Would it be possible to take some existing data-model, "engine", written > entirely in C++, wrap it with some PDO, and get a userinterface-less > server process, accessible using PDO from any OpenStep application > (Mach/NT)? > > Michael Radio Erivan says: as a matter of principle, YES. :-) If your C++ engine can be compiled (or at least linked) with gcc for PDO this should be possible. There are even ways to "talk" from c++ objects to Obj-C objects, C-style functions should work without problems. I've never done such a project, so these are assumptions based on the product specifications and documentation. Can somebody validate this? Yepp, there are promising capabilities in OpenStep... Patrick. -- Patrick Schulz; 1704 Nelms Dr. #2025; Austin, TX 78744 email: pschulz@symphony.amd.com (MIME welcome) - vmunix: panic - no coffee detected, user halted.
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: 12 Jun 1997 17:14:30 GMT Organization: WolfWare Message-ID: <5npapm$k9j$1@vader.wolfware.ipc.net> References: <5lvn4l$tje@src-news.pa.dec.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> <5nmfjt$9423@eccws1.dearborn.ford.com> <maury-1206971121440001@199.166.204.230> In-Reply-To: <maury-1206971121440001@199.166.204.230> On 06/12/97, Maury Markowitz wrote: >In article <5nmfjt$9423@eccws1.dearborn.ford.com>, tph@rmi.net wrote: > >> John Norstad (j-norstad@nwu.edu) wrote: >> : example, the server version will support spanning, striping, and mirroring > > This is interesting (sorry about the reply to the reply guys). Does >anyone know how this is being offered? AIX's drive management system >perhaps (it's very good), Veritas maybe? > >Maury I don't think anyone knows yet (even at Apple). At WWDC these were listed as features that they thought were important for the server version but they emphasized that they had not yet begun even looking at how they were actually going to be implemented. - Chris -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer,comp.sys.mac.misc,comp.sys.mac.programmer.misc Subject: Re: Rhapsody features? Date: Thu, 12 Jun 1997 14:54:24 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <onc4LU200iWT04uuI0@andrew.cmu.edu> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> <19970611212335461585@marconi.physik.tu-berlin.de> <5nnres$390$2@kaopala.mhpcc.edu> [ ...newsgroups trimmed... ] Excerpts from netnews.comp.sys.next.misc: 12-Jun-97 Re: Rhapsody features? by L e e Altenberg@no_junk_ > Could anyone explain why an OS optimized for use as a server would not > also be optimal for a workstation? Because the two roles are different? Let's consider just one tunable parameter for the sake of example-- the scheduler quantum. Longer quantums result in more throughput due to reduced context-switching overhead. Shorter quantums have more context switching overhead, but they make the machine feel more responsive to user input, since more processes get timeslices over a given interval. There are lots of similar tradeoffs available depending on how one wishes to tune the system-- the size of the buffer cache, the size of the inode cache, the size of the process table (although hopefully, that will not be static but dynamicly allocated), tweaking the scheduling algorithm, et cetera. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Susan Hovde <hovde@al.noaa.gov> Newsgroups: comp.sys.next.programmer Subject: tabbing in Currency Converter Date: Thu, 12 Jun 1997 15:05:24 -0600 Organization: NOAA Aeronomy Lab Message-ID: <33A06476.7606@al.noaa.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit In the Prelude To Rhapsody Currency Converter developer tutorial, I have set and verified the nextKeyView connections so that the tab key should shift only between the first two fields. However, running the interface both under Interface Builder and as the full-blown app shows that the tabbing connections are not working. With the third field set as not selectable, tabbing from the second field causes the Convert button to be highlighted, rather than causing the insertion point to move to the first field as it should. (If I make the third field selectable, tabbing from the second field causes the insertion point to move to the third field.) It seems that the nextKeyView connections are being totally ignored. Has anyone else had this problem? Am I missing something? TIA, Susan Hovde
From: "Amanda Walker" <amanda_walker@ascend.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Wed, 11 Jun 1997 11:07:49 -0400 Organization: A poorly-installed InterNetNews site Message-ID: <5nmf6e$ffg$1@news.intercon.com> References: <5mvclb$jb2$1@news.digifix.com> <AFB89A0D-2FF75B@206.165.44.70> Lawson English wrote ... >GX provides functionality that DPS + AppKit doesn't and won't for quite >some time. True but not necessarily relevant. DPS+AppKit provides functionality that GX doesn't and won't for some time. The question is which set is more relevant to the actual marketplace, and this is not a question that can be answered by looking at a manual. For example, I would much rather use DPS for multilingual text rendering than GX, if only because there are more supporting resources (example: Morisawa's unequaled Japanese typeface library is available for PostScript, but not for GX (aside from a couple fonts that Apple has converted to TrueType GX); similarly for most other non-Roman typefaces). PostScript, warts and all, is a more mature and field-proven techology base than GX. >E.G. full implementation of GX Typography. Full implementation of GX >Printing. GX Printing is dead, according to Apple. MacOS 8 will support GX graphics for the screen but not for printers. And if I have to generate PostScript anyway, might as well pipe it to the screen instead of having two rendering methods for everything. I also like the ability of DPS to support user-defined PostScript procedures that are loaded into the window server (close enough to "retained mode" as not to matter for 2-D, IMNSHO). This allows the PS interpreter to deal with hardware acceleration and such, one of GX's big blind spots. Amanda Walker Senior Software Engineer Ascend Communications, Inc., Client Software Group
From: MaRK_BeSSeY@NeXT.CoM (Mark Bessey) Newsgroups: comp.sys.next.programmer Subject: Re: tabbing in Currency Converter Date: 12 Jun 1997 22:43:18 GMT Organization: Apple Computer, Inc. Message-ID: <5npu26$jok$1@news.apple.com> References: <33A06476.7606@al.noaa.gov> Susan Hovde <hovde@al.noaa.gov> writes > In the Prelude To Rhapsody Currency Converter developer tutorial, I have > set and verified the nextKeyView connections so that the tab key should > shift only between the first two fields. However, running the interface > both under Interface Builder and as the full-blown app shows that the > tabbing connections are not working. You need to set the "initialFirstResponder" outlet of the window to point at the first text field. If the initialFirstResponder isn't set, then the Appkit ignores the nextKeyView connections. A related issue you might run into later is that the initialFirstResponder must be an NSView, not an NSCell. If you want to hook up initialFirstResponder to a matrix of cells (for instance, an NSForm), make sure that you connect in to the containing view, rather than one of the cells inside. -Mark -- Mark Bessey Apple Computer, Inc. -->I DON'T SPEAK FOR APPLE<-- until the WWDC that they realized they should support the 9500 and 9600. Like developers interested in leading- edge software wouldn't buy top-of-the-line hardware. Sheesh... Not that this affects me and my Performa 580. Cliff ============= Maui no ka oi =============
From: quinlan@intergate.bc.ca (Brian Quinlan) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: Thu, 12 Jun 1997 14:43:17 -0700 Organization: Internet Gateway Corporation Message-ID: <quinlan-1206971443170001@pm14s11.intergate.bc.ca> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> <19970611212335461585@marconi.physik.tu-berlin.de> <5nnres$390$2@kaopala.mhpcc.edu> In article <5nnres$390$2@kaopala.mhpcc.edu>, decoy_id@no_junk_on_the.net (L e e Altenberg) wrote: > Could anyone explain why an OS optimized for use as a server would not > also be optimal for a workstation? The schedualer might need to be different to give priority to interactive processes. Server processes tend to block on IO frequently but a lot of GUI applications do not. Probably other stuff. -- Brian Quinlan quinlan@intergate.bc.ca
From: dnelson@slip.net (Dru Nelson) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Followup-To: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Date: 12 Jun 1997 23:41:19 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5nq1ev$c85$1@owl> References: <5lvn4l$tje@src-news.pa.dec.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> <5nmfjt$9423@eccws1.dearborn.ford.com> <maury-1206971121440001@199.166.204.230> IBM doesn't use veritas, and yes it is good. sun is going to use veritas, but who knows what NeXT will use.... Maury Markowitz (maury@softarc.com) wrote: > This is interesting (sorry about the reply to the reply guys). Does > anyone know how this is being offered? AIX's drive management system > perhaps (it's very good), Veritas maybe? > Maury
From: gad@eclipse.its.rpi.edu (Garance A. Drosehn) Newsgroups: comp.sys.next.programmer Subject: Re: Rhapsody features? Date: 12 Jun 1997 23:43:36 GMT Organization: Rensselaer Polytechnic Institute, Troy NY, USA Distribution: world Message-ID: <5nq1j8$6d@usenet.rpi.edu> References: <5nnres$390$2@kaopala.mhpcc.edu> Lee Altenberg writes: > Could anyone explain why an OS optimized for use as a server > would not also be optimal for a workstation? Because they will see different workloads. They will also have different criteria for what "acceptable performance" means. It is also reasonable to assume that servers are going to be higher-end configurations than user workstations would be (more RAM, etc), and you might change tuning parameters based on that difference. There can be a lot of reasons. -- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu Senior Systems Programmer (MIME & NeXTmail capable) Rensselaer Polytechnic Institute; Troy NY USA
From: goldsmith@apple.com (David Goldsmith) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Thu, 12 Jun 1997 17:52:09 -0700 Organization: Apple Computer, Inc. Message-ID: <goldsmith-ya02408000R1206971752090001@news.apple.com> References: <5mvclb$jb2$1@news.digifix.com> <AFB89A0D-2FF75B@206.165.44.70> <5nmf6e$ffg$1@news.intercon.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit In article <5nmf6e$ffg$1@news.intercon.com>, "Amanda Walker" <amanda_walker@ascend.com> wrote: >GX Printing is dead, according to Apple. MacOS 8 will support GX graphics >for the screen but not for printers. No, not quite. MacOS 8 will support GX graphics for printers, just through the classic printing architecture. -- David Goldsmith Architect International, Text, and Graphics Department Apple Computer, Inc. goldsmith@apple.com
From: embuck@palmer.cca.rockwell.com (Erik M. Buck) Newsgroups: comp.sys.next.programmer Subject: 4.2 incompatibilities Date: 13 Jun 1997 01:12:39 GMT Organization: Rockwell Collins Message-ID: <5nq6q7$bda2@castor.cca.rockwell.com> 4.2 incompatibilities Has anyone else noticed the following ? Is it just me ? 4.2 can load 4.1 bundles but not visa-versa (Was this in the release notes ?) debugging of autorelease pools has changed and is less powerful in 4.2 See rant below object files are sometimes MUCH larger in 4.2 the semantics of NSClassNamed() have changed. It did not used to call the unknown class handler under any circumstances and now does. It is now impossible to break on _freedhandler() The makefiles changed again in some incompatible ways. Variables that used to have relative paths now have absolute paths. This interfears with frameworks that are shared over a network because paths are sometimes stored in the framework. Does anybody know when or why ? The after_install:: target is not always called after install...It is sometimes called before install due to line 172 in common.make (I am writing this from memory and may have the line number or file name slightly wrong) finally, the BIG one: NSAutorelease pools now ALWAYS call _objc_error and abort WITHOUT raising an exception if a freed object is in the pool. The various NSDebug.h tricks to prevent this and debug this DO NOT WORK. This makes it extremely difficult to debug these things! The environment variables controlling this stuff do not work either.
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer Subject: Re: NSDictionary grammar? Date: 11 Jun 1997 21:18:09 GMT Organization: Digital Fix Development Message-ID: <5nn4mh$d8i$1@news.digifix.com> References: <5nn35k$hmq$1@owl> In-Reply-To: <5nn35k$hmq$1@owl> On 06/11/97, Dru Nelson wrote: > What is the grammar for an OpenStep NSDictionary object? > (or where is the documentation) > > Are the initFromFile:/writeToFile:atomically: methods the best > way of storing little configurations in an ascii format or is > there a better serializer/persistence methodology? > I think Project Builder uses the same method in it's PB.project > file. > >Dru Nelson - dnelson@slip.net - Redwood City, California > > Actually, you want the Persistent Property List grammar.. PropertyList(5) UNIX Programmer's Manual PropertyList(5) NAME PropertyList - ASCII Property List format DESCRIPTION A property list organizes data into named values and lists of values. Property lists are used by the NEXTSTEP user defaults system (among other things). This man page pro- vides a brief and incomplete overview for casual users; see <<xref>> for a detailed description of the ASCII property list format. In simple terms, a property list contains strings, binary data, arrays of items, and dictionaries. These four kinds of items can be combined in various ways, as described below. A string is enclosed in double quotation marks; for example, "This is a string." (The period is included in this string.) The quotation marks can be omitted if the string is composed strictly of alphanumeric characters and contains no white space (numbers are handled as strings in property lists). Though the property list format uses ASCII for strings, note that NEXTSTEP uses Unicode. You may see strings containing unreadable sequences of ASCII characters; these are used to represent Unicode characters. <<Details?>> Binary data is enclosed in angle brackets and encoded in hexadecimal ASCII; for example, <0fbd777 1c2735ae>. Spaces are ignored. An array is enclosed in parentheses, with the elements separated by commas; for example, ("San Francisco", "New York", "London"). The items don't all have to be of the same type (for example, all strings) - but they normally should be. Arrays can contain strings, binary data, other arrays, or dictionaries. A dictionary is enclosed in curly braces, and contains a list of keys with their values. Each key-value pair ends with a semicolon. Here's a sample dictionary: { user = maryg; "error string" = "core dump"; code = <fead0007>; }. (Note the omission of quotation marks for single-word alphanumeric strings.) Values don't all have to be the same type, since their types are usually defined by whatever pro- gram uses them (in this example, the program using the dic- tionary knows that user is a string and code is binary data). Dictionaries can contain strings, binary data, arrays, and other dictionaries. Below is a sample of a more complex property list, taken from a user's defaults system (see defaults(1)). The pro- perty list itself is a dictionary with keys "Clock," "NSGlobalDomain," and so on; each value is also a diction- ary, which contains the individual defaults. { Clock = {ClockStyle = 3; }; NSGlobalDomain = {24HourClock = Yes; Language = English; }; NeXT1 = {Keymap = /NextLibrary/Keyboards/NeXTUSA; }; Viewer = {NSBrowserColumnWidth = 145; "NSWindow Frame Preferences" = "5 197 395 309 "; }; Workspace = {SelectedTabIndex = 0; WindowOrigin = "-75.000000"; }; pbs = {}; } SEE ALSO defaults(1) <<TM notice for Unicode?>> -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: arti@address.in.signature (Art Isbell) Newsgroups: comp.sys.next.programmer Subject: Re: Hair Pulling with my first stab at IB! (please help :-)) Date: 13 Jun 1997 04:37:13 GMT Organization: LavaNet, Inc. Distribution: world Message-ID: <5nqipp$n94@mochi.lava.net> References: <5nev5t$bqg$1@news.sas.ab.ca> <5ng6au$acj$1@leonie.object-factory.com> dirk@object-factory.com (Dirk Olmes) wrote: > the NSReturnSign seems to have been lost in OpenStep for Mach 4.1 (It > was in fact lost in the 4.0 beta but I assumed it would find it's way > home in the sharp release). I have been able to locate the image in the > file /usr/lib/NextStep/Resources/SharedGray.tiff, but the problem lies > within InterfaceBuilder, namely the file > InterfaceBuilder.app/Resources/images.table where the NSReturnSign > entry is missing. Just add the entry and everything works as it should. > > Just our two cents, > Urban and Malte > > > --- > Urban Nilsson, Oops Art: urban@oops.se, d7urban@dtek.chalmers.se, > un@cd.chalmers.se, d7urban@mdstud.chalmers.se > Hiroshima 45, Tjernobyl 86, Windows 95 > ------------------------ schnapp --------------------------------------- > > So you just have to edit > /NextDeveloper/Apps/InterfaceBuilder.app/Resources/images.table and add > NSReturnSign. Restart IB and everything should be fine. No, no, no!!! Don't recommend this sort of hack to new developers. The NSReturnSign is no longer used to signify the default button - the one that will be pressed when the RETURN/ENTER key is pressed. Forget about NSReturnSign. Instead, enter "\r" as the button's or buttonCell's key in its Attributes Inspector. At runtime, this will result in a dark border around this button. -- 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
Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc From: Don McGregor <mcgredo@NOSPAM.stl.nps.navy.mil> Subject: Re: Rhapsody features? Content-Type: text/plain; charset=us-ascii Message-ID: <33A0D544.58CB@NOSPAM.stl.nps.navy.mil> Sender: news@taurus.cs.nps.navy.mil Content-Transfer-Encoding: 7bit Organization: Naval Postgraduate School References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> <19970611212335461585@marconi.physik.tu-berlin.de> <5nnres$390$2@kaopala.mhpcc.edu> <quinlan-1206971443170001@pm14s11.intergate.bc.ca> Mime-Version: 1.0 Date: Fri, 13 Jun 1997 05:06:12 GMT Brian Quinlan wrote: > > In article <5nnres$390$2@kaopala.mhpcc.edu>, decoy_id@no_junk_on_the.net > (L e e Altenberg) wrote: > > > Could anyone explain why an OS optimized for use as a server would not > > also be optimal for a workstation? > > The schedualer might need to be different to give priority to interactive > processes. Server processes tend to block on IO frequently but a lot of > GUI applications do not. Probably other stuff. In most flavors of Unix there are a bunch of tables in the kernel for things like process table slots and filesystem buffers. The tables are constant size; you can set them small at boot time, and take up less memory, or set them big, so they can handle anything you throw at them, but also suck up memory. What a "reasonable" table size is differs for personal workstations and servers, since they have vastly different usage profiles. A workstation usually won't be serving up some high-traffic NFS mounts or dealing with a big news feed, so configuring a workstation as if it will handle these tasks is a waste of hardware. Usually it's no big deal to change the kernel parameters; many unices have kernel directories that let users recompile and relink the kernel with modified parameters. Mach isn't quite Unix, but I presume the ideas are the same. Microsoft famously ships NT "Workstation" and NT "Server", the difference in the kernels consisting only of a couple modifications to kernel tuning parameters and several hundred dollars in retail price. With a little smoke and mirrors (aka "market segementation"), Microsoft is able to add a few megabucks to Bill's bank account. -- Don McGregor mcgredo@stl.nps.navy.mil http://www.stl.nps.navy.mil/~mcgredo
From: Alex Blakemore <alex@genoa.com> Newsgroups: comp.sys.next.programmer Subject: Re: tabbing in Currency Converter Date: 13 Jun 1997 01:43:37 GMT Organization: Genoa Software Systems Message-ID: <5nq8k9$30k@saturn.genoa.com> References: <33A06476.7606@al.noaa.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: hovde@al.noaa.gov, mwyner@apple.com In <33A06476.7606@al.noaa.gov> Susan Hovde wrote: > In the Prelude To Rhapsody Currency Converter developer tutorial, I have > set and verified the nextKeyView connections ... > However, running the interface both under Interface Builder and as the > full-blown app shows that the tabbing connections are not working. ... > It seems that the nextKeyView connections are being totally ignored. you must set the connection for initialFirstResponder (or something named similar from the window to the first field) it would be nice if the tutorial mentioned this. -- Alex Blakemore alex@genoa.com NeXT, MIME and ASCII mail accepted
Date: Fri, 13 Jun 1997 03:51:04 -0600 From: amas@lhr-sys.dhl.com Subject: [Q] Main executable in application folder Newsgroups: comp.sys.next.programmer Message-ID: <866191282.19621@dejanews.com> Organization: Deja News Usenet Posting Service Maybe I am not using the right terminology (correct me where you can), though that should not be a problem. When the main executable is placed in the application's folder, how does the system know which it is, is there a special name or something. Also, if I wish to make the main code say a Perl script or csh script does Open Step support the possibility and if it does, how does it know which execution environment to start? If this is not currently a possiblity, would anyone see this as an idea for future implementations - especially with Java coming along. Thanks - Please CC me your reply Andre -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: tzs@halcyon.com (Tim Smith) Newsgroups: comp.sys.next.programmer,comp.sys.next.advocacy,comp.sys.mac.programmer.codewarrior,comp.sys.mac.programmer.help Subject: Re: Looking for Prelude to Rhapsody (It should be free!) Date: 13 Jun 1997 02:38:59 -0700 Organization: Northwest Nexus Inc. Message-ID: <5nr4fj$qfm$1@halcyon.com> References: <338b6960.76768104@news.pdnt.com> <5ni5ot$81v$1@ocoee.iac.net> <paulmeyers-1206971301110001@177.orlando-011.fl.dial-access.att.net> <33A07946.13DD@research.canon.com.au> <cstory@research.canon.com.au> wrote: >What's amazing about this is that it wasn't until the WWDC that they realized >they should support the 9500 and 9600. Like developers interested in leading- >edge software wouldn't buy top-of-the-line hardware. Sheesh... Well, one problem with buying top-of-the-line hardware for development is that you might find that the tools you want to use don't quite work yet. With each new Mac, Apple likes to do things that break The Debugger, and if you are too eager, you may end up getting your top-of-the-line Mac before Steve Jasik gets one and comes up with an update to support it. --Tim Smith
From: jak@asu.edu (John Kestner) Newsgroups: comp.sys.next.hardware,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.programmer Subject: 3.3 or 4.x? Date: Fri, 13 Jun 1997 02:34:35 -0700 Organization: Arizona State University Message-ID: <jak-1306970234350001@ss7-16.inre.asu.edu> Apologies for the newbie question: I'm planning to get a mono turbo slab, and I'll be using it for some light web/ftp serving, print serving (to a small Mac network), and hopefully for developing. But I'm wondering: What version of NS should I get? What are the differences between 3.0, 3.1, 3.2, 3.3, (which I guess is the minimum I should get) and 4.x? Rhapsody is supposed to be a superset of NS, so I can develop now on black hardware and recompiling will be all that's necessary, right? Will stuff I develop in 3.3 carry over, or would I have to have 4.x? I've been looking through the www resources for OS version comparisons, but haven't found anything. If someone could point me to a resource with the answer, or just tell me, I'd appreciate it. Oh yeah, and cost is a factor. If you have NS Developer (or even User) to sell, I'll consider buying it. thanks john -- --- - ------- ------- The real in us is silent; the acquired is talkative. - Kahlil Gibran jak@asu.edu http://www.public.asu.edu/~jkestner/
From: Susan Hovde <hovde@al.noaa.gov> Newsgroups: comp.sys.next.programmer Subject: Re: tabbing in CC -- THANKS! Date: Fri, 13 Jun 1997 07:57:50 -0600 Organization: NOAA Aeronomy Lab Message-ID: <33A151DD.974@al.noaa.gov> References: <33A06476.7606@al.noaa.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Thanks to Mark Bessey and Alex Blakemore for clueing me in to initialFirstResponder; Currency Converter works great now. Onward! Susan Hovde
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] Main executable in application folder Date: 13 Jun 1997 16:54:02 GMT Organization: WolfWare Message-ID: <5nrtva$9co$1@vader.wolfware.ipc.net> References: <866191282.19621@dejanews.com> In-Reply-To: <866191282.19621@dejanews.com> On 06/12/97, amas@lhr-sys.dhl.com wrote: >Maybe I am not using the right terminology (correct me where you can), >though that should not be a problem. When the main executable is placed >in the application's folder, how does the system know which it is, is >there a special name or something. The name of the executable must be the same as the name of the application wrapper (folder) minus the app extension. I.E. If you have Foo.app the main executable is named Foo. >Also, if I wish to make the main code >say a Perl script or csh script does Open Step support the possibility Alas, no. The main code has to be an executable not a script file. You can kludge it of course by writing a very simple executable which calls system("myScriptFile") as it's only code to get it to run your script though. >Thanks - Please CC me your reply > >Andre > >-------------------==== Posted via Deja News ====----------------------- > http://www.dejanews.com/ Search, Read, Post to Usenet > -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: John Eric Ivancich <ivancich@quip.eecs.umich.edu> Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: 13 Jun 1997 13:00:04 -0400 Organization: University of Michigan EECS Department Message-ID: <yvx2066h20b.fsf@quip.eecs.umich.edu> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> <19970611212335461585@marconi.physik.tu-berlin.de> <5nnres$390$2@kaopala.mhpcc.edu> <quinlan-1206971443170001@pm14s11.intergate.bc.ca> <33A0D544.58CB@NOSPAM.stl.nps.navy.mil> Don McGregor <mcgredo@NOSPAM.stl.nps.navy.mil> writes: > In most flavors of Unix there are a bunch of tables in the kernel > for things like process table slots and filesystem buffers. The > tables are constant size; you can set them small at boot time, and > take up less memory, or set them big, so they can handle anything you > throw at them, but also suck up memory. What a "reasonable" table size > is differs for personal workstations and servers, since they have > vastly different usage profiles. A workstation usually won't be > serving up some high-traffic NFS mounts or dealing with a big > news feed, so configuring a workstation as if it will handle > these tasks is a waste of hardware. > > Usually it's no big deal to change > the kernel parameters; many unices have kernel directories that let > users > recompile and relink the kernel with modified parameters. What about an adaptive scheme? As the OS runs, the kernel makes note of the typical useage of various tables (e.g., on average the process table is 30% full), along with maximum usage, minimum usage etc. The kernel then figures out a good size for these tables to take effect on the next boot. Obviously if this were done, the tables would have to be allocated dynamically at boot to circumvent the need for a recompile/relink. Eric P.S. I hate spam.
From: mtrombin@ix.netcom.com (Mark Trombino) Newsgroups: comp.sys.next.programmer Subject: HELP: WMInspector changed in 4.1? Date: 13 Jun 1997 17:40:10 GMT Organization: Egghead Billy, Inc. Message-ID: <5ns0lq$m2i@dfw-ixnews10.ix.netcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Hello, I would like to write a Workspace Manager inspector, but am unable to follow the directions given in the NeXT docs. The file describing this procedure doesn't seem to be up to date. It claims that my inspector must inherit from a class named WMInspector. I've combed my entire filesystem and was unable to locate any reference to a WMInspector class. Has this class been renamed or something? I've searched NeXTAnswers but was unable to find any info there. I hope someone here can help! Please? BTW, I'm running 4.1 for mach. -- |-------------------------------------------------------------------------| | Mark Trombino | J A M S o f t | | mtrombin@ix.netcom.com | Audio DSP Tools for Openstep & Rhapsody | |-------------------------------------------------------------------------|
From: eric@skatter.USask.Ca Newsgroups: comp.sys.next.programmer Subject: Does OS4.2 finally fix the printf %g bug? Date: 12 Jun 1997 21:07:49 GMT Organization: University of Saskatchewan Message-ID: <5npof5$jlf@tribune.usask.ca> Would someone out there running OPENSTEP 4.2 try my famous printf test program? For those of you coming in late, this program tests for a bug that has been present since NeXTstep version 1.0 and which I first reported in December, 1989. I even have the NeXT bug tracking reference numbers [60697, 98369] and an e-mail message from NeXT (September 1994) indicating that, ``it looks like it will finally be fixed in NEXTSTEP release 4.0.'' ================================================================ #include <stdio.h> int main (int argc, char **argv) { printf ("Expect 0.00123: %.3g\n", 0.001234567); printf ("Expect 123: %.3g\n", 123.4567); printf ("Expect 123.5: %.4g\n", 123.4567); printf ("Expect 1e+03: %.3g\n", 999.6); return 0; } ================================================================= To forestall the usual flurry of responses that I get about this, please read the following before sending me a message that my test program is incorrect: The ANSI standard (X3.159-1989) says (Section 4.9.6.1, page 134, line 33): g,G The double argument is converted in style f or e (or in style E in the case of a G conversion specifier), with the precision specifying the number of significant digits. If the precision is zero, it is taken as 1. The style used depends on the value converted; style e (or E) will be use only if the exponent resulting from the conversion is less than -4 or greater than or equal to the precision. Trailing zeros are removed from the fractional portion of the result; a decimal-point character appears only if it is followed by a digit. *** NOTE THE WORDS `precision specifying the number of significant digits'. *** Let's have a look at some examples from the test program: format "%.3g" .001234567 ==== Result was: 0.001 ---- Result should have been: 0.00123 The zero's in 0.00123 are *not* significant. Score 1 against NeXT. format "%.3g" 999.6 ==== Result was: 999.6 ---- Result should have been: 1e+03 The format calls for 3 SIGNIFICANT DIGITS. Rounding 999.6 to three significant digits leaves 1000, which has an exponent equal to the precision and therefore should be printed in e format. Score 2 against NeXT. Also, - This is a LIBRARY issue, not a compiler issue. It doesn't matter *what* version of cc/gcc I use on the NeXT, the problem is in the library supplied by NeXT - K&R II does *not* conflict with the ANSI standard -- it just doesn't `tell the *whole* truth and nothing but the truth'. It mentions when to switch between e and f, but it doesn't say *anything* else. In particular, it *doesn't* say that %.3g should be rendered as %.3e or %.3f it just says when %e or %f style should be used. The appendix at the end of K&R II is, ``a summary of the library defined by the ANSI standard''. As such, it leaves the authors/publisher off the hook when they omit (for space reasons, I presume) parts of the standard. - The ANSI standard isn't that expensive, and isn't that hard to read. I can't figure why more people don't just go out and buy it. Here are some results from the machines I have access to at our lab: On a SUN 4/380 (SUNOS 4.1.1): Expect 0.00123: 0.00123 Expect 123: 123 Expect 123.5: 123.5 Expect 1e+03: 1e+03 So it would appear that at least *one* vendor can get it right! On an HP-715 (HP-UX V8 -- I think): Expect 0.00123: 0.00123 Expect 123: 123 Expect 123.5: 123.5 Expect 1e+03: 1e+03 Hmmm...looks like there are at least *two* vendors that can get it right. On a DEC MicroVAX II (Ultrix version 2.2): Expect 0.00123: 0.00123 Expect 123: 123 Expect 123.5: 123.5 Expect 1e+03: 1e+03 Wow, even an antique computer can do this properly! -- Eric Norum eric@skatter.usask.ca Saskatchewan Accelerator Laboratory Phone: (306) 966-6308 University of Saskatchewan FAX: (306) 966-6058 Saskatoon, Canada. NeXTMail accepted.
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer,comp.sys.mac.misc,comp.sys.mac.programmer.misc Subject: Re: Rhapsody features? Date: Fri, 13 Jun 1997 15:45:28 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <kncOBMS00iWn07BtM0@andrew.cmu.edu> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <*jc-2305972243290001@192.0.2.1> <338DB457.2579@abacus.com> <5ndo6o$d4n@slip.net> <cdoutyEBJpJs.64@netcom.com> <REMOVE.TO.REPLY.drg-1006972343570001@bmrip01.mdacc.tmc.edu> <j-norstad-1106970817050001@legume186154.nuts.nwu.edu> <19970611212335461585@marconi.physik.tu-berlin.de> <5nnres$390$2@kaopala.mhpcc.edu> <quinlan-1206971443170001@pm14s11.intergate.bc.ca> <33A0D544.58CB@NOSPAM.stl.nps.navy.mil> <yvx2066h20b.fsf@quip.eecs.umich.edu> In-Reply-To: <yvx2066h20b.fsf@quip.eecs.umich.edu> [ ...newsgroups snipped... ] Excerpts from netnews.comp.sys.next.misc: 13-Jun-97 Re: Rhapsody features? by John Eric Ivancich@quip. >> Usually it's no big deal to change the kernel parameters; many unices have >> kernel directories that let users recompile and relink the kernel with >> modified parameters. > > What about an adaptive scheme? As the OS runs, the kernel makes note > of the typical useage of various tables (e.g., on average the process > table is 30% full), along with maximum usage, minimum usage etc. The > kernel then figures out a good size for these tables to take effect on > the next boot. If the kernel can dynamicly allocate its' internal tables, there would be no need to reboot, and yes, it is generally desirable to have the kernel do this as much as possible. The problem comes in that it's difficult to get dynamic kernel table allocation working perfectly, and staticly-sized arrays tend to be faster to work with than linked lists or similar dynamic data structures. Being able to access and manipulate process table entries _fast_ is important, although the CPU speed of modern machines makes a little additional overhead less important than it used to be. Then again, supporting an unlimited number of processes can be important, too-- especially for a server; this is another area where tradeoffs can be made depending on the intended usage of the machine. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: wu-ftpd: link statically 'ls' etc ? Date: Fri, 13 Jun 1997 15:54:39 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <wncOJzC00iWn07BvM0@andrew.cmu.edu> References: <Pine.NXT.3.96.970612112105.6213D-100000@cc344191-a> In-Reply-To: <Pine.NXT.3.96.970612112105.6213D-100000@cc344191-a> Excerpts from netnews.comp.sys.next.programmer: 12-Jun-97 wu-ftpd: link statically 'l.. by "Timothy J. Luoma"@peak. > wu-ftpd suggests statically linked binaries > > Can this be done under NS/OS (3.3/4.1)? Yes, but most likely you're going to have to do a lot more work than you really want to. > (I was planning on using the GNU versions, since I have source code for > them ;-) > > If so, how? If NeXT/Apple made static versions of their shared libraries publicly available [hint, hint], or released the tools they use to manage shared libraries [ditto on the "hint"s], you'd have no problem. Unfortunately, neither is the case, so the next solution is to build yourself the GNU C library and/or libiberty.a, which will give you with a staticly-linkable standard C library in place of NeXT's shared libraries. Unfortunately (again), building libgcc.a under NEXTSTEP was a major pain in the posterior the last time I tried. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Benhur.Stein@imag.fr (Benhur Stein) Newsgroups: comp.sys.next.programmer Subject: hit test in postscript Date: 13 Jun 1997 20:50:07 GMT Organization: IMAG, Grenoble, France Message-ID: <5nsbpv$qgq$1@imag.imag.fr> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit Hi. I'm developping a program in OS4.1/intel, and it has to test for cursor hits in some displayed figures. I'm using postscript's infill operator to test for the hit, but am having some problems when the figures are too big. For example, the code below fails for a hit test inside a triangle (the triangle is drawn in the screen with the same code, with a fill instead of the infill). If the values 10000 are changed to 1000, for example, the hit succeeds. Is this a documented bug? Is it present in 4.2? Is it a limitation of the infill op? Does anybody have pointers for a good code for testing a hit in a line? newpath 10000 34 moveto -2 6 rlineto 4 0 rlineto closepath 10000 38 infill Thanks, Benhur
From: luomat@peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.hardware,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.programmer Subject: cmsg cancel <5ns28e$88n$2@bashir.peak.org> Control: cancel <5ns28e$88n$2@bashir.peak.org> Date: 13 Jun 1997 18:11:33 GMT Organization: Public Electronic Access to Knowlege,Inc Message-ID: <5ns2gl$88n$3@bashir.peak.org>
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: GX OOP class proposal & contest (was Re: GX game graphics Date: 13 Jun 1997 18:56:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFC74CB8-1F634@206.165.44.35> 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 This reply was sent to me from Joel Breckinridge, who doesn't have newsgroup access: Joel Breckinridge <velvetj@tactgraphic.co.jp> said: > > Amanda wrote: > > For example, I would much rather use DPS for multilingual text rendering > > than GX, if only because there are more supporting resources (example: > > Morisawa's unequaled Japanese typeface library is available for > > PostScript, > > but not for GX (aside from a couple fonts that Apple has converted to > > TrueType GX); similarly for most other non-Roman typefaces). > PostScript, > > warts and all, is a more mature and field-proven techology base than GX. > > While this is true, it does not answer the fact that PostScript (I'm not > talking DPS here) is insufficient for Japanese DTP needs or any 2 byte > needs. I have been with it since the start in 1989 and Japanese PS is > difficult to kern, mutiple weights are large, unweildy and take up too > much memory, this is just for starters. As I said earlier on the GX list: > > 'If anybody cares to know, there are no Mutiple Master Japanese fonts, > (or any 2 byte MM fonts, but there are some GX Japanese fonts) and there > is no way to embed most Japanese type 1 fonts in the recently released > Acrobat 3.0 J (the first to support 2 byte characters). Outside of GX, 2 > byte PDA is a kludge and sucks, to put it mildly.' > > Torsten Buck of FontWorks Inc. (the largest Japanese font vendor next to > Morisawa with many quality and popular typefaces) answered this post > most > perceptively, giving us some compelling reasons for GX Typography: > > >Now we are talking. We at Fontworks invested heavily over the last > couple > >of years in GX technology because it is/was addressing exactly the issues > >which were/are hindering DTP in Japan to take off as it did in the United > >States. > > > >There is for example the big file size of Japanese fonts (not to mention > >Chinese fonts); we are talking of 3 to 6 MB per weight. That means 30MB > >of hard disk space for one family in six weights. > > > >Or there is the need to be able to make up characters which are not in > >the standard character set. If you come across an "unusual" character > >when giving a document to be printed in a typesetting bureau they will > >make up this characters as they go; and simply charge you for the extra > >work. This is expensive but at least you get your job done. In the DTP > >world however the user is stuck. > > > >GX with the OTA gives a font vendor the possibility to address these > >issues. I'm not just talking about taking a PostScript or TrueType font > >and wrapping it up is a GX wrapper to give it some more features. We can > >(and will) do that with an ordinary 'sfnt' now as well. > > > >I'm talking about innovative font data structures. In order to get the > >file size down you have to find a way to avoid storing full outlines for > >every single of the over 6000 glyphs. > >We can now store one weight in about 1.5 MB without loosing any design > >detail compared with our "standard" PostScript font. This loss-less > >compression is extremely (I repeat: extremely) important in the so overly > >quality oriented Japanese market. > > > >But what's the point having a small data structure if the operation > >system doesn't know what to do with it. And this is where I see the > >biggest advantage of GX. It allows us to develop a scaler (i.e. a > >rasterizer with runs side by side with the build in TT or PS rasterizers) > >which knows how to extract the outlines from our compact data format. > And > >it goes even much further. I can produce weight variations within the > >scaler as well. A whole font family with weights anywhere between Light > >and UltraBold (not just the six default weight instances but ANY weight, > >like for example 35% between Light and Medium) can now fit in a suitcase > >of roughly 9 MB. > > > >This compact data structure happens to be a result of the way we > >construct characters in our production process. What's the benefit for > >the user? Well, we take a stripped down version of our in-house > >production software and, swupp, we have a Gaiji-editor. The DTP user can > >create it's own characters with such ease, you will have to see it to > >believe it. It's a real Macintosh user interface gem, I would say. All > >the technical details are hidden and the user can concentrate on the > >design process. > > Joel Breckinridge > Tact Graphic KK > Shizuoka, Japan > joel@tactgraphic.co.jp > > ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: Dru Nelson <dnelson@slip.net> Newsgroups: comp.sys.next.programmer Subject: Version control in 4.2prerelease? Date: 14 Jun 1997 08:04:04 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5ntj9k$ajg$1@owl> How can one add RCS/whatever so project builder uses version control? Dru Nelson - Redwood City - California
Newsgroups: comp.sys.next.programmer From: dfevans@bbcr.uwaterloo.ca (David Evans) Subject: Re: [Q] Main executable in application folder Sender: news@novice.uwaterloo.ca (Mr. News) Message-ID: <EBpyqL.61n@novice.uwaterloo.ca> Date: Fri, 13 Jun 1997 14:54:20 GMT References: <866191282.19621@dejanews.com> Organization: University of Waterloo In article <866191282.19621@dejanews.com>, <amas@lhr-sys.dhl.com> wrote: >Maybe I am not using the right terminology (correct me where you can), >though that should not be a problem. When the main executable is placed >in the application's folder, how does the system know which it is, is >there a special name or something. Yes. If you have an app wrapper called "Blah.app" then "Blah.app/Blah" will be run. >Also, if I wish to make the main code >say a Perl script or csh script does Open Step support the possibility >and if it does, how does it know which execution environment to start? > That should work fine, provided your script starts with the "#!/bin/csh" (or whatever) magic cookie. The kernel understands such things. However, the utility of this inside an app wrapper is questionable, since stdout will go to the console and there's no stdin (as far as I know). -- David Evans (NeXTMail/MIME OK) dfevans@bbcr.uwaterloo.ca Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/ University of Waterloo "Default is the value selected by the composer Ontario, Canada overridden by your command." - Roland TR-707 Manual
Newsgroups: comp.sys.next.hardware,comp.sys.next.misc,comp.sys.next.software,comp.sys.next.programmer From: dfevans@bbcr.uwaterloo.ca (David Evans) Subject: Re: 3.3 or 4.x? Sender: news@novice.uwaterloo.ca (Mr. News) Message-ID: <EBqGCo.IHH@novice.uwaterloo.ca> Date: Fri, 13 Jun 1997 21:14:48 GMT References: <jak-1306970234350001@ss7-16.inre.asu.edu> Organization: University of Waterloo In article <jak-1306970234350001@ss7-16.inre.asu.edu>, John Kestner <jak@asu.edu> wrote: > >But I'm wondering: What version of NS should I get? What are the >differences between 3.0, 3.1, 3.2, 3.3, (which I guess is the minimum I >should get) and 4.x? 3.2 is the bare minimum you should get, with 3.3 perferable. 4.x (4.2 preferably) should be OK on a Turbo machine as long as you put 64MB or more RAM in it. >Rhapsody is supposed to be a superset of NS, so I can develop now on black >hardware and recompiling will be all that's necessary, right? Will stuff I >develop in 3.3 carry over, or would I have to have 4.x? > Well, it's hard to say exactly what will be required in addition to a recompile. But, at a minimum, your code must be OPENSTEP-based, not NeXTSTEP-based. In that case 4.x is the only way to go. -- David Evans (NeXTMail/MIME OK) dfevans@bbcr.uwaterloo.ca Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/ University of Waterloo "Default is the value selected by the composer Ontario, Canada overridden by your command." - Roland TR-707 Manual
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: Re: Does OS4.2 finally fix the printf %g bug? Date: Sat, 14 Jun 1997 05:25:41 -0600 Organization: ErgoTech Message-ID: <33A27FB5.6635@ergotech.com> References: <5npof5$jlf@tribune.usask.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: eric@skatter.USask.Ca eric@skatter.USask.Ca wrote: > > Would someone out there running OPENSTEP 4.2 try my famous printf test program? > ================================================================ > #include <stdio.h> > > int main (int argc, char **argv) > { > printf ("Expect 0.00123: %.3g\n", 0.001234567); > printf ("Expect 123: %.3g\n", 123.4567); > printf ("Expect 123.5: %.4g\n", 123.4567); > printf ("Expect 1e+03: %.3g\n", 999.6); > return 0; > } > ================================================================= Since 4.2 is still prerelease and still under NDA, I would just tell Eric to port any 3.3 patches he has for printf to 4.2. What surprises me is that NeXT/Apple consider this bug unimportant. This is not just Eric and a particular broken example. We get complaints from users about the appearance of numbers formatted with %g. Jim
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: hit test in postscript Date: 14 Jun 1997 12:55:40 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5nu4cc$qte$1@brachio.zrz.TU-Berlin.DE> References: <5nsbpv$qgq$1@imag.imag.fr> In article <5nsbpv$qgq$1@imag.imag.fr> Benhur.Stein@imag.fr (Benhur Stein) writes: [Hit testing on OS 4.1/Intel fails] > newpath > 10000 34 moveto > -2 6 rlineto > 4 0 rlineto > closepath > 10000 38 infill This infil succeeds on my 4.0/Intel system. Maybe I shouldn't have ordered 4.2 yesterday? Marcel
From: davep@nospam.com (Dave Polaschek) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Sat, 14 Jun 1997 08:16:33 -0500 Organization: Polaschek Publishing Message-ID: <davep-1406970816330001@dialup75.mn.uswest.net> References: <5mvclb$jb2$1@news.digifix.com> <AFB89A0D-2FF75B@206.165.44.70> <5nmf6e$ffg$1@news.intercon.com> <goldsmith-ya02408000R1206971752090001@news.apple.com> In article <goldsmith-ya02408000R1206971752090001@news.apple.com>, goldsmith@apple.com (David Goldsmith) wrote: > In article <5nmf6e$ffg$1@news.intercon.com>, "Amanda Walker" > <amanda_walker@ascend.com> wrote: > >GX Printing is dead, according to Apple. MacOS 8 will support GX graphics > >for the screen but not for printers. > > No, not quite. MacOS 8 will support GX graphics for printers, just through > the classic printing architecture. David, You need to be VERY careful about raising people's expectations here. MacOS 8 will support GX graphics for printers, but only as bitmaps. Support for PostScript generation will not be in 8.0. -DaveP -- THE "REPLY TO" BUTTOM WILL NOT WORK WITH THIS POSTING The reply-to address has been altered to thwart junk e-mailers. Dave Polaschek - davep@best.com PGP key and other spiffy things at <http://www.best.com/~davep/>
From: gtupar@ctp.com (Georg Tuparev) Newsgroups: comp.sys.next.programmer Subject: Re: Version control in 4.2prerelease? Date: 14 Jun 1997 14:57:27 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5nubgn$d4r$1@concorde.ctp.com> References: <5ntj9k$ajg$1@owl> By adding CVS or PVCS, making some hacks to IB and PB (mine will be submitted one day), and cross your fingers. Being religios might also help especially if you pray very hard before start using the SCM module ;-) In article <5ntj9k$ajg$1@owl> Dru Nelson <dnelson@slip.net> writes: > > > How can one add RCS/whatever so project builder uses version > control? -- ------- /\/\ Georg Tuparev <georg_tuparev@ctp.com> | Currently in Dublin / /_ \ Cambridge Technology Partners 118/119 Lower Baggot Street \ / / Apollo House, Apollolaan 15 Dublin 2, Republic of Ireland \/\/ 1077 AB Amsterdam, The Netherlands Tel: +353(1)607-9083 Tel: +31(20)575-0492 Fax: +31(20)575-0500 WWW: http://www.ctp.com
From: Benhur.Stein@imag.fr (Benhur Stein) Newsgroups: comp.sys.next.programmer Subject: Re: hit test in postscript Date: 14 Jun 1997 14:50:43 GMT Organization: IMAG, Grenoble, France Distribution: world Message-ID: <5nub43$k0t$1@imag.imag.fr> References: <5nsbpv$qgq$1@imag.imag.fr> <5nu4cc$qte$1@brachio.zrz.TU-Berlin.DE> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit Cc: marcel@system.de In <5nu4cc$qte$1@brachio.zrz.TU-Berlin.DE> marcel@system.de wrote: > In article <5nsbpv$qgq$1@imag.imag.fr> Benhur.Stein@imag.fr (Benhur Stein) > writes: > > [Hit testing on OS 4.1/Intel fails] > > > newpath > > 10000 34 moveto > > -2 6 rlineto > > 4 0 rlineto > > closepath > > 10000 38 infill > > This infil succeeds on my 4.0/Intel system. Maybe I shouldn't have > ordered 4.2 yesterday? Well, I was testing with BeYap; with Yap it works. But changing 10000 with 30000, it doesn't work (that was my original problem, I simplified to 10000 to have more beautiful numbers..) thanks, benhur
From: Tony Scott<summer@laoffices.com> Newsgroups: comp.sys.next.programmer Subject: Distributors Wanted NOW!! Date: 14 Jun 1997 15:07:02 GMT Organization: World Wide Pants Promotion Message-ID: <5nuc2m$84p$7662@cadmium.aware.nl> Hi, This is a Solicitation Email, we are looking for new TUFF'97 DISTRIBUTORS. TUFF'97 is a promotional concept for a prospect follow-up system called "Simple Track". Software designed for Windows 95, 3.11 and NT machines. (Macintosh users read the NOTE below(*)) Concept: The TUFF'97 software includes a copy of Simple Track, so you can use this for your own personal administration (free). However, the TUFF'97 software is designed to give you BIG TIME credits for just taking the efforts to PROMOTE it. How do I get those credits?. Easy, just distribute/promote it. How? (Give away the software for free, put it on your website/homepage and let everyone download it. Give it to your friends, attach it by email etc). With a one time $40.00 (USD) validation code purchase, YOUR name becomes part of all sofware you sent out to your customers. When your TUFF'97 software gets validated, all your personal information is automatically integrated in the software, so everyone knows how to reach you by phone/fax or email and where to sent the check/money or cash orders. Basically this is the concept. Offcourse you need the software to grasp the whole thing. Believe it, there is absolutely no marketing knowledge required, just ask for the software package, join and start promoting your own TUFF'97 line. The only thing you DO need is a PC or Mac(*) and Internet. (*) TUFF'97 is only available for MacOS users running Connectix VirtualPC or Insignia's Sofwindows 95/3.11. A true MacOS version is not available. When you're interested in becoming a TUFF'97 distributor and create your own distribution line, sent a mailto:FiReStArTeR@frodo.com and ask for your copy of the TUFF software (We'll sent you a download location and install instructions). This is a serious opportunity, so only serious responses are processed. If you're not interested, have a nice day. Best Regards, Tony Scott Worldwide Pants Promotions 24H Support/Download requests, mailto:FiReStArTeR@frodo.com voice:(+01)779 698 4655 Fax :(+01)779 698 4555
From: rbraver@ohww.norman.ok.us Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5nuc2m$84p$7662@cadmium.aware.nl> Date: 14 Jun 1997 17:40:06 GMT Control: cancel <5nuc2m$84p$7662@cadmium.aware.nl> Message-ID: <cancel.5nuc2m$84p$7662@cadmium.aware.nl> Sender: Tony Scott<summer@laoffices.com> Spam cancelled. Notice ID: 19970614.03. See news.admin.net-abuse.announce or http://spam.ohww.norman.ok.us/spam_notices/19970614.03.html for complete report. Original Subject: Distributors Wanted NOW!!
From: mpaque@wco.com (Mike Paquette) Newsgroups: comp.sys.next.programmer Subject: Re: Problems compiling after OpenStep conversion Date: 14 Jun 1997 11:37:32 -0700 Organization: Electronics Service Unit No. 16 Sender: mpaque@mpaque Distribution: world Message-ID: <5nuodc$kr@mpaque.mpaque> References: <5nkred$llc@ucsbuxb.ucsb.edu> In article <5nkred$llc@ucsbuxb.ucsb.edu> goldwass@lifesci.lscf.ucsb.edu (Lloyd Goldwasser) writes: > First is yet another header problem. When I try to compile, I get > link errors due to undefined symbols: > > /bin/ld: Undefined symbols: > .objc_class_name_NSArray > .objc_class_name_NSBundle > ... > > although, for instance, every file that has NSArrays definitely > #imports <Foundation/NSArray.h>. The #import brings in the declarations for your source code, but not the implementations. To get the implementations, you'll need to get your app to link with the appropriate frameworks. In ProjectBuilder you can double-click on the 'Frameworks' line in the project browser, and use the resulting 'Open' panel to add the Foundation framework (where NSArray and NSBundle are implemented). Alternatively, you can manually put the needed frameworks on the OTHER_LIBS variable in your Makefile.preamble: # Additional libs to link apps against ('app' target) OTHER_LIBS = -framework Foundation If you ar using home-grown makefiles, or just typing on the commandline: cc -o Foo -O -arch m68k -arch sparc Foo.m -framework Foundation > The list of undefined symbols > encompasses even things like _SNDAlloc, although I never go near > anything to do with sound. Something you are doing might be indirectly referencing sound code. Use the following frameworks to bring in sound support along with everything else. -framework SoundKit -framework AppKit -framework Foundation > Second, the NAN problem that I mentioned still persists. Sorry , but I don't have any cluse on this one. -- Mike Paquette (mpaque@wco.com) 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: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: hit test in postscript Date: 14 Jun 1997 19:02:40 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5nupsg$ep4$1@brachio.zrz.TU-Berlin.DE> References: <5nub43$k0t$1@imag.imag.fr> In article <5nub43$k0t$1@imag.imag.fr> Benhur.Stein@imag.fr (Benhur Stein) writes: > In <5nu4cc$qte$1@brachio.zrz.TU-Berlin.DE> marcel@system.de wrote: > > In article <5nsbpv$qgq$1@imag.imag.fr> Benhur.Stein@imag.fr (Benhur Stein) > > writes: > > > > [Hit testing on OS 4.1/Intel fails] > > > > > newpath > > > 10000 34 moveto > > > -2 6 rlineto > > > 4 0 rlineto > > > closepath > > > 10000 38 infill > > > > This infil succeeds on my 4.0/Intel system. Maybe I shouldn't have > > ordered 4.2 yesterday? > Well, I was testing with BeYap; with Yap it works. But changing 10000 with > 30000, it > doesn't work (that was my original problem, I simplified to 10000 to have > more > beautiful numbers..) OK, I did some more tests and the cross-over is between 16001 and 16002. Strange. Also, it seems to depend on the absolute device coordinate. The only thing I can think of is that 16000 is near the maximum size for windows, so maybe the rationale (or bug) is that the point cannot possibly be filled if it is outside the window's bounds. Still seems strange. Marcel
From: goldsmith@apple.com (David Goldsmith) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Sat, 14 Jun 1997 12:25:28 -0700 Organization: Apple Computer, Inc. Message-ID: <goldsmith-ya02408000R1406971225280001@news.apple.com> References: <5mvclb$jb2$1@news.digifix.com> <AFB89A0D-2FF75B@206.165.44.70> <5nmf6e$ffg$1@news.intercon.com> <goldsmith-ya02408000R1206971752090001@news.apple.com> <davep-1406970816330001@dialup75.mn.uswest.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit In article <davep-1406970816330001@dialup75.mn.uswest.net>, davep@nospam.com (Dave Polaschek) wrote: >In article <goldsmith-ya02408000R1206971752090001@news.apple.com>, >goldsmith@apple.com (David Goldsmith) wrote: >> In article <5nmf6e$ffg$1@news.intercon.com>, "Amanda Walker" >> <amanda_walker@ascend.com> wrote: >> >GX Printing is dead, according to Apple. MacOS 8 will support GX graphics >> >for the screen but not for printers. >> No, not quite. MacOS 8 will support GX graphics for printers, just through >> the classic printing architecture. > >David, > >You need to be VERY careful about raising people's expectations here. >MacOS 8 will support GX graphics for printers, but only as bitmaps. >Support for PostScript generation will not be in 8.0. > >-DaveP Yes, that's correct. Unfortunately, the LaserWriter 8 driver that ships with Mac OS 8 will send GX data to Postscript printers as printer-resolution images (slow, but the output will appear correct). There wasn't time for a better solution. We are planning to move the GX-to-Postscript code that was in the LaserWriter GX driver into a library that can be used by the LaserWriter 8 driver, so that GX data gets sent as Postscript. Unfortunately, I can't say when that will be delivered; I hope it will be soon. Apple is kind of resource-constrained right now. We're trying. Really. -- David Goldsmith Architect International, Text, and Graphics Group Apple Computer, Inc. goldsmith@apple.com
From: "Dhani Nugraha" <dhani@dnet.net.id> Newsgroups: comp.sys.next.programmer Subject: I need help Date: 15 Jun 1997 00:37:08 GMT Organization: Voice Underground Daily Message-ID: <01bc7923$0c15c1e0$a20294ca@dhani.dnet.net.id> Please help me,i need a IBM Aptiva.Where is the place to buy it? Please reply. -- Dhani Nugraha
Date: Sat, 14 Jun 1997 19:51:41 -0600 From: Rene Berber <rberber@spin.com.mx> Subject: Re: nib/InterfaceBuilder.h Newsgroups: comp.sys.next.programmer Message-ID: <866335365.8210@dejanews.com> Organization: Deja News USENET Posting Service beauvois@usa.net wrote: > > I'm trying to port some code from NS 2.0/1 to 3.3 but can't figure > out what to replace the old <nib/InterfaceBuilder.h> header file > with. > > Anyone know ? Try #import <apps/InterfaceBuilder.h> --- Rene Berber rberber@spin.com.mx MIME / NeXT Mail welcomed -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: luomat@peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.programmer Subject: Re: wu-ftpd: link statically 'ls' etc ? Date: 14 Jun 1997 17:39:07 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Message-ID: <5nukvr$asn$1@bashir.peak.org> References: <Pine.NXT.3.96.970612112105.6213D-100000@cc344191-a> <wncOJzC00iWn07BvM0@andrew.cmu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: cs4w+@andrew.cmu.edu In <wncOJzC00iWn07BvM0@andrew.cmu.edu> Charles William Swiger wrote: > Excerpts from netnews.comp.sys.next.programmer: 12-Jun-97 wu-ftpd: link > statically 'l.. by "Timothy J. Luoma"@peak. > > wu-ftpd suggests statically linked binaries > > > > Can this be done under NS/OS (3.3/4.1)? > > Yes, but most likely you're going to have to do a lot more work than you > really want to. Ok.... it sounds like the alternative is to make the shlib's available on the ftp site itself..... For some systems this is not considered a safe plan. I hate to ask but any chance that OS4.1 is a safe one? Thanks TjL -- TjL <luomat@peak.org> / http://www.peak.org/~luomat/next/ NeXT bookmarks: http://www.peak.org/~luomat/next/bookmarks.html sed 's/End\ of\ sig/pithy\ quotation/g'
From: luomat@peak.org (Timothy J. Luoma) Newsgroups: comp.sys.next.programmer Subject: Re: Does OS4.2 finally fix the printf %g bug? Date: 14 Jun 1997 18:03:37 GMT Organization: The PEAK FTP Archive for NEXTSTEP/OpenStep Message-ID: <5numdp$asn$3@bashir.peak.org> References: <5npof5$jlf@tribune.usask.ca> <33A27FB5.6635@ergotech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: jim@ergotech.com In <33A27FB5.6635@ergotech.com> Jim Redman wrote: > Since 4.2 is still prerelease and still under NDA I was told (by NeXT) that 4.2 was publically available as of 13 May 1997. This message seems to have appeared at my news server in 14 June, but they have had some old posts come back from time to time, so I may just be behind on when this thread was around.... TjL -- TjL <luomat@peak.org> / http://www.peak.org/~luomat/next/ NeXT bookmarks: http://www.peak.org/~luomat/next/bookmarks.html sed 's/End\ of\ sig/pithy\ quotation/g'
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] Main executable in application folder Date: 15 Jun 1997 01:45:21 GMT Organization: WolfWare Message-ID: <5nvhfh$5fq$1@vader.wolfware.ipc.net> References: <866191282.19621@dejanews.com> <EBpyqL.61n@novice.uwaterloo.ca> In-Reply-To: <EBpyqL.61n@novice.uwaterloo.ca> On 06/13/97, David Evans wrote: >>Also, if I wish to make the main code >>say a Perl script or csh script does Open Step support the possibility >>and if it does, how does it know which execution environment to start? > > That should work fine, provided your script starts with the "#!/bin/csh" (or >whatever) magic cookie. The kernel understands such things. However, the >utility of this inside an app wrapper is questionable, since stdout will go to >the console and there's no stdin (as far as I know). I thought it would work fine too -- until I tried it. It didn't work under NeXTSTEP 3.3 for me. - Chris > >-- >David Evans (NeXTMail/MIME OK) dfevans@bbcr.uwaterloo.ca >Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/ >University of Waterloo "Default is the value selected by the composer >Ontario, Canada overridden by your command." - Roland TR-707 Manual > -- _______________________________________________________________________ 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: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: Version control in 4.2prerelease? Date: 15 Jun 1997 01:48:34 GMT Organization: WolfWare Message-ID: <5nvhli$5jr$1@vader.wolfware.ipc.net> References: <5ntj9k$ajg$1@owl> <5nubgn$d4r$1@concorde.ctp.com> In-Reply-To: <5nubgn$d4r$1@concorde.ctp.com> On 06/14/97, Georg Tuparev wrote: >By adding CVS or PVCS, making some hacks to IB and PB (mine will be submitted >one day), and cross your fingers. Being religios might also help especially if >you pray very hard before start using the SCM module ;-) > >In article <5ntj9k$ajg$1@owl> Dru Nelson <dnelson@slip.net> writes: >> >> >> How can one add RCS/whatever so project builder uses version >> control? Another option you may want to look into is DevMan from VNP (www.vnp.com) (although Devman was recently bought by Sun.) It doesn't offer full integration with DevMan but it's a very robust and nice to use GUI front-end for RCS under NeXTSTEP and OpenStep. - Chris > >-- >------- > /\/\ 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 > -- _______________________________________________________________________ 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: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: Version control in 4.2prerelease? Date: 15 Jun 1997 03:37:10 GMT Organization: WolfWare Message-ID: <5nvo16$6ub$1@vader.wolfware.ipc.net> References: <5ntj9k$ajg$1@owl> <5nubgn$d4r$1@concorde.ctp.com> <5nvhli$5jr$1@vader.wolfware.ipc.net> In-Reply-To: <5nvhli$5jr$1@vader.wolfware.ipc.net> On 06/14/97, Christopher Wolf wrote: >On 06/14/97, Georg Tuparev wrote: >>By adding CVS or PVCS, making some hacks to IB and PB (mine will be submitted >>one day), and cross your fingers. Being religios might also help especially if >>you pray very hard before start using the SCM module ;-) >> >>In article <5ntj9k$ajg$1@owl> Dru Nelson <dnelson@slip.net> writes: >>> >>> >>> How can one add RCS/whatever so project builder uses version >>> control? > >Another option you may want to look into is DevMan from VNP (www.vnp.com) >(although Devman was recently bought by Sun.) It doesn't offer full >integration with DevMan but it's a very robust and nice to use GUI ^^^^^^ d'oh I meant Project Builder here of course >front-end for RCS under NeXTSTEP and OpenStep. > -- _______________________________________________________________________ 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: 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: <14012865742421@digifix.com> Date: 15 Jun 1997 03:57:19 GMT Organization: Digital Fix Development Message-ID: <3108866347221@digifix.com> Topics include: Stepwise NEXTSTEP/OpenStep Information WWW site eduSTEP WWW site NeXT Computer, Inc. WWW site comp.sys.next newsgroups related newsgroups comp.sys.next newsgroups mailing list ftp sites NeXTanswers Stepwise NEXTSTEP/OpenStep Information WWW site =============================================== This online community resource includes - ISV company pages - ISV product descriptions - NEXTSTEP Developer Directory - NEXTSTEP Community WhitePages - Mailing List archives and information You can connect via the world wide web at: http://www.stepwise.com/ Suggestions or comments can be directed to me at sanguish@digifix.com If you would like to get your company and product information on Stepwise, please contact me at sanguish@digifix.com. eduSTEP WWW site ================ http://www.nmr.embl-heidelberg.de/eduStep/ eduStep aims to provide up-to-date information on: - NextStep tools and projects for scientists. - Third-party products interesting for the educational and scientific community (with educational discounts noted, where they exist). - A listing of resellers and shops interested in working with customers in the educational community. - Conferences, meetings, workshops - Major projects, such as SciTools, EMBL's project to develop a NextStep scientific work environment - Status reports on GNUStep, a freely-available implementation of OpenStep now being developed NeXT Computer, Inc. WWW site ============================ http://www.next.com comp.sys.next.* newsgroups ========================== news: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. news: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. news: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. news: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. news: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. news: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!!! news: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. news:comp.sys.next.software This is a place to talk about [third party] software products that run on NEXTSTEP systems. news:comp.sys.next.sysadmin Stuff relating to NeXT system administration issues; in rare cases this will spill over into .programmer or .software. Related Newsgroups ================== news: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. news:comp.lang.objective-c Technical talk about the Objective-C language. Implemetations discussed include NeXT, Gnu, Stepstone, etc. news: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 The ftp sites ============= ftp://ftp.next.peak.org - The main site for North American submissions formerly ftp.cs.orst.edu ftp://ftp.informatik.uni-muenchen.de: - (Peanuts) Located in Germany. ftp://ftp.dn.net/pub/next - Peanuts mirror in the US ftp://terra.stack.urc.tue.nl - (Dutch NEXTSTEP User Group) 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: "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: GX OOP class proposal & contest (was Re: GX game graphics Date: 14 Jun 1997 19:30:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFC8A645-31EFC@206.165.44.17> References: <goldsmith-ya02408000R1406971225280001@news.apple.com> To: "David Goldsmith" <goldsmith@apple.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 David Goldsmith <goldsmith@apple.com> said: > Yes, that's correct. Unfortunately, the LaserWriter 8 driver that ships > with Mac OS 8 will send GX data to Postscript printers as > printer-resolution images (slow, but the output will appear correct). There > wasn't time for a better solution. We are planning to move the > GX-to-Postscript code that was in the LaserWriter GX driver into a library > that can be used by the LaserWriter 8 driver, so that GX data gets sent as > Postscript. Unfortunately, I can't say when that will be delivered; I hope > it will be soon. Apple is kind of resource-constrained right now. We're > trying. Really. This is good to hear. Keep in mind the possibility for at least SOME GX Printing like functinality by providing a pipeline with hooks for 3rd parties to add GX Display List editing and so on. The ideal solution would be to add GX Print driver information to the classic drivers that only the GX print library routines would know about. This would give GX applications at least some of the functionality that they are loosing. Don't make this a private revision closed to the public. Let the GX developers and wannabe developers and users provide feedback and input on GX-Talk about what kind of functionality *could* be preserved and how. ++++++++++++++++++++++++++++++ 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: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer Subject: Re: Does OS4.2 finally fix the printf %g bug? Date: 15 Jun 1997 04:06:02 GMT Organization: Digital Fix Development Message-ID: <5nvpna$hro$1@news.digifix.com> References: <5npof5$jlf@tribune.usask.ca> <33A27FB5.6635@ergotech.com> <5numdp$asn$3@bashir.peak.org> In-Reply-To: <5numdp$asn$3@bashir.peak.org> On 06/14/97, Timothy J. Luoma wrote: >In <33A27FB5.6635@ergotech.com> Jim Redman wrote: > >> Since 4.2 is still prerelease and still under NDA > >I was told (by NeXT) that 4.2 was publically available as of 13 May 1997. > >This message seems to have appeared at my news server in 14 June, but they >have had some old posts come back from time to time, so I may just be behind >on when this thread was around.... > 4.2 was 'released' the first week of June according to something that was mailed to EAPs.. -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer Subject: Re: Version control in 4.2prerelease? Date: 15 Jun 1997 04:07:28 GMT Organization: Digital Fix Development Message-ID: <5nvpq0$hrq$1@news.digifix.com> References: <5ntj9k$ajg$1@owl> <5nubgn$d4r$1@concorde.ctp.com> <5nvhli$5jr$1@vader.wolfware.ipc.net> In-Reply-To: <5nvhli$5jr$1@vader.wolfware.ipc.net> On 06/14/97, Christopher Wolf wrote: >On 06/14/97, Georg Tuparev wrote: >>By adding CVS or PVCS, making some hacks to IB and PB (mine will be submitted >>one day), and cross your fingers. Being religios might also help especially if >>you pray very hard before start using the SCM module ;-) >> >>In article <5ntj9k$ajg$1@owl> Dru Nelson <dnelson@slip.net> writes: >>> >>> >>> How can one add RCS/whatever so project builder uses version >>> control? > >Another option you may want to look into is DevMan from VNP >(www.vnp.com) (although Devman was recently bought by Sun.) It >doesn't offer full integration with DevMan but it's a very robust >and nice to use GUI front-end for RCS under NeXTSTEP and OpenStep. > I'd be surprised if Sun is still selling DevMan on OpenStep... After all, OpenStep isn't Java.. -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: beatty@netcom.com (Derek Lee Beatty) Newsgroups: comp.sys.mac.graphics,comp.sys.next.programmer,comp.sys.mac.comm,comp.sys.mac.programmer.misc,comp.sys.mac.misc,comp.sys.next.misc Subject: Re: Rhapsody features? Date: 14 Jun 1997 04:40:16 GMT Organization: none Message-ID: <5nt7bh$9q0@beatty.slip.netcom.com> References: <5lvn4l$tje@src-news.pa.dec.com> <3385FA0F.5439@abacus.com> <5nnres$390$2@kaopala.mhpcc.edu> decoy_id@no_junk_on_the.net (L e e Altenberg) wrote: >Could anyone explain why an OS optimized for use as a server would not >also be optimal for a workstation? To be concise but rather general: - On a workstation, you want to respond to the user with low latency. - On a server, you want to maximize overall throughput. -- Derek Lee Beatty _ Death beatty@netcom.com _| ~-, Taxes Austin, Texas \, * } C++ \_(
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: GX OOP class proposal & contest (was Re: GX game graphics Date: 14 Jun 1997 13:17:06 -0700 Organization: Primenet Services for the Internet Message-ID: <AFC84ECE-E467@206.165.44.37> References: <davep-1406970816330001@dialup75.mn.uswest.net> 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 Dave Polaschek <davep@nospam.com> said: goldsmith@apple.com (David Goldsmith) wrote: > > No, not quite. MacOS 8 will support GX graphics for printers, just through > > the classic printing architecture. > > David, > > You need to be VERY careful about raising people's expectations here. > MacOS 8 will support GX graphics for printers, but only as bitmaps. > Support for PostScript generation will not be in 8.0. Which is just plain stupid. The GX => PS translation algorithms already exist. In fact, a substantial portion of the functionality of GX Printing could be re-implemented as a set of shared libraries and components that printed to the Classic Printing model rather than to individual drivers. WIll Apple do this? No. Why? A combination of money and politics. GX Printing is far, FAR more useful than what Rhapsody will provide for the forseeable future. It isn't good for Gil Amelio's career if non-RHapsody printing offers more power than Rhapsody printing. Apple has essentially crippled its "Windows 95" OS in order to make its "Windows 95" OS look good in comparison. Not so impressive, Dr. Gil. ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
Date: Sat, 14 Jun 1997 06:51:42 -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: GX OOP class proposal & contest (was Re: GX game graphics Message-ID: <joe.ragosta-ya02408000R1406970651420001@news.dol.net> References: <AFC74CB8-1F634@206.165.44.35> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-transfer-encoding: 8bit In article <AFC74CB8-1F634@206.165.44.35>, "Lawson English" <english@primenet.com> wrote: > This reply was sent to me from Joel Breckinridge, who doesn't have > newsgroup access: > > Joel Breckinridge <velvetj@tactgraphic.co.jp> said: > > > > > While this is true, it does not answer the fact that PostScript (I'm not > > talking DPS here) is insufficient for Japanese DTP needs or any 2 byte > > needs. I have been with it since the start in 1989 and Japanese PS is > > difficult to kern, mutiple weights are large, unweildy and take up too > > much memory, this is just for starters. As I said earlier on the GX list: > > > > 'If anybody cares to know, there are no Mutiple Master Japanese fonts, > > (or any 2 byte MM fonts, but there are some GX Japanese fonts) and there > > is no way to embed most Japanese type 1 fonts in the recently released > > Acrobat 3.0 J (the first to support 2 byte characters). Outside of GX, 2 > > byte PDA is a kludge and sucks, to put it mildly.' Snip, snip, snip. But isn't this all just another red herring since Apple has announced that GX typography will make it to Tempo? -- Regards, Joe Ragosta joe.ragosta@dol.net Visit the Complete Macintosh Web Site http://www.dol.net/~Ragosta/complmac.htm
From: 38433@ix93384.net Newsgroups: comp.sys.next.programmer Subject: Cheerleaders are so sweet...! Date: 15 Jun 1997 08:58:38 GMT Organization: Pinnacle Online Message-ID: <5o0aru$go1$152@nw001.infi.net> Hey, just thought i'd share with everyone, I found a site with loads of nude CHEERLEADERs. The address is: http://www.mid-night.com --Jason-- (Sorry for the intrusion, everyone needs some short skirts in their life) P.S. They also a have spanish / latina women archive
From: 38433@ix93384.net Organization: Pinnacle Online Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5o0aru$go1$152@nw001.infi.net> Message-ID: <cancel.5o0aru$go1$152@nw001.infi.net> Control: cancel <5o0aru$go1$152@nw001.infi.net> References: <5o0aru$go1$152@nw001.infi.net> Date: Sun, 15 Jun 1997 11:04:12 +1 EMP/ECP spam cancelled by hweede@berlin.snafu.de. This is an ongoing spam whose Breidbart index already is above 20. See my report "www.mid-night.com" or "summary of auto-cancellations" in news.admin.net-abuse.bulletins. Subject was: Cheerleaders are so sweet...!.
From: szallies@gehirnbrei.energotec.de (Constantin Szallies) Newsgroups: comp.sys.next.programmer Subject: Re: DO question: trapping client deaths Date: 9 Jun 1997 16:12:45 GMT Organization: Technet GmbH Message-ID: <5nha1t$s70$1@oxygen.technet.net> References: <EB68yM.9I0@gateway.ali.bc.ca> frank@ali.bc.ca (Frank Pang) wrote: >In NS3.3: >I have a program using DO where a client can send messages for the server >to do some fetching and processing. In the meantime, there's a >possibility that the client dies before the server is able to send a reply >to the client. In this scenario, the server will try to send a message to >a dead client, causing [myNXConnection run] to exit. Is this an exception >condition? It doesn't seem to raise exceptions. What's a clean way to >handle client deaths / invalid connections? And where can I get some >sample code for this? Thanks. What do you mean by "sending a reply" to the client? There are two possibilities: 1) The client calls a remove method on the server. The client dies before the invokation is completed. In this case the connection should not exit! The reply is just thrown away. 1) The client calls a remove method on the server, the server does a callback to the client and the cleint dies. The server receives an exception which must be caught by the server. Greetings CS
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: GX OOP class proposal & contest (was Re: GX game graphics Date: 15 Jun 1997 11:57:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFC98D96-5A1DE@206.165.44.11> References: <joe.ragosta-ya02408000R1406970651420001@news.dol.net> 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 Joe Ragosta <joe.ragosta@dol.net> said: [discussion of GX font features that make them superior to PostSCript] > > Snip, snip, snip. > > But isn't this all just another red herring since Apple has announced that > GX typography will make it to Tempo? But FontWorks won't be able to use its compressed Japanese fonts because Rhapsody's implementation of GX Typography won't allow plug-in renderers in [at least] the first release, and no time-table has been set to include this feature (not surprising -why should Adobe allow anyone to compete with them on type fonts?). Since Japanese fonts can take up to 6 MB, whereas FOntWork's custom Japanese fonts only take up 1.6 MB, this means that Rhapsody will require nearly 4x the space for Japanese fonts that GX would since FontWorks has a custom GX rendering engine for displaying these fonts. When you add to the equation the fact that GX allows one to parse the actual characters used in a print job and download to a printer ONLY those characters that are actually used on a page whereas Rhapsody apparently won't support GX-like functionality in this area (since it requires a GX-like display list to be the standard way in which print jobs are created), this means that Rhapsody printing of Japanese won't be as efficient or as fast as GX printing. WHich means that Amanda's claim that GX wasn't as worthwhile for Japanese DTP makes no sense in this context. -Especially since Apple ships Ready, Set, Go GX as THE desktop publishing app bundled with every Macintosh sold in China, which I'm willing to be has much of the same requirements for DTP apps as Japan. Obviously Apple felt that GX offered a distinct advantage for Chinese DTP or they would have bundled a PostSCript-using DTP app, right? Face it, Joe, as Tom Bayly said on Carpe.Diem: there's no doubt which is a better graphics system. And it is NOT Display PostScript. Or, as Steve Wozniak said in an interview: "They may think that they've got the better technology, but they don't." And the Woz's official job at Apple right now is to seek out the best technologies available, so he is speaking directly to his official "expertise" at Apple (of course, you can argue that Steve Wozniak really isn't competent to evaluate Apple vs Other technologies, but that's what they say that they're using him for, so they appear to be ignoring the advice of their own in-house expert about technologies, which I'm guessing includes GX/Taligent vs Display PostSCript, since it is so obviously the case). Don't worry, Joe. Very shortly, you too can become totally incensed at the lack of willingness to use superior Apple technologies over NeXT technologies. You're going to love the abilities of GX printing that Apple is giving up with MacOS 8 that will be IMPOSSIBLE to implement in Rhapsody without decreeing and enforcing some kind of GX-like shape-type strategy. You may even be inspired to write a letter to Apple complaining about your loss. If you've ever been able to express disapproval of anything that ANY current Apple management team has ever done, of course. BTW, Joe, GX Typography includes the ability to assign a custom layer to a type font that may include a 3D perspective. Since Display PostSCript doesn't support 3D perspectives, how is it possible for RHapsody Graphics to implement ALL of GX Typography? ANd if it isn't possible, how can you automatically assume that any specific (and possibly valueable) feature of GX Typography will make it into Rhapsody Graphics? Blind loyalty is, well, blind... ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: attainment <abiogen@abiogenesis.com> Newsgroups: comp.sys.next.programmer,comp.sys.amiga.programmer,comp.sys.be.programmer Subject: Free online pubication - Attainment adding Software Developers' column Date: Sun, 15 Jun 1997 12:27:58 -0700 Organization: CalWeb Internet Services, Inc. Message-ID: <Pine.BSF.3.95q.970615122406.4073B-100000@web2.calweb.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII NNTP-Posting-User: ident=abiogen Subject: Free online pubication - Attainment adding Software Developers' column I thought programmers who are software entrepreneurs would appreciate knowing that we've added a column to our free online publication that will deal with software issues, and how to build a business selling software online and off. It's located at http://www.abiogenesis.com/attainment Look for the article labelled Selling Your Software. The first installment is written by Julie Benson, who has a decidedly 'individual' take on these things. More are in the works, and if you're interested in contributing, writers' guidelines are listed there, as well. Attainment Editor =========================================================================== ATTAINMENT Business Startup Journal(tm) A N A B I O G E N E S I S P U B L I C A T I O N A practical, how-to resource for business owners and entrepreneurs who are seeking day-to-day financing, production and marketing resources for establishing and running their startup businesses. http://www.abiogenesis.com/attainment
From: Benhur.Stein@imag.fr (Benhur Stein) Newsgroups: comp.sys.next.programmer Subject: Re: hit test in postscript Date: 15 Jun 1997 19:07:02 GMT Organization: IMAG, Grenoble, France Distribution: world Message-ID: <5o1egm$6t$1@imag.imag.fr> References: <5nub43$k0t$1@imag.imag.fr> <5nupsg$ep4$1@brachio.zrz.TU-Berlin.DE> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit Cc: marcel@system.de In <5nupsg$ep4$1@brachio.zrz.TU-Berlin.DE> marcel@system.de wrote: > In article <5nub43$k0t$1@imag.imag.fr> Benhur.Stein@imag.fr (Benhur Stein) > writes: > > In <5nu4cc$qte$1@brachio.zrz.TU-Berlin.DE> marcel@system.de wrote: > > > In article <5nsbpv$qgq$1@imag.imag.fr> Benhur.Stein@imag.fr (Benhur > Stein) > > > writes: > > > > > > [Hit testing on OS 4.1/Intel fails] > > > > > > > newpath > > > > 30000 34 moveto > > > > -2 6 rlineto > > > > 4 0 rlineto > > > > closepath > > > > 30000 38 infill > OK, I did some more tests and the cross-over is between 16001 and 16002. > > [...] > Still seems strange. > What I think is the most strange is that it actually draws to the screen with these same values! Benhur
From: bchin@NOSPAM.richmond.freedomnet.com (Bill Chin) Newsgroups: comp.sys.next.programmer Subject: Re: Does OS4.2 finally fix the printf %g bug? Date: 16 Jun 1997 04:25:28 GMT Organization: FreedomNet - Your Full Service Internet Provider Message-ID: <5o2f7o$2d7$2@news.freedomnet.com> References: <5npof5$jlf@tribune.usask.ca> eric@skatter.USask.Ca wrote: >Would someone out there running OPENSTEP 4.2 try my famous printf >test program? With the released version of OPENSTEP 4.2/Mach for Intel: nikita> cc -v Reading specs from /lib/i386/specs NeXT Software, Inc. version cc-744.13, gcc version 2.7.2.1 nikita> a.out Expect 0.00123: 0.001 Expect 123: 123.457 Expect 123.5: 123.4567 Expect 1e+03: 999.6 This is with a fresh install. -- Bill Chin - bchin@richmond.freedomnet.com - NeXTmail/MIME welcomed
From: taustin@ozemail.com.au (T. Austin) Newsgroups: comp.sys.next.programmer Subject: Australian OpenStep commercial developers Date: Mon, 16 Jun 1997 16:04:39 +1100 Organization: Swinburne University Message-ID: <taustin-1606971604400001@d185-1.cpe.melbourne.aone.net.au> If there are any Australian developers out there. I'd like to hear from you.
From: gary-nospam-@screaming.org (Gary W. Longsine) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.programmer Subject: Re: Prelude2Rhapsody on black hardware! Date: 16 Jun 1997 15:43:32 GMT Organization: Save the Skeet Foundation Message-ID: <5o3mv4$rh7$1@news.platinum.com> References: <rbarris-ya023280001106971149270001@news.intelenet.com> Cc: rbarris@quicksilver.com In <rbarris-ya023280001106971149270001@news.intelenet.com> it appeared that Rob Barris wrote: > I finally got around to trying the Prelude OpenStep 4.2 CDROM on my > 68040 NextStation Turbo. After mounting the disc, running Upgrader.app and > then clearing off some disk space by its advice, the install worked! > > After having heard from any number of people "nahhh, it's an Intel only > release" it was a pleasant surprise indeed to see that the discs from WWDC > were actually fat builds for both Intel and "black" NeXT hardware. (So far > I've only installed the User side FWIW) > > However I only have a 240MB HD in that machine, and now it's full. Looks > like I need to swap in a 1GB drive or something before I continue with > installing the "Developer" disc. Hi Rob, You may know this already, but when slipping a new disk into a NeXTstation, remember to stay away from the newest (7200RPM) disks like the SeaGate Baracuda. It's an excellent disk, but it's too hot for the machine. If you keep the room temperature very cool (less than 70 or so) it would be fine, but if the room temp gets up to about 80, you'll probably start to see kernel panics which result from an overheated CPU. /gary -- Gary W. Longsine, Systems Engineer | ____/| OpenStep MachOS PLATINUM technology, inc. | \ o.O| Objective-C l_o_n_gsine@platinum.com (NeXTmail | =(_)= the Dock (Can i have his spam?) & MIME) |. U Elegance is Relevant.
From: droege@informatik.uni-koblenz.de (Detlev Droege) Newsgroups: comp.sys.next.programmer Subject: Re: Version control in 4.2prerelease? Date: 16 Jun 1997 17:19:58 GMT Organization: University Koblenz / Germany Distribution: world Message-ID: <5o3sju$l4n$1@newshost.uni-koblenz.de> References: <5ntj9k$ajg$1@owl> In article <5ntj9k$ajg$1@owl> Dru Nelson <dnelson@slip.net> writes: > > > How can one add RCS/whatever so project builder uses version > control? Have a look in /NextLibrary/Documentation/NextDev/ReleaseNotes/ProjectBuilderSCM.rtfd Detlev -- Detlev Droege, Uni Koblenz, FB Informatik, Rheinau 1, D-56075 Koblenz, Germany Tel:+49 261 9119-421,Fax:-497,NeXT/MIME/Emil: droege@informatik.uni-koblenz.de C++ is the only current language making COBOL look good. --Bertrand Meyer
From: rbarris@quicksilver.com (Rob Barris) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.programmer Subject: Re: Prelude2Rhapsody on black hardware! Date: Mon, 16 Jun 1997 10:56:06 -0700 Organization: Quicksilver Software, Inc. Message-ID: <rbarris-ya023280001606971056060001@news.intelenet.com> References: <rbarris-ya023280001106971149270001@news.intelenet.com> <5o3mv4$rh7$1@news.platinum.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit In article <5o3mv4$rh7$1@news.platinum.com>, gary-nospam-@screaming.org (Gary W. Longsine) wrote: > In <rbarris-ya023280001106971149270001@news.intelenet.com> it appeared that > Rob Barris wrote: > > I finally got around to trying the Prelude OpenStep 4.2 CDROM on my > > 68040 NextStation Turbo. After mounting the disc, running Upgrader.app and > > then clearing off some disk space by its advice, the install worked! > > > > After having heard from any number of people "nahhh, it's an Intel only > > release" it was a pleasant surprise indeed to see that the discs from WWDC > > were actually fat builds for both Intel and "black" NeXT hardware. (So far > > I've only installed the User side FWIW) > > > > However I only have a 240MB HD in that machine, and now it's full. Looks > > like I need to swap in a 1GB drive or something before I continue with > > installing the "Developer" disc. > > Hi Rob, > > You may know this already, but when slipping a new disk into a NeXTstation, > remember to stay away from the newest (7200RPM) disks like the SeaGate > Baracuda. It's an excellent disk, but it's too hot for the machine. If you > keep the room temperature very cool (less than 70 or so) it would be fine, > but if the room temp gets up to about 80, you'll probably start to see kernel > panics which result from an overheated CPU. > Thanks Gary - yes, looking inside the NeXT machine I was reminded of the Power Mac 6100 in terms of "close quarters" spacing and airflow, definitely not the best place for one of those scorcher drives. BTW I have a Conner 1080S; it came out of my 8500 and it doesn't play friendly when attached to the external NeXT bus (parity errors on boot, system can't talk to the primary internal drive, etc). I found the Conner PDF file for this drive but there do not appear to be any jumpers for parity or anything else besides SCSI ID... any ideas? The next thing I will try is pulling the terminating resistor packs off of the Conner and see if that makes any difference. (I only paid $450 for the NextStation-Turbo-mono, something about paying $200 for a new 1GB drive to go with it rubs me the wrong way, but if I can't get this old drive working with it I might just give up and do it) Rob Barris Quicksilver Software Inc. rbarris@quicksilver.com * Opinions expressed not necessarily those of my employer *
From: mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 16 Jun 1997 13:23:42 GMT Organization: P&L Systems Message-ID: <5o3eou$rkt$1@ironhorse.plsys.co.uk> References: <davep-1406970816330001@dialup75.mn.uswest.net> <AFC84ECE-E467@206.165.44.37> In-Reply-To: <AFC84ECE-E467@206.165.44.37> On 06/14/97, "Lawson English" wrote: >A combination of money and politics. GX Printing is far, FAR more useful >than what Rhapsody will provide for the forseeable future. It isn't good >for Gil Amelio's career if non-RHapsody printing offers more power than >Rhapsody printing. > What drivel. We have discussed ad nauseum why it makes sense for Rhapsody to use DPS; you've conceded yourself (unless I misunderstood) that it would not be in AppLE's interests to delay Rhapsody's release until GX is debugged and shoehorned in. I look forward to seeing enhanced GX-like/derived capabilities in Rhapsody *in the fullness of time*. Best wishes, mmalc. --
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 16 Jun 1997 13:35:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFCAF619-128736@206.165.44.75> References: <5o3eou$rkt$1@ironhorse.plsys.co.uk> 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 mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> said: > We have discussed ad nauseum why it makes sense for Rhapsody to use > DPS; > you've conceded yourself (unless I misunderstood) that it would not be in > AppLE's interests to delay Rhapsody's release until GX is debugged and > shoehorned in. > > I look forward to seeing enhanced GX-like/derived capabilities in Rhapsody > *in the fullness of time*. But when? And in the meantime, GX is being crippled on MacOS as of MacOS 8. Here's a [partial] list of what you will NOT see with Rhapsody graphics in the near future, as far as I can tell: 3D perspective for shapes, collections of shapes and views. 3D perspective for individual glyphs in a line of text. Transparency modes using Porter-Duff compositing on a PER COLOR CHANNEL basis. Retained mode interface for both text and graphics without using a kludgey, disorganized, pseudo-object-oriented class design that doesn't acknowledge that a bitmap is a first class graphical shape, but instead implements it in an entirely different part of the framework. Ditto with text. No text-oriented hit-testing that makes use of a shape's transform matrix automatically. No auto-hit-testing of glyphs during font rendering to make sure that Hoefler Italic's more complicated glyphs aren't overlapping with the rest of a text-string and allowing for simplified font-substitution when such a case occurs. Etc. Etc. Etc. Now, on the MacOS side, MacOS 8 loses ALL of GX Printing. That means that an application can no longer grab the print job before it is spooled to disk and make further modifications. That means that GX Printing extensions can't grab the job AFTER it is spooled to disk and make automatic modifications to the display-list-based GX Print job. There are over 100 messages defined that applications and extensions can override to modify the print stream. Since there is a well-defined, *SIMPLE TO EDIT*, universally available display-list for GX printing used in all GX Printing, one can do just about anything that one desires to that print stream. As long as it can be automated using GX graphics/typography calls, it can be accomplished. If GX Extensions can send/receive AppleSCript messages, you can even alter the contents of a print job, IN THE PRINT STREAM, from within my HyperCard stack (not sure if this last is possible -depends on the AE-ability of GX printing extensions -HyperCard can certainly handle it). The irony of all this is that OpenStep's dynamic messaging would make all of this MUCH easier than the kludgey System 7.x way of doing things. You could take GX drivers and make the info available to Rhapsody graphics to allow for a much more flexable printing system that what is currently possible using GX. But you need a well-defined display list that stays constant throughout the entire print process. You can't do GX-like things automatically using EPS and Display PostScript. Presumably they are aware of this and are working on it, but given statements from Apple management like "Java does what OpenDoc does" and so on, I'm not convinced that they have a clue. Like as not, they'll use PDF and call the resultant subset of GX Printing capabilities "a miracle of Rhapsody engineering." And regardless of what Rhapsody can do, that doesn't help ANY current MacOS user for the next 12-18 months. GX, until MacOS 8, can do things like print to etching machines. Print transparency to pen plotters. Redirect print output from one page type to another as needed. Embedd multiple layers of transparent water marks in different layers of the graphics/text on the page. Apply 3D perspective to designated elements of the page. Print to level 1 PostScript printers using Level 2-type features. Apply multiple printing extensions in a user-sequenced cascade to a single print job. Print to any device resolution up to 20" on -a-side with 1 million DPI. Etc. As I said: >A combination of money and politics [killing GX Printing]. GX Printing is far, >FAR more useful >than what Rhapsody will provide for the forseeable future. It isn't good >for Gil Amelio's career if non-RHapsody printing offers more power than >Rhapsody printing. And so they cripple GX Printing to make Rhapsody look better. Sorry Gil, but it ain't going to work. ++++++++++++++++++++++++++++++ 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: alvin@cse.ucsc.edu (Alvin Jee) Newsgroups: comp.sys.next.programmer Subject: Re: hit test in postscript Date: 16 Jun 1997 20:15:07 GMT Organization: UC Santa Cruz CS/CE Message-ID: <5o46sb$jvk@darkstar.ucsc.edu> References: <5nsbpv$qgq$1@imag.imag.fr> <5nu4cc$qte$1@brachio.zrz.TU-Berlin.DE> <5nub43$k0t$1@imag.imag.fr> In article <5nub43$k0t$1@imag.imag.fr>, Benhur Stein <Benhur.Stein@imag.fr> wrote: >Well, I was testing with BeYap; with Yap it works. But changing 10000 with >30000, it >doesn't work (that was my original problem, I simplified to 10000 to have >more >beautiful numbers..) Ahhh.. there's the problem. I ran into this problem also. I found this note in the PSOperators man page (GeneralRef/05_DisplayPS/Operators): Repositions and resizes the specified window, effectively allowing it to be resized from any corner or point. x, y, width, and height are given in the screen coordinate system. No portion of the repositioned window can have an x or y coordinate with an absolute value greater than 16000; width and height must be in the range from 0 to 10000. The four operands can be integer or real numbers; however, they are converted to integers in the Window Server by rounding toward 0. This was from the "placewindow" operator. -- Alvin Jee alvin@neander.com http://www.neander.com NeXTMail gleefully accepted!
From: gtupar@ctp.com (Georg Tuparev) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] Main executable in application folder Date: 16 Jun 1997 23:18:20 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5o4hjs$h6q$1@concorde.ctp.com> References: <5nul7q$asn$2@bashir.peak.org> In article <5nul7q$asn$2@bashir.peak.org> luomat@peak.org (Timothy J. Luoma) writes: > > Am I nutty in the head or do I have some vague recollection of something in > 4.x that allowed a script to be wrapped as an app? Probably you've read the stuff about how to wrap a script or any other (Unix) utility in a service?!?! This is really nifty. later -- 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: "Robert H. Mann" <bobmann@anv.net> Newsgroups: comp.sys.next.programmer Subject: Device Driver Dilemma Date: Mon, 16 Jun 1997 18:38:09 +0000 Organization: All USENET -- http://www.Supernews.com Message-ID: <33A58811.79A5@anv.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I would like to run OS 4.2 on my PowerMac 6100 that has a 486DX processor in it but the right SCSI drivers don't exist to my knowledge. Any tips on making these beasts. I can run OS entreprise on Win95 if I have to but that sounds pretty disgusting. Thanks so much Robert H. Mann Director of Technology Nevada Wallboards
From: mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 16 Jun 1997 22:12:21 GMT Organization: P&L Systems Message-ID: <5o4do5$77s$1@ironhorse.plsys.co.uk> References: <5o3eou$rkt$1@ironhorse.plsys.co.uk> <AFCAF619-128736@206.165.44.75> In-Reply-To: <AFCAF619-128736@206.165.44.75> On 06/16/97, "Lawson English" wrote: >But when? And in the meantime, GX is being crippled on MacOS as of MacOS 8. > Are you sure? >Here's a [partial] list of what you will NOT see with Rhapsody graphics in >the near future, as far as I can tell: > GOOD! This means that AppLE will be concentrating on what they should be doing -- getting Rhapsody out of the door as quickly as possible to address the market sector to which it is relevant. It has clearly escaped your attention that this is, primarily, *not* AppLE's traditional marketplace, it's the enterprise market, which in general cares little about the multimedia aspects of their system, and more about stability and productivity. Really, get a grip. Get a clue. And get out of csna. Please. mmalc. --
From: dirk@object-factory.com (Dirk Olmes) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] Main executable in application folder Date: 17 Jun 1997 06:08:29 GMT Organization: Object Factory GmbH (Germany) Message-ID: <5o59kt$brt$1@leonie.object-factory.com> References: <866191282.19621@dejanews.com> <EBpyqL.61n@novice.uwaterloo.ca> <5nvhfh$5fq$1@vader.wolfware.ipc.net> cwolf@wolfware.com (Christopher Wolf) wrote: > On 06/13/97, David Evans wrote: > > >>Also, if I wish to make the main code > >>say a Perl script or csh script does Open Step support the possibility > >>and if it does, how does it know which execution environment to start? > > > > That should work fine, provided your script starts with the > > "#!/bin/csh" (or > >whatever) magic cookie. The kernel understands such things. > >However, the utility of this inside an app wrapper is questionable, > >since stdout will go to the console and there's no stdin (as far > >as I know). > > I thought it would work fine too -- until I tried it. It didn't work under > NeXTSTEP 3.3 for me. It never worked for me. Neither does it in 4.1. If I double click the "app" the console shows Jun 17 08:06:30 Workspace: Cannot exec /tmp/foo.app: (not a valid program) :-( I would have loved to have such a scripted app. -dirk --- ______________________________________________________________________ Dirk Olmes OBJECT FACTORY Gesellschaft fuer Informatik und Datenverarbeitung mbH Otto-Hahn Str. 18, 44227 Dortmund, Germany Telephon +49 (0) 231 975 137 0 Telefax +49 (0) 231 975 137 99 dirk@object-factory.com http://www.object-factory.com/ Hiroshima 45, Tschernobyl 86, Windows 95
From: dirk@object-factory.com (Dirk Olmes) Newsgroups: comp.sys.next.programmer Subject: Re: Hair Pulling with my first stab at IB! (please help :-)) Date: 17 Jun 1997 06:17:32 GMT Organization: Object Factory GmbH (Germany) Message-ID: <5o5a5s$brt$2@leonie.object-factory.com> References: <5nev5t$bqg$1@news.sas.ab.ca> <5ng6au$acj$1@leonie.object-factory.com> <5nqipp$n94@mochi.lava.net> arti@address.in.signature (Art Isbell) wrote: > > > > So you just have to edit > > /NextDeveloper/Apps/InterfaceBuilder.app/Resources/images.table and add > > NSReturnSign. Restart IB and everything should be fine. > > No, no, no!!! Don't recommend this sort of hack to new > developers. The > NSReturnSign is no longer used to signify the default button - > the one that will be pressed when the RETURN/ENTER key is pressed. > Forget about NSReturnSign. Since I don't care too much about the windows-like new style I will continue to use good old NSReturnSign. The last time we had this question here it was quite hard to explain that everything will work just fine without the NSReturnSign. So give'em the answer and everything is fine. > Instead, enter "\r" as the button's or buttonCell's key in > its Attributes > Inspector. At runtime, this will result in a dark border around > this button. Yes, but this isn't by far as intuitive as the NSReturnSign. BTW: What if you enter '\r' in more than one button's key attribute? -dirk --- ______________________________________________________________________ Dirk Olmes OBJECT FACTORY Gesellschaft fuer Informatik und Datenverarbeitung mbH Otto-Hahn Str. 18, 44227 Dortmund, Germany Telephon +49 (0) 231 975 137 0 Telefax +49 (0) 231 975 137 99 dirk@object-factory.com http://www.object-factory.com/ Hiroshima 45, Tschernobyl 86, Windows 95
From: csaldanh@mae.carleton.ca (Chris Saldanha) Newsgroups: comp.sys.next.programmer Subject: Sending "terminate" to an app Date: 16 Jun 1997 22:14:36 GMT Organization: Carleton University, Ottawa, Canada Message-ID: <5o4dsc$oua@bertrand.ccs.carleton.ca> I have a system running 3.3 but we've replaced the workspace with another restricted shell (essentially a button bar which launches only a select set of apps). We run it by changing the "loginwindow Workspace" default. I'm trying to figure out how to send a terminate: message to the various running apps in the way that the workspace does upon logout, so that I can cleanly shutdown all the apps, even though I don't have source for the third-party ones. Any ideas? I couldn't find any documentation on this one... --Chris Chris Saldanha | DAR HARM FISH EM LIVE ROSY Carleton University (Comp. Sci) | csaldanh@mae.carleton.ca (NeXT/MIME) | -Another knee-slapper http://www.mae.carleton.ca/~csaldanh | from S/Key
From: d89cb@efd.lth.se (Christian Brunschen) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 17 Jun 1997 08:54:08 GMT Organization: Lund Institute of Technology, Sweden Message-ID: <5o5jbg$gra$1@news.lth.se> References: <5o3eou$rkt$1@ironhorse.plsys.co.uk> <AFCAF619-128736@206.165.44.75> NNTP-Posting-User: d89cb In article <AFCAF619-128736@206.165.44.75>, Lawson English <english@primenet.com> wrote: >mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> said: > >> We have discussed ad nauseum why it makes sense for Rhapsody to use >> DPS; >> you've conceded yourself (unless I misunderstood) that it would not be in > >> AppLE's interests to delay Rhapsody's release until GX is debugged and >> shoehorned in. >> >> I look forward to seeing enhanced GX-like/derived capabilities in >Rhapsody >> *in the fullness of time*. > > >But when? And in the meantime, GX is being crippled on MacOS as of MacOS 8. > >Here's a [partial] list of what you will NOT see with Rhapsody graphics in >the near future, as far as I can tell: > >3D perspective for shapes, collections of shapes and views. > >3D perspective for individual glyphs in a line of text. Neither of which GX supplies either. GX, while using a full 3x3 transform matrix, which allows a number of funny transforms other than the affine transforms available with the 3x2 matrix which PostScript uses, still only does 2D. For 3D, you need at least a 4x3 matrix (or 4x4 -- like the difference between the 2D 3x2 and 3x3 cases). GX does _2D_. Some of the trapezoid transforms that a 3x2 matrix allows may look _vaguely_ like 3D, but they're still _2D_. > >Transparency modes using Porter-Duff compositing on a PER COLOR CHANNEL >basis. I would characterize this as marginally useful, since it > >Retained mode interface for both text and graphics without using a kludgey, >disorganized, pseudo-object-oriented The AppKit is not object-oriented ? Wow. All those classes I took on Object-oriented programming must have fooled me completely. Also, the NSBezier class, which has been announced, looks suspiciously similar to what you claim Rhapsody won't have .... Even further, a retained-mode interface gets really useless when you are displaying over a network rather than just blitting to a locally-connected screen. Where shall we put the database ? In the application ? Then the graphics subsystem has to talk to the display somehow -- either by blitting large images (which will suck up bandwidth, be highly sensitive to latency, and generally suck), or by sending some sort of commands to a display server -- you know, like X or DPS already do it .. ie, no gain over an immediate-mode interface, but rather a loss, since retained-mode has to be emulated over immediate-mode. Or should we put the database in the display ? That way, whenever we want to change something in the database, we have to transmit whatever changes we make to the database, across the network, making all database changes susceptible to network bandwidth limitations, and latency. Or perhaps we should have a copy of the database in the application, and one in the display .. ? Thus cutting the apllication some away from the latency and bandwidth limits in the previous case, but wasting memory for two identical databases ... and we still have to keep them synchronized, resulting in lots of fun network traffic. >class design that doesn't acknowledge >that a bitmap is a first class graphical shape, but instead implements it >in an entirely different part of the framework. Since when can NSImage not be trated like any other graphical object in OpenStep ? >Ditto with text. > >No text-oriented hit-testing that makes use of a shape's transform matrix >automatically. No auto-hit-testing of glyphs during font rendering to make >sure that Hoefler Italic's more complicated glyphs aren't overlapping with >the rest of a text-string and allowing for simplified font-substitution >when such a case occurs. Um, in case you hadn't noticed, Apple has said that GX' typography features will be added to DPS. > >Etc. > >Etc. > >Etc. > >Now, on the MacOS side, MacOS 8 loses ALL of GX Printing. That means that >an application can no longer grab the print job before it is spooled to >disk and make further modifications. That means that GX Printing extensions >can't grab the job AFTER it is spooled to disk and make automatic >modifications to the display-list-based GX Print job. There are over 100 >messages defined that applications and extensions can override to modify >the print stream. Since there is a well-defined, *SIMPLE TO EDIT*, >universally available display-list for GX printing used in all GX Printing, >one can do just about anything that one desires to that print stream. As >long as it can be automated using GX graphics/typography calls, it can be >accomplished. > >If GX Extensions can send/receive AppleSCript messages, you can even alter >the contents of a print job, IN THE PRINT STREAM, from within my HyperCard >stack (not sure if this last is possible -depends on the AE-ability of GX >printing extensions -HyperCard can certainly handle it). well, since GX was never adopted by the grat majority of Mac users / developers, apparently those features were not in such high demand .. ? Furthermore, in publishing, 'WYSIWYG' is usually a Good Thing(TM). Having the print engine go around rearranging things for you behind your back is rather unwanted. Also, if you want certain things to be done to your printjob, there is a more general way of doing it than inserting 'filters' in the print queue: namely, performing whatever modifications you want, on the original document. Because then you can either view it on-screen, print it, email it .... whatever. Your example of translating Japanese text to some sort of English in the print queue is cute, even useful ... but it might just as easily be done _in the application_, before I send the job to the printer at all. In fact, that is where it _should_ be done, because then I don't have to _print_ the document to get the pseudo-translation. So, putting lots of fun filters in the printing pipeline is arguably _wrong_, conceptually. > >The irony of all this is that OpenStep's dynamic messaging would make all >of this MUCH easier than the kludgey System 7.x way of doing things. You >could take GX drivers and make the info available to Rhapsody graphics to >allow for a much more flexable printing system that what is currently >possible using GX. > >But you need a well-defined display list that stays constant throughout the >entire print process. You can't do GX-like things automatically using EPS >and Display PostScript. Presumably they are aware of this and are working >on it, but given statements from Apple management like "Java does what >OpenDoc does" and so on, I'm not convinced that they have a clue. Like as >not, they'll use PDF and call the resultant subset of GX Printing >capabilities "a miracle of Rhapsody engineering." From your persistent misinformation, misunderstandings, and outright _lies_, one is left to wonder if _you_ have a clue. > >And regardless of what Rhapsody can do, that doesn't help ANY current MacOS >user for the next 12-18 months. GX, until MacOS 8, can do things like print >to etching machines. Print transparency to pen plotters. Redirect print >output from one page type to another as needed. Embedd multiple layers of >transparent water marks in different layers of the graphics/text on the >page. Apply 3D perspective to designated elements of the page. Print to >level 1 PostScript printers using Level 2-type features. Apply multiple >printing extensions in a user-sequenced cascade to a single print job. >Print to any device resolution up to 20" on -a-side with 1 million DPI. > >Etc. > >As I said: > >>A combination of money and politics [killing GX Printing]. GX Printing is >far, >>FAR more useful >>than what Rhapsody will provide for the forseeable future. It isn't good >>for Gil Amelio's career if non-RHapsody printing offers more power than >>Rhapsody printing. > >And so they cripple GX Printing to make Rhapsody look better. Um, no. They _get rid of_ GX, and use those things in it that are _useful_ to make Rhapsody indeed _better_. > >Sorry Gil, but it ain't going to work. ALso, there is this: PostScript is _the_ graphics engine in the DTP world. GX is not, was not, going to change that (as could be seen from the number of users using, and developers developing for, GX). Now, instead of trying to 'topple' DPS, Apple and Adobe are going to 'merge' the good parts of GX and PostScript, giving us all something that is _better_ than either was before. Except, of course, for you -- since onle 'pure' GX is good enough for you. (hideously long signature deleted) // Christian Brunschen
From: Konstantin Wiesel <kwiesel@mailhost.jura.uni-bonn.de> Newsgroups: comp.sys.next.programmer Subject: gcc 2.7.2.1 for OS 4.1? Date: Tue, 17 Jun 1997 12:36:19 +0000 Organization: RHRZ - University of Bonn (Germany) Message-ID: <Pine.NXT.3.95.970617123325.4331F-100000@pollux.jura.uni-bonn.de> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I just tried to compile the gcc source that comes with os4.2 on OS 4.1 but without success. Configure detects a Ns 3 machine and it cant compile the first stage compiler. Did anybody have more luck? Regards Konstantin Wiesel Email:kwiesel@mailhost.jura.uni-bonn.de
Newsgroups: comp.sys.next.programmer From: gabriel@trigger.ali.bc.ca (Gabriel Musatescu) Subject: ftp source code Message-ID: <EBxLI0.MtB@gateway.ali.bc.ca> Sender: nobody@gateway.ali.bc.ca Organization: A.L.I. Technologies, Inc. Date: Tue, 17 Jun 1997 17:49:11 GMT Hi Does anyone know where I could find ftp source code on the Internet? Ideally if it was on OPENSTEP but anything would do. thanks. --gabriel
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: GX OOP class proposal & contest (was Re: GX game graphics Date: 17 Jun 1997 15:40:02 -0700 Organization: Primenet Services for the Internet Message-ID: <AFCC64EB-738DD@206.165.44.18> References: <5o5jbg$gra$1@news.lth.se> 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 Christian Brunschen <d89cb@efd.lth.se> said: > > Um, in case you hadn't noticed, Apple has said that GX' typography > features will be added to DPS. Some of those typographical features use the 3x3 matrix transform. Others require hit-testing to be done for every glyph drawn on the screen, causing a round-trip from the AppKit to DPS and back with every glyph drawn unless you implement hit-testing of glyphs on the AppKit side (or eliminate this from the port of GX typography). And if NSImage is a first class graphical object, howcome there's no display list mechanism ala a GX picture shape? Or is there, and does it include bitmap images? BTW, having 3D perspective available for vector shapes and text is considered a big plus for high-end DTP apps, even though it isn't "real" 3D, or so the reviewers of the latest Adobe apps have indicated. ++++++++++++++++++++++++++++++ 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: robertg@trg.saic.com (Robert Gottlieb) Newsgroups: comp.sys.next.programmer Subject: PB.project file format MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Message-ID: <12342024@NEWS.SAIC.COM> Date: Wed, 18 Jun 1997 01:53:37 GMT Hi all, I searched the archives to see if this has been asked before and as far as I can tell it hasn't: I'm trying to parse a PB.project file (NS 3.3) and I want to put the results in a dictionary. Ultimately I just need the PROJECTTYPE and PROJECTNAME attributes. I know I could get both of those via grep or some other shell utility, but I really want to try to make this work in ObjC. I've also heard that NS 4.? changes it to be a true property list, but I'm not upgrading just yet. Anyway, if anyone knows of a way of reading the PB.project file into an NSDictionary or some similiar structure, I'd love to hear your ideas. Thanks, Robert -- robertg@trg.saic.com NeXTMail preferred, others accepted "I believe that what I'm feeling Changes how the world appears" - Neil Peart (Rush) "The above comments/opinions are not that of SAIC"
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 18 Jun 1997 06:45:27 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5o8068$bbt$1@brachio.zrz.TU-Berlin.DE> References: <AFCC64EB-738DD@206.165.44.18> In article <AFCC64EB-738DD@206.165.44.18> "Lawson English" <english@primenet.com> writes: > And if NSImage is a first class graphical object, howcome there's no > display list mechanism ala a GX picture shape? Because the two have nothing to do with each other? Because you don't need to have a display list mechanism to have 1st class graphical objects? @implementation NSArray(DisplayList) -(void)draw { [self contentsPerformSelector:@selector(draw) withObject:nil]; } @end Now arrays can draw themselves. Whew, that was hard! And don't bother pointing out that this is not a GX picture shape. It doesn't need to be, and isn't trying either. > Or is there, and does it include bitmap images? You obviously don't understand dynamic OO if you even ask that question. Marcel
From: "Lee Byeong-ho" <bhlee@cnt.co.kr> Newsgroups: comp.sys.next.programmer Subject: Q: Creating a Compiled Application in WO Date: Wed, 18 Jun 1997 20:24:37 -0000 Organization: Hansol Telecom Message-ID: <5o9fot$345$1@news.hansol.net> Hi all my friends, I searched the WO Enterprise samples created with Objective-C (or C, C++, Perl) compiled code. I saw 'Register Now' samples. But it didn't help me not so much ,so If you have any concerned samples or code, Please send me some tips, I'd love to hear your ideas. Thanks. ---------------------- Lee, Byeong-ho. Yuhan C&T Korea, R&D mail: bhlee@cnt.co.kr ----------------------
From: droege@informatik.uni-koblenz.de (Detlev Droege) Newsgroups: comp.sys.next.programmer Subject: Re: ftp source code Date: 18 Jun 1997 11:32:55 GMT Organization: University Koblenz / Germany Distribution: world Message-ID: <5o8h17$df$1@newshost.uni-koblenz.de> References: <EBxLI0.MtB@gateway.ali.bc.ca> In article <EBxLI0.MtB@gateway.ali.bc.ca> gabriel@trigger.ali.bc.ca (Gabriel Musatescu) writes: > Hi > Does anyone know where I could find ftp source code on the Internet? > Ideally if it was on OPENSTEP but anything would do. > thanks. Binaries come with every NS/OS system: /usr/ucb/ftp If you really need some source, try e.g. the GNU Internet Utilities inetutils-1.3a.tar.gz from your nearest GNU archive Detlev -- Detlev Droege, Uni Koblenz, FB Informatik, Rheinau 1, D-56075 Koblenz, Germany Tel:+49 261 9119-421,Fax:-497,NeXT/MIME/Emil: droege@informatik.uni-koblenz.de C++ is the only current language making COBOL look good. --Bertrand Meyer
From: Timothy J Luoma <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: Re: ftp source code Date: Wed, 18 Jun 1997 09:13:51 -0400 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.NXT.3.96.970618091037.10489C-100000@cc344191-a.ewndsr1.nj.home.com> References: <EBxLI0.MtB@gateway.ali.bc.ca> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII To: Gabriel Musatescu <gabriel@trigger.ali.bc.ca> In-Reply-To: <EBxLI0.MtB@gateway.ali.bc.ca> On Tue, 17 Jun 1997, Gabriel Musatescu wrote: > Does anyone know where I could find ftp source code on the Internet? > Ideally if it was on OPENSTEP but anything would do. > thanks. I just uploaded the newest beta release (with most up to date security patches) of wu-ftp to ftp://ftp.next.peak.org/pub/next/apps/internet/ftp/wu-ftpd.2.4.2-beta-13.NIHS.bs .tar.gz ftp://ftp.next.peak.org/pub/next/apps/internet/ftp/wu-ftpd.2.4.2-beta-13.README That is source code and compiled (under NS 3.3) binaries. This is probably the most up to date ftp stuff you are going to find. Here's a new wu-ftp link I just found also: http://www.landfield.com/wu-ftpd/ That has the wu-ftp FAQ etc..... TjL ps -- I've made cursory efforts of actually getting this working (ie config files, etc) but have not yet succeeded. If you do, please drop me a note ;-) -- TjL <luomat@peak.org> / http://www.peak.org/~luomat/next/ NeXT bookmarks: http://www.peak.org/~luomat/next/bookmarks.html "Everything is easy when you know what you are doing." - Dr Robert Cupper, Department of CS, Allegheny College
From: d89cb@efd.lth.se (Christian Brunschen) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 18 Jun 1997 13:41:30 GMT Organization: Lund Institute of Technology, Sweden Message-ID: <5o8oia$c4m$1@news.lth.se> References: <5o5jbg$gra$1@news.lth.se> <AFCC64EB-738DD@206.165.44.18> NNTP-Posting-User: d89cb In article <AFCC64EB-738DD@206.165.44.18>, Lawson English <english@primenet.com> wrote: >Christian Brunschen <d89cb@efd.lth.se> said: > >> >> Um, in case you hadn't noticed, Apple has said that GX' typography >> features will be added to DPS. > > > >Some of those typographical features use the 3x3 matrix transform. Which _typography_ features require non-affine tranforms ? I am not talking about the ability to apply arbitrary 3x3 transform matrices to everythin, including text -- that has nothing to do with _typography_. >Others >require hit-testing to be done for every glyph drawn on the screen, causing >a round-trip from the AppKit to DPS and back with every glyph drawn unless >you implement hit-testing of glyphs on the AppKit side (or eliminate this >from the port of GX typography). Firstly, since the AppKit is the OO API that shields programmers from bare-bones PostScript programming, I would suggest that this is indeed a rather good place to put the typography functionality. Secondly, even if it is implemented on the server side, hit-testing would not have to include a round-trip for every glyph drawn, since hit-tests for multiple shapes could be easily coalesced, thus generating a round-trip for every hit-testing done, not one for every glyph. Thirdly, you seem to be determined to think that, since Apple decided not to go with Pure GX, they must have lost their brains completely. I am quite confident that > >And if NSImage is a first class graphical object, howcome there's no >display list mechanism ala a GX picture shape? Um, because DPS isn't a retained-mode graphics engine ... ? >Or is there, and does it include bitmap images? Bitmap images can be told to draw themselves to a PostScript view with one (1) Objective-C method call. This will typically be done inside the view's '- drawSelf:' method, which is called when the view needs updating. So far, programmers under NEXTSTEP and OPENSTEP have managed to keep track of whatever objects, or PostScript calls, they need to for this to happen, without the help of a Display List ala GX. Immediate-mode graphics work _very well_, and in a networked context they avoid a _number_ of problems that retained-mode graphics systems have when the machine which composes the image, and the machine that blits the pixels to the screen, are not the same. Giving explicits commands that are executed one-at-a-time is always going to be simpler, and more efficient, to implement _correctly_ than any solution where a database has to be transported, or two databases have to be synchronized, over a network. Besides, if you want to have an explicit display list API, noone is going to stop you from writing a 'DisplayListView' which holds a display list, and in its 'drawSelf' method draws each item in the display list to the screen. You could then make all _your_ views subclasses of DisplayListView, and you'd never have to touch a PostScript operator again. Remeber, Rhapsody is going to include classes for Bezier paths, Graphics contexts, etc. No need for you to touch those disgusting PostScript operators. > >BTW, having 3D perspective available for vector shapes and text 2D is 2D, 3D is 3D. GX offers 2D. Saying '3D perspective' is spreading a _falsehood_. >is >considered a big plus for high-end DTP apps, even though it isn't "real" >3D, or so the reviewers of the latest Adobe apps have indicated. Perhaps you would care to include some pointers to these reviews so that we might read them for ourselves, rather than hearing your interpretation ? (same hideously long advertising .sig snipped) Best regards // Christian Brunschen /* Disclaimer: Everything I write is my own opinion, which is based on my interpretation of those facts to which I have access. Errors may obviously creep in, but I do try to avoid them; however, feel free to correct me. */
From: Timothy J Luoma <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: Re: ftp source code Date: Wed, 18 Jun 1997 09:43:29 -0400 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.NXT.3.96.970618094118.11021C-100000@cc344191-a.ewndsr1.nj.home.com> References: <EBxLI0.MtB@gateway.ali.bc.ca> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII To: Gabriel Musatescu <gabriel@trigger.ali.bc.ca> In-Reply-To: <EBxLI0.MtB@gateway.ali.bc.ca> On Tue, 17 Jun 1997, Gabriel Musatescu wrote: > Does anyone know where I could find ftp source code on the Internet? Oh! Did you mean 'ftp' or 'ftpd' ? If you meant source code for ftp-ing OUT from your machine, you want ftp. If you meant for ftp-ing INTO your machine, then you want wu-ftp. Source code could be found for ftp in the GNU stuff as someone else already suggested, or you could try ncftp at ftp.probe.net which is a much nicer commandline ftp program... TjL -- TjL <luomat@peak.org> / http://www.peak.org/~luomat/next/ NeXT bookmarks: http://www.peak.org/~luomat/next/bookmarks.html "Everything is easy when you know what you are doing." - Dr Robert Cupper, Department of CS, Allegheny College
Newsgroups: comp.sys.next.programmer From: kevinc@netcom.com (kevinc) Subject: GNU binutils-2.8.1 Message-ID: <kevincEBz5Mq.Hn1@netcom.com> Organization: Netcom On-Line Services Date: Wed, 18 Jun 1997 14:01:38 GMT Sender: kevinc@netcom2.netcom.com Hi guys, Has anyone compiled 2.8.1 for NextStep 3.3? configure complains thjat Intel 3.3 is not supported by BFD. Any assistance will be most appreciated. Rgds, Kevin
From: shess@one.net (Scott Hess) Newsgroups: comp.sys.next.programmer Subject: Re: Strangeness... Date: 18 Jun 1997 14:13:51 GMT Organization: Is a sign of weakness Message-ID: <SHESS.97Jun4115843@slave.one.net> References: <beauvois-0306970231110001@amour.la.utk.edu> <EB7KLq.KJL@gateway.ali.bc.ca> <5n2r2b$l6@saturn.genoa.com> In-reply-to: Alex Blakemore's message of 4 Jun 1997 04:31:07 GMT In article <5n2r2b$l6@saturn.genoa.com> Alex Blakemore <alex@genoa.com> writes: In <EB7KLq.KJL@gateway.ali.bc.ca> Gabriel Musatescu wrote: > Don't expect an unallocated object to be nil. true enough when speaking of _automatic_ variables, (those defined local to a method or block) but just to be pedantic, it is safe to assume that instance variables are initialized to nil - that is defined by the language. so code in the "init" method that sets ivars to zero or nil is redundant, unless of course you expect init to be sent more than once to the same object. On the other hand, explicitely setting instance variables to nil is a form of implicit documentation, -- scott hess <scott@doubleu.com> (606) 578-0412 http://www.doubleu.com/ <Favorite unused computer book title: The Idiots Guide to the Zen of Dummies in a Nutshell in Seven Days, Unleashed>
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: GX remote drawing (Was Re: GX OOP class Date: 18 Jun 1997 07:26:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFCD4284-ACA1@206.165.44.53> References: <5o5jbg$gra$1@news.lth.se> 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 Christian Brunschen <d89cb@efd.lth.se> said: > Also, the NSBezier class, which has been announced, looks suspiciously > similar to what you claim Rhapsody won't have .... > Are you so familiar with GX's capabilities, as described in about 1500 pages of manuals that you can make that claim? With GX, I can do a "GXSImplifyShape(myBitmapShape)", and if myBitmapShape is only one color, it will turn it into a GXRectangleShape. Will NSImage and NSBezier support such a thing? With GX, I can turn off dithering/halftoning on a per-shape basis. How's NSImage and NSBezier do for that level of coordination? There is a LOT of stuff you can do with GX. A NeXT developer told me that GX is "over-engineered" but a NeXTStep hobbyist programmer who was hoping to use GX in Rhapsody snorted at that and commented that it was "powerful, not 'over-engineered.'" > Even further, a retained-mode interface gets really useless when you > are displaying over a network rather than just blitting to a > locally-connected screen. Who says? Where shall we put the database ? Where it has always been: in the GX Graphics Heap. In the > application ? No. Where it has always been: in the GX Graphics Heap. Then the graphics subsystem has to talk to the display > somehow -- either by blitting large images (which will suck up > bandwidth, be highly sensitive to latency, and generally suck), or by > sending some sort of commands to a display server -- you know, like X > or DPS already do it .. ie, no gain over an immediate-mode interface, > but rather a loss, since retained-mode has to be emulated over > immediate-mode. Obviously, you don't know squat about GX. > Or should we put the database in the display ? Obviously, you don't know squat about GX. That way, whenever we > want to change something in the database, we have to transmit > whatever changes we make to the database, across the network, making > all database changes susceptible to network bandwidth limitations, and > latency. Obviously, you don't know squat about GX. > Or perhaps we should have a copy of the database in the application, > and one in the display .. ? Thus cutting the apllication some away > from the latency and bandwidth limits in the previous case, but > wasting memory for two identical databases ... and we still have to > keep them synchronized, resulting in lots of fun network traffic. > Obviously, you don't know squat about GX. If you want to draw into an application's window over a network by using GX, you need to do a few things: 1) Establish some kind of communications channel over the network to send function calls to the copy of GX that exists on that MacOS machine. 2) procure the appropriate GX Graphics Heap reference from an existing application. 3) procure the appropriate GX viewport 4) set the remote GX heap to the appropriate heap 5) make the calls via the communications channel to the copy of the GX library on the other machine, with reference to the window's viewport object, when needed (which might be only once -to set the default viewport, if it isn't already set). The communications channel would be AppleSCript for convenience' sake. Any app that wanted to allow remote drawing would simply respond to the query for the appropriate graphics heap, windows viewport, etc. The reason why I say that you don't know squat about GX is that GX *ALREADY* maintains the database in its own private segment of memory that no application has access to (leaving aside MacOS's non-protected memory space, of course). Any modifications made to the retained-mode shapes would be made in the private memory space created by GX on the remote machine. In fact, it is funny that you should mention this remote drawing issue. ONce I get the GXFCN debugged and shipping, my NeXT project is to create a GX OSA that uses FaceSpan for the windows. Developers can actually create faceless apps that draw via FaceSpan, or they can create quick and dirty, drawing scripts, or pictures, or whatevers from within any app that can attach an AppleScript and print them out using GX Printing (or whatever kludge Apple comes up with). Since AppleScript is already networkable, this would make my OSA a network-server for GX. [thanks for the idea, Eric King!] Of course, since AppleScript is slow, this solution would be slow, but obviously the same principle could be used for a speedier communications channel... ... like the one used in Rhapsody, maybe? And of course, one could always replace DPS's forth-like language with Java and use Java to call GX or Taligent graphics... [thanks to my brother for this idea!!!] Hey! What an idea: make Rhapsody the premier internet graphics system by making its main graphics engine the same 2D graphics engine that will be used in Java applets starting in Java 2.0... Call it Apple Internet Graphics, or some such... BTW, I don't know how Taligent handles allowing other graphics systems to draw directly to the screen, but GX allows one to specify a clip shape for a view device, which means that an app that wanted to draw directly to the screen could reserve a section of the screen for its own use by supplying a clip-shape, and not even window-frames drawn using GX would obscure it. Huh! Sounds like Interceptor, but better since it would never interfere with anything GX does and could still track the cursor and so on... But you can do all of this using Rhapsody graphics based on DPS, despite the kludgeyness of it all, so therefore DPS + AppKit is "just as good" as GX... Yeparoonies, them folks at Apple shore know what they're doing with this GX vs DPS vs Taligent vs whatever issue... ++++++++++++++++++++++++++++++ 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.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 18 Jun 1997 08:10:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFCD4CFC-3223A@206.165.44.53> References: <5o8oia$c4m$1@news.lth.se> 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 > In article <AFCC64EB-738DD@206.165.44.18>, > Lawson English <english@primenet.com> wrote: > >Christian Brunschen <d89cb@efd.lth.se> said: > > > >> > >> Um, in case you hadn't noticed, Apple has said that GX' typography > >> features will be added to DPS. > > > > > > > >Some of those typographical features use the 3x3 matrix transform. > > Which _typography_ features require non-affine tranforms ? I am not > talking about the ability to apply arbitrary 3x3 transform matrices to > everythin, including text -- that has nothing to do with > _typography_. > The ability to apply a 3x3 transform matrix to each glyph in a text shape in layers called a "textface." > >Others > >require hit-testing to be done for every glyph drawn on the screen, > causing > >a round-trip from the AppKit to DPS and back with every glyph drawn > unless > >you implement hit-testing of glyphs on the AppKit side (or eliminate this > >from the port of GX typography). > > Firstly, since the AppKit is the OO API that shields programmers from > bare-bones PostScript programming, I would suggest that this is indeed > a rather good place to put the typography functionality. > > Secondly, even if it is implemented on the server side, hit-testing > would not have to include a round-trip for every glyph drawn, since > hit-tests for multiple shapes could be easily coalesced, thus > generating a round-trip for every hit-testing done, not one for every > glyph. > OK, so how do you do this with text? Every adjacent character might be overlapping, so how do you manage to reduce the number of round trips? > Thirdly, you seem to be determined to think that, since Apple decided > not to go with Pure GX, they must have lost their brains completely. I > am quite confident that > You keep on trailing off... And yes, given remarks that Hancock and AMelio have made (e.g. Java can replace OpenDoc's functionality, we don't need no stinkin' TV ads -they cost too much, etc), I generally agree with your characterization about Apple management having lost their brains. Or do you think that their performance in overseeing the fall of Apple has been an example of *competence*? > > > >And if NSImage is a first class graphical object, howcome there's no > >display list mechanism ala a GX picture shape? > > Um, because DPS isn't a retained-mode graphics engine ... ? > > >Or is there, and does it include bitmap images? > > Bitmap images can be told to draw themselves to a PostScript view with > one (1) Objective-C method call. This will typically be done inside > the view's '- drawSelf:' method, which is called when the view needs > updating. So far, programmers under NEXTSTEP and OPENSTEP have managed > to keep track of whatever objects, or PostScript calls, they need to > for this to happen, without the help of a Display List ala GX. > > Immediate-mode graphics work _very well_, and in a networked context > they avoid a _number_ of problems that retained-mode graphics systems > have when the machine which composes the image, and the machine that > blits the pixels to the screen, are not the same. > Who said that this was necessary? > Giving explicits commands that are executed one-at-a-time is always > going to be simpler, and more efficient, to implement _correctly_ than > any solution where a database has to be transported, or two databases > have to be synchronized, over a network. > Who said that this was necessary? > > Besides, if you want to have an explicit display list API, noone is > going to stop you from writing a 'DisplayListView' which holds a > display list, and in its 'drawSelf' method draws each item in the > display list to the screen. You could then make all _your_ views > subclasses of DisplayListView, and you'd never have to touch a > PostScript operator again. Remeber, Rhapsody is going to include > classes for Bezier paths, Graphics contexts, etc. No need for you to > touch those disgusting PostScript operators. > But for some of the more interesting features of GX Printing to be workable, you need to have some kind of display-list mechanism that is universally defined. > > > >BTW, having 3D perspective available for vector shapes and text > > 2D is 2D, 3D is 3D. GX offers 2D. Saying '3D perspective' is spreading > a _falsehood_. > Calling me a liar? Please explain why Loren Petrich's analysis of GX's 3D perspective mapping is invalid. GX implements a 3x3 matrix that allows one to apply a 3D perspective to 2D textures such as vector graphics, bitmaps and text. > >is > >considered a big plus for high-end DTP apps, even though it isn't "real" > >3D, or so the reviewers of the latest Adobe apps have indicated. > > Perhaps you would care to include some pointers to these reviews so > that we might read them for ourselves, rather than hearing your > interpretation ? Every high-end DTP person that *I* know wants 3D perspective capabilties. In fact, Level 3 PostScript supports "3D effects," so obviously Adobe felt that this was important to support. BTW, is it a _falsehood_ for Adobe to call 3D perspective "3D effects?" > > (same hideously long advertising .sig snipped) > ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: Tim Pugh <tpugh@nospams.OCE.ORST.EDU> Newsgroups: comp.sys.next.programmer Subject: Re: Does OS4.2 finally fix the printf %g bug? Date: 18 Jun 1997 17:12:25 GMT Organization: Oregon State University Message-ID: <5o94tp$atk@news.orst.edu> References: <5npof5$jlf@tribune.usask.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: eric@skatter.USask.Ca In <5npof5$jlf@tribune.usask.ca> eric@skatter.USask.Ca wrote: > Would someone out there running OPENSTEP 4.2 try my famous printf test program? > > For those of you coming in late, this program tests for a bug that has been > present since NeXTstep version 1.0 and which I first reported in December, 1989. > > I even have the NeXT bug tracking reference numbers [60697, 98369] and an e-mail > message from NeXT (September 1994) indicating that, ``it looks like it will > finally be fixed in NEXTSTEP release 4.0.'' > > ================================================================ > #include <stdio.h> > > int main (int argc, char **argv) > { > printf ("Expect 0.00123: %.3g\n", 0.001234567); > printf ("Expect 123: %.3g\n", 123.4567); > printf ("Expect 123.5: %.4g\n", 123.4567); > printf ("Expect 1e+03: %.3g\n", 999.6); > return 0; > } > ================================================================= > rarebear> a.out Expect 0.00123: 0.001 Expect 123: 123.457 Expect 123.5: 123.4567 Expect 1e+03: 999.6 rarebear> hostinfo Mach kernel version: NeXT Mach 4.2: Wed Apr 16 13:44:57 PDT 1997; root(rcbuilder):Objects/mk-183.34.obj~2/RELEASE_I386 Kernel configured for a single processor only. 1 processor is physically available. Processor type: I386 (Intel 486) Processor active: 0 Primary memory available: 64.00 megabytes. Default processor set: 64 tasks, 150 threads, 1 processors Load average: 0.34, Mach factor: 0.76 The answer is NO. I guess we'll have to wait until Rhapsody and BSD4.4 . - Tim -
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 18 Jun 1997 16:24:26 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5o923q$ceg$1@brachio.zrz.TU-Berlin.DE> References: <AFCD4CFC-3223A@206.165.44.53> In article <AFCD4CFC-3223A@206.165.44.53> "Lawson English" <english@primenet.com> writes: > > Which _typography_ features require non-affine tranforms ? I am not > > talking about the ability to apply arbitrary 3x3 transform matrices to > > everythin, including text -- that has nothing to do with > > _typography_. > > The ability to apply a 3x3 transform matrix to each glyph in a text shape > in layers called a "textface." The question was not: what can GX do to typgraphy, but which typgraphic features require 3x3 transforms? Typography != GX feature-list. > OK, so how do you do this with text? Every adjacent character might be > overlapping, so how do you manage to reduce the number of round trips? How many characters can actually surround one point on the screen? > Or do you think that their performance in overseeing the fall of Apple has > been an example of *competence*? They are cleaning up the mess that was created before they were on board. (Lawson probably also shoots messengers for bad news...) > > >BTW, having 3D perspective available for vector shapes and text > > > > 2D is 2D, 3D is 3D. GX offers 2D. Saying '3D perspective' is spreading > > a _falsehood_. > > > > Calling me a liar? Please explain why Loren Petrich's analysis of GX's 3D > perspective mapping is invalid. Because 2D effects that look a little 3D-ish are not 3D. And if we're going to argue from authority, instead of simple logic, I'd rather put my lot in with what every computer-graphics textbook says on the subject, instead of with what Loren Petrich says. Marcel
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: GX remote drawing (Was Re: GX OOP class Date: 18 Jun 1997 17:05:34 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5o94gu$f3b$1@brachio.zrz.TU-Berlin.DE> References: <AFCD4284-ACA1@206.165.44.53> In article <AFCD4284-ACA1@206.165.44.53> "Lawson English" <english@primenet.com> writes: > > Even further, a retained-mode interface gets really useless when you > > are displaying over a network rather than just blitting to a > > locally-connected screen. > > Who says? Foley van Dam, Computer Graphics, 2nd Edition. > Where shall we put the database ? > > Where it has always been: in the GX Graphics Heap. That isn't answering the question. Where is the GX graphics heap? > The reason why I say that you don't know squat about GX is that GX > *ALREADY* maintains the database in its own private segment of memory that > no application has access to (leaving aside MacOS's non-protected memory > space, of course). So they chose one of the options Christian described, and thus have the problems that Christian described (inefficient access to shape-DB, which is inacceptable for most programs, or duplication, which is also problematic). So much ignorance in one package... Marcel to do with anything? Adobe is a software vendor. They do not determine what features are or are not in the core operating system and frameworks. Features such as font support and font architectures are determined by Apple, to meed the needs of the developer and end user community. NeXT had a long relationship with Adobe. At no time did Adobe try to block the use or development of other technologies, or prevent NeXT from integrating these technologies into their system. I would expect this to continue with Apple. > Since Japanese fonts can take up to 6 MB, whereas FOntWork's custom > Japanese fonts only take up 1.6 MB, this means that Rhapsody will require > nearly 4x the space for Japanese fonts that GX would since FontWorks has a > custom GX rendering engine for displaying these fonts. I sure hope they did this as an OFA plug-in... > When you add to the equation the fact that GX allows one to parse the > actual characters used in a print job and download to a printer ONLY those > characters that are actually used on a page whereas Rhapsody apparently > won't support GX-like functionality in this area [munch] You don't know what you are talking about. Really. [munch] > Face it, Joe, as Tom Bayly said on Carpe.Diem: there's no doubt which is a > better graphics system. > > And it is NOT Display PostScript. Duh. Display PostScript is just one small part (the 2D vector graphics bit) of the Rhapsody Graphics Architecture. There's a heck of a lot more there than you seem to realize, covering fonts, typography, raster graphics, and imaging, along with a whole slew of new technologies. [munch] > BTW, Joe, GX Typography includes the ability to assign a custom layer to a > type font that may include a 3D perspective. Since Display PostSCript > doesn't support 3D perspectives, how is it possible for RHapsody Graphics > to implement ALL of GX Typography? Alas, what GX supports is just a 3x3 transform matrix, permitting some obscure distortion effects in addition to the usual affine transforms. Even the fluffy marketing papers never referred to this as anything more than "2 1/2 D" graphics. Unfortunately, some people are easily confused by all these threes... True 3D requires a minimum of a 4x3 transform operation, and a means of specifying three coordinate dimensions. [munch] > Blind loyalty is, well, blind... ..and ignorance is strength, eh? -- Mike Paquette (mpaque@wco.com) 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: jrudd@cygnus.com (John Rudd) Newsgroups: comp.sys.next.programmer Subject: Loading Elements into NSPopUpButton Date: 18 Jun 1997 17:44:34 GMT Organization: Cygnus Solutions Message-ID: <5o96q2$ars$1@cronkite.cygnus.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Ok, so I've been a Nextstep User and Admin for close to 6 years now, and I even bought several of the books (I even have a copy of the NeXT Bible, and 2 copies of the Garfinkle Step One book, and the entire black & gold NeXT reference set (and even a few copies of the white & black cover versions of those books) for NS 3.x) to learn how to do NS programming, but I haven't actaully gotten around to it until now. So, the other day I threw together a quick and dirty calculator -- it had 3 text fields and 5 buttons. I used my own "MyCalcObj" (which became MyCalc2Obj in one iteration of the design) to control things -- 2 of the text fields would send operands to it, and then it would send an answer to the third when you pushed one of the buttons (which had integer math operators + - * / % ). That worked fine. I even have it set up so that the answer says "Error" if you try to devide/modulo by zero. I didn't line up the GUI elements too well, or anything -- but this is just my practice at the programming side of things. I decided to redo it using a NSPopUpButton for the operator selection. Everything works EXCEPT the contents of the pop up button are wrong (ie. "Item 1" does add, but there's only 3 items, not 5) I double-clicked on the button in Interface builder, and it reveals all of the button elements, but if I entered "-" for a button title it disappeared -- but when I went to run the finished app, all of those "-" buttons were there, but in a weird order compared to the other buttons. And not all of the correct button titles show up -- just the initial +, several -'s, and one /. The first 5 buttons perform as expected, but it just doesn't look the way its supposed to. So I tried to fix it programatically -- when MyCalc2Obj init's, I tried to have it send messages to its Operation outlet, which is the NSPopUpButton. That didn't do anything. I tried making the NSPopUpButton be a subclass and I modified its init to do those same messages (only to self this time). Still no dice. So then I had the NSPopUpButton target a method in MyCalcObject, and if it doesn't have the right number of elements, it tells the popup button to remove all of its items, and then inserts the correct list of elements. This one lets me pop up the list once (with all of the wrong elements), and then the pop up becomes either a) empty or b) unresponsive -- I can't tell which because when I click on it nothing shows other than the lead item, so I can't tell if it is in fact responding, or just empty. But it still works (ie. when I change the values in the two operands, the answer still comes out right). I assume that the reason the init's didn't work has to do with a) the order in which things are extracted from the .nib file (when MyCalcObject extracts itself, and I try to use init to send messages to Operation, Operation is later overwritten when it gets pulled out of the .nib file), or that init isn't invoked because objects extracted from the .nib file aren't being created, they're being unarchived. But that doesn't explain the strange behavior of the last case. You can look at the entire project at: http://www.cygnus.com/~jrudd/MyCalc3 I haven't done any commenting, but there is very little code, too. If it's hard to read, let me know and I'll pretty it up. I'm using Openstep 4.1/Mach on a Sparcstation 4/110MHz (technically not supported, but the only problem I've encountered so far is sound -- this model doesn't come with a built in sound card). So, what is the right way to specify the fields of the NSPopUpButton? Either staticly through IB or programatically is fine (though the static/IB mechanism is really more appriate to this program). -- John "kzin" Rudd jrudd@cygnus.com http://www.cygnus.com/~jrudd =========Intel: Putting the backward in backward compatible.============ Thought for the day: According to the supreme court, proof of innocence isn't enough to avoid execution if you've exhausted your appeals.
From: mpaque@yummyspam.wco.com (Mike Paquette) Newsgroups: comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 18 Jun 1997 12:12:00 -0700 Organization: Electronics Service Unit No. 16 Sender: mpaque@mpaque Distribution: world Message-ID: <5o9bu0$1pg@mpaque.mpaque> References: <AFCC64EB-738DD@206.165.44.18> In article <AFCC64EB-738DD@206.165.44.18> "Lawson English" <english@primenet.com> writes: > Some of those typographical features use the 3x3 matrix transform. Others > require hit-testing to be done for every glyph drawn on the screen, causing > a round-trip from the AppKit to DPS and back with every glyph drawn No. > And if NSImage is a first class graphical object, howcome there's no > display list mechanism ala a GX picture shape? OPENSTEP doesn't use a display list architecture. This is generally considered a feature. > Or is there, and does it include bitmap images? The OPENSTEP API provides a rich set of classes supporting both vector and raster based images. The NSImage superclass supports a number of built-in image representations, including in OPENSTEP 4.2 support for EPS files, TIFF images (including a variety of TIFF compression formats), and raw bitmaps. Applications ma register additional representations with the system. > BTW, having 3D perspective available for vector shapes and text is > considered a big plus for high-end DTP apps, even though it isn't "real" > 3D, or so the reviewers of the latest Adobe apps have indicated. Yes. In the past, NeXT shipped a full 3D Kit and both Interactive and Photorealistic Renderman from PIXAR. I expect that the excellent 3D technologies developed within Apple will be integrated into Rhapsody by the Unified release, so developers can continue to take advantage of 3D. -- Mike Paquette (mpaque AT wco.com ; Yank that yummyspam 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: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.mac.programmer,comp.sys.next.programmer Subject: Computer Users of the World, Share Your Dreams for Rhapsody Date: 18 Jun 1997 19:45:50 GMT Organization: Digital Fix Development Message-ID: <5o9dte$1lv$1@news.digifix.com> Dear (potential) members of the Rhapsody Community, Hi, my name is Ernie Prabhakar, and on Monday June 9th I will become the Product Marketing Manager for Rhapsody, Apple's NeXT Generation Operating System. I am only filling in for two months while the "real" Product Manager is on sabbatical, after which I intend to resume my normal life of management consultant by day, computer groupie by night. However, before I disappear behind the Silicon Curtain, before I am an Official Apple Employee, I just wanted to share with you my vision of the world I would like to help create. Or rather, a world that I would like you to help create. I am not a professional programmer or computer scientist. I am just a guy who believes computers can be a powerful tool which people can use to change the world. And I want to do what I can to help computers reach their potential, primarily by helping computer -users- reach their potential. Computers are amazing devices in their ability to perform a wide variety of tasks, but all too often this comes at the expense of failing to perform a single task very well - the task that you want it to do! There are many reasons for this mismatch between technology and user needs, but let me focus on one key area where I hope that I - and you - can make a difference. Software development has traditionally been such an expensive, arduous and time consuming process that a) developers had to be experts on technology, not on human problems b) products were designed to satisfy static requirements, not dynamic communities c) the development process largely excluded user involvement Rhapsody, as the only mass-market operating system designed from the ground-up to support object-oriented software development, has the potential to break these compromises and change the rules of software development. If it succeeds, it should provide incredible power and ease of use to end users, developers, and system administrators. I believe Apple has assembled one of the greatest teams of engineers, scientists, and design specialists ever seen in the software industry. I believe they have the tools, support, and vision to pull it off. However, they lack one thing, the one thing that makes the crucial difference between a product that revolutionizes the world and one that merely looks impressive. It is the one thing I most want to give to Apple: You. You, the guy who fights with his mail application, or labors over updating web pages, or dreams of writing a "killer app" that crushes the Big Guys, or merely wants to transform her little corner of the world. I want Rhapsody, along with the tools, programs, and community it engenders, to be the vehicle for making your dreams come true. Not just a dream of how a computer should work, but dreams of community, freedom, learning, love - the stuff people live and die for all over the world. Apple has always been about helping people realize their dreams, and it has made many dreams come true. But in all honesty, Apple was hampered both by its technology, and its culture. Like most great insitutions, it lost sight of its original focus - you, the ultimate consumer - and got caught up in esoteric dreams of greatness. That's changed. Apple has been humbled, and it is reaching out. The acquisition of NeXT, and Apple's commitment to Java and open systems are the signs of a sea change. In a tiny way, bringing in an outsider like myself to work alongside marketing is also a sign of that change. I am no Silicon Valley insider, or hot shot programmer. I'm just like you. I use a computer to balance my checkbook, write love letters, and send email. I've written games for my friends, technical analyses as a student, and mission critical custom apps for a multinational corporation. I've lugged notebooks across the world to share presentations and datafiles with clients, and I've setup PPP and Ethernet networks in my home. And all along, I've often thought: there has to be a better way. If I could, I'd love to sit down with each and every one of you, to hear your dreams for your life and the world, and figure out what small part a computer might play in helping you realize them. But I can't. Even if I had the time, my loyalty is to Apple, to be part of their Voice. I can't be your voice. But, what I can do, what I hope to do, is help you find your voice. To bring people together, where they can fight through the tradeoffs and priorities that are an inevitable part of any democratic community. And then, to be Apple's Eyes and Ears, to see your concerns, hear what you're saying, feel your pain. And then to do whatever I can within Apple, to make sure your voices are heard. To do this, I have asked some friends of mine to setup two mail aliases For developers, devideas@stepwise.com by Don Yacktman, of The Object Foundation For users, userideas@nula.com by Tim Byars of NeXT Users Los Angeles These two have committed to sorting through all the mail that is sent to these lists, and identifying the biggest concerns in those communities. Whether it is a matter of technology, communication, or business. And I have gauranteed them my time and attention to understand those concerns, and my passion and energy to communicate those to the Apple community. I have to be honest with you. Product Managers have no real authority, and as a short-termer and outsider I could very well have less than usual. I am also a realist and a capitalist: Apple has to make money, and soon, and that means shipping a product as quickly as possible, not catering to everyone's hopes and dreams. However, I am enough of an idealist to believe that a company like Apple can only survive if it succeeds in truly understanding its customers. More than almost any other American company, Apple isn't about products, it is about people. It is about changing the world. It did it once with the Apple II, then again with the Macintosh. WIth your help, I believe it may yet have one more revolution in it. I don't know for sure if I can do anything, or even if you can do anything. But I know I have to try. Will you join me? Sincerely, Dr. Ernie Prabhakar Rhapsody Product Manager Designee
From: adrissmn@gopher.science.wayne.edu (Avi Drissman) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] Main executable in application folder Date: 18 Jun 1997 20:35:53 GMT Organization: Wayne State University Message-ID: <5o9gr9$e9s@cwis-20.wayne.edu> References: <866191282.19621@dejanews.com> <EBpyqL.61n@novice.uwaterloo.ca> David Evans (dfevans@bbcr.uwaterloo.ca) wrote: : Yes. If you have an app wrapper called "Blah.app" then "Blah.app/Blah" will : be run. Does that mean that if I rename an app it stops working? (Um, that would be bad.) Avi -- Avi Drissman Preferred address: drissman@acm.org +-------------------------------------------------------------------+ http://www.science.wayne.edu/~adrissmn/ And we'll have fun, fun, fun till my daddy takes the keyboard away...
From: Timothy J Luoma <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: What needs to be FAT to compile FAT? Date: Wed, 18 Jun 1997 16:29:43 -0400 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.NXT.3.96.970618162432.19293B-100000@cc344191-a.ewndsr1.nj.home.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII All the binaries in /usr/ucb are 3-fat on my 4.1 install! Why? Is this an general installation bug or did I select something wrong? I want to be able to compile things MAB, but don't need MAB binaries myself. What can I thin and what can't I? I'm afraid of will-nilly 'lipo'-ing files that will mean I can't compile MAB, but also don't want to have 3-fat around when I don't need it. Please help me reclaim lost diskspace.... Thanks! TjL
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.mac.programmer,comp.sys.next.programmer Subject: cmsg cancel <5o9dte$1lv$1@news.digifix.com> Control: cancel <5o9dte$1lv$1@news.digifix.com> Date: 18 Jun 1997 20:39:51 GMT Organization: Digital Fix Development Message-ID: <cancel.5o9dte$1lv$1@news.digifix.com> Cancelled by comp.sys.next.announce Moderator sanguish@digifix.com
From: sanguish@digifix.com Newsgroups: comp.sys.next.announce,comp.sys.mac.programmer,comp.sys.next.programmer Subject: Computer Users of the World, Share Your Dreams for Rhapsody Date: 18 Jun 1997 20:46:48 GMT Organization: Digital Fix Development Message-ID: <5o9hfo$2u7$1@news.digifix.com> Dear (potential) members of the Rhapsody Community, Hi, my name is Ernie Prabhakar, and on Monday June 9th I will become the Product Marketing Manager for Rhapsody, Apple's NeXT Generation Operating System. I am only filling in for two months while the "real" Product Manager is on sabbatical, after which I intend to resume my normal life of management consultant by day, computer groupie by night. However, before I disappear behind the Silicon Curtain, before I am an Official Apple Employee, I just wanted to share with you my vision of the world I would like to help create. Or rather, a world that I would like you to help create. I am not a professional programmer or computer scientist. I am just a guy who believes computers can be a powerful tool which people can use to change the world. And I want to do what I can to help computers reach their potential, primarily by helping computer -users- reach their potential. Computers are amazing devices in their ability to perform a wide variety of tasks, but all too often this comes at the expense of failing to perform a single task very well - the task that you want it to do! There are many reasons for this mismatch between technology and user needs, but let me focus on one key area where I hope that I - and you - can make a difference. Software development has traditionally been such an expensive, arduous and time consuming process that a) developers had to be experts on technology, not on human problems b) products were designed to satisfy static requirements, not dynamic communities c) the development process largely excluded user involvement Rhapsody, as the only mass-market operating system designed from the ground-up to support object-oriented software development, has the potential to break these compromises and change the rules of software development. If it succeeds, it should provide incredible power and ease of use to end users, developers, and system administrators. I believe Apple has assembled one of the greatest teams of engineers, scientists, and design specialists ever seen in the software industry. I believe they have the tools, support, and vision to pull it off. However, they lack one thing, the one thing that makes the crucial difference between a product that revolutionizes the world and one that merely looks impressive. It is the one thing I most want to give to Apple: You. You, the guy who fights with his mail application, or labors over updating web pages, or dreams of writing a "killer app" that crushes the Big Guys, or merely wants to transform her little corner of the world. I want Rhapsody, along with the tools, programs, and community it engenders, to be the vehicle for making your dreams come true. Not just a dream of how a computer should work, but dreams of community, freedom, learning, love - the stuff people live and die for all over the world. Apple has always been about helping people realize their dreams, and it has made many dreams come true. But in all honesty, Apple was hampered both by its technology, and its culture. Like most great insitutions, it lost sight of its original focus - you, the ultimate consumer - and got caught up in esoteric dreams of greatness. That's changed. Apple has been humbled, and it is reaching out. The acquisition of NeXT, and Apple's commitment to Java and open systems are the signs of a sea change. In a tiny way, bringing in an outsider like myself to work alongside marketing is also a sign of that change. I am no Silicon Valley insider, or hot shot programmer. I'm just like you. I use a computer to balance my checkbook, write love letters, and send email. I've written games for my friends, technical analyses as a student, and mission critical custom apps for a multinational corporation. I've lugged notebooks across the world to share presentations and datafiles with clients, and I've setup PPP and Ethernet networks in my home. And all along, I've often thought: there has to be a better way. If I could, I'd love to sit down with each and every one of you, to hear your dreams for your life and the world, and figure out what small part a computer might play in helping you realize them. But I can't. Even if I had the time, my loyalty is to Apple, to be part of their Voice. I can't be your voice. But, what I can do, what I hope to do, is help you find your voice. To bring people together, where they can fight through the tradeoffs and priorities that are an inevitable part of any democratic community. And then, to be Apple's Eyes and Ears, to see your concerns, hear what you're saying, feel your pain. And then to do whatever I can within Apple, to make sure your voices are heard. To do this, I have asked some friends of mine to setup two mail aliases For developers, devideas@stepwise.com by Don Yacktman, of The Object Foundation For users, userideas@nula.org by Tim Byars of NeXT Users Los Angeles These two have committed to sorting through all the mail that is sent to these lists, and identifying the biggest concerns in those communities. Whether it is a matter of technology, communication, or business. And I have gauranteed them my time and attention to understand those concerns, and my passion and energy to communicate those to the Apple community. I have to be honest with you. Product Managers have no real authority, and as a short-termer and outsider I could very well have less than usual. I am also a realist and a capitalist: Apple has to make money, and soon, and that means shipping a product as quickly as possible, not catering to everyone's hopes and dreams. However, I am enough of an idealist to believe that a company like Apple can only survive if it succeeds in truly understanding its customers. More than almost any other American company, Apple isn't about products, it is about people. It is about changing the world. It did it once with the Apple II, then again with the Macintosh. WIth your help, I believe it may yet have one more revolution in it. I don't know for sure if I can do anything, or even if you can do anything. But I know I have to try. Will you join me? Sincerely, Dr. Ernie Prabhakar Rhapsody Product Manager Designee
From: mpaque@yummyspam.wco.com (Mike Paquette) Newsgroups: comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 18 Jun 1997 14:07:52 -0700 Organization: Electronics Service Unit No. 16 Sender: mpaque@mpaque Distribution: world Message-ID: <5o9in8$35f@mpaque.mpaque> References: <AFCD4CFC-3223A@206.165.44.53> In article <AFCD4CFC-3223A@206.165.44.53> "Lawson English" <english@primenet.com> writes: [marcel's explanation of immediate mode graphics deleted] > > Who said that this was necessary? I did. > But for some of the more interesting features of GX Printing to be > workable, you need to have some kind of display-list mechanism that is > universally defined. Actually' what's needed is a machine parsable description of the output. A display list is just one approach to this. (Be very careful what you say. This is a public forum. The folks posting here are not public figures. Libel and slander laws do apply!) -- Mike Paquette (mpaque AT wco.com ; Yank that yummyspam to reply.) Well, if there *were* anything to say, it would be with the understanding that the PR/Marketing people want to make the announcements on products, so anything I have to say wouldn't actually exist until after then, so what I might have to say now doesn't exist, and what I may say in future can't be said, so theoretically what exists, doesn't, for the immediate future. (With apologies to Joe Straczynski)
From: Fam.Moser@t-online.de (Familie Moser) Newsgroups: comp.sys.next.misc,comp.sys.next.sysadmin,comp.sys.next.programmer,comp.sys.next.software Subject: glib++ for NextStep 3.0 Date: 17 Jun 1997 10:50:05 GMT Organization: Telekom Online Internet Gateway Message-ID: <5o5q4t$q0s$1@news02.btx.dtag.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi everybody, I am looking for a port of glib++ on NS 3.0. At had a look at the sources of glib2.0, but I am afraid that the effort in porting it is too much for me. So I was wondering if there is somebody out there who has a port available. It need not be the latest version, I am just interested in hacking a little bit C++. Any kind of advice or pointer is welcome. Please mail to either martin@shiratori.riec.tohoku.ac.jp or fam.moser@t-online.de. Thanks in advance for your kindness, Martin
From: David Young <daver@jacobs.Geeks.ORG> Newsgroups: comp.sys.next.programmer Subject: IBFramework Documentation? Date: 18 Jun 1997 22:13:05 GMT Organization: Geeks Organizations Message-ID: <5o9mhh$igi$1@darla.visi.com> NNTP-Posting-Date: 18 Jun 1997 17:13:05 CDT After install of 4.2PR2 Developer, I've noticed that it looks like the InterfaceBuilder.framework Resources directory doesn't have any documentation. Did I miss something, or is this normal? The reason I ask is that I have a custom subclass of NSBox on a palette, and I can't seem to get IB to want to drop other objects into it. advTHANKSance Dave -- :: d a v i d y o u n g ::::: smtp dwy@ace.net http www.ace.net :: :: PGP fingerprint :: 89F5 E75D 4749 3FF4 :: ED92 1B6D 9871 9B93 ::
From: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer Subject: Re: GX remote drawing (Was Re: GX OOP class Date: 18 Jun 1997 19:28:00 GMT Organization: Digital Fix Development Message-ID: <5o9cs0$1ah$1@news.digifix.com> References: <AFCD4284-ACA1@206.165.44.53> <5o94gu$f3b$1@brachio.zrz.TU-Berlin.DE> In-Reply-To: <5o94gu$f3b$1@brachio.zrz.TU-Berlin.DE> On 06/18/97, marcel@system.de wrote: >In article <AFCD4284-ACA1@206.165.44.53> "Lawson English" ><english@primenet.com> writes: > >> > Even further, a retained-mode interface gets really useless when you >> > are displaying over a network rather than just blitting to a >> > locally-connected screen. >> >> Who says? > >Foley van Dam, Computer Graphics, 2nd Edition. Anyone want to take bets that Lawson hasn't read this bible? -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: mpaque@yummyspam.wco.com (Mike Paquette) Newsgroups: comp.sys.next.programmer Subject: Re: GX remote drawing (Was Re: GX OOP class Followup-To: comp.sys.next.advocacy Date: 18 Jun 1997 16:18:39 -0700 Organization: Electronics Service Unit No. 16 Sender: mpaque@mpaque Distribution: world Message-ID: <5o9qcf$3m4@mpaque.mpaque> References: <5o9cs0$1ah$1@news.digifix.com> In article <5o9cs0$1ah$1@news.digifix.com> sanguish@digifix.com (Scott Anguish) writes: > On 06/18/97, marcel@system.de wrote: > >In article <AFCD4284-ACA1@206.165.44.53> "Lawson English" > ><english@primenet.com> writes: > >> Who says? > > > >Foley van Dam, Computer Graphics, 2nd Edition. > > > Anyone want to take bets that Lawson hasn't read this bible? Oh, heck. All bets are off on that one. <TONGUE_IN_CHEEK> If Foley and van Dam really knew what they were doing, they'd have just implemented all those examples as GX code, rather than fob off their own incomplete examples as a standard for the world to follow. None of their wimpy code can draw 3D perspective text along arbitrary n-dimensional curves in six different transfer modes and three different languages with different orientations? Remember, Lawson has achieved Oneness with all 1500 pages of GX documentation. He is Beyond such trivialities as a computer graphics primer. </TONGUE_IN_CHEEK> -- Mike Paquette (mpaque AT wco.com ; Yank that yummyspam 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: MaRK_BeSSeY@NeXT.CoM (Mark Bessey) Newsgroups: comp.sys.next.programmer Subject: Re: What needs to be FAT to compile FAT? Date: 18 Jun 1997 23:46:35 GMT Organization: Apple Computer, Inc. Message-ID: <5o9s0r$5ik$1@news.apple.com> References: <Pine.NXT.3.96.970618162432.19293B-100000@cc344191-a.ewndsr1.nj.home.com> Timothy J Luoma <luomat@peak.org> writes > > All the binaries in /usr/ucb are 3-fat on my 4.1 install! > > Why? > > Is this an general installation bug or did I select something wrong? > > I want to be able to compile things MAB, but don't need MAB binaries > myself. What can I thin and what can't I? I'm afraid of will-nilly > 'lipo'-ing files that will mean I can't compile MAB, but also don't want > to have 3-fat around when I don't need it. I think you maybe did something wrong along the way, though I'm not sure what. The installation process should lipo those files automatically. The only fat files you should need to do fat compiles are the libraries and precompiled headers. On my 4.2 system, the only diectories with fat files in them are /lib, /usr/lib, NextLibrary/Frameworks, and /NextDeveloper/Headers. -- Mark Bessey Apple Computer, Inc. -->I DON'T SPEAK FOR APPLE<--
From: goldsmith@apple.com (David Goldsmith) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: Wed, 18 Jun 1997 17:36:13 -0700 Organization: Apple Computer, Inc. Message-ID: <goldsmith-ya02408000R1806971736130001@news.apple.com> References: <5o5jbg$gra$1@news.lth.se> <AFCC64EB-738DD@206.165.44.18> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit In article <AFCC64EB-738DD@206.165.44.18>, "Lawson English" <english@primenet.com> wrote: >Some of those typographical features use the 3x3 matrix transform. Others >require hit-testing to be done for every glyph drawn on the screen, causing >a round-trip from the AppKit to DPS and back with every glyph drawn unless >you implement hit-testing of glyphs on the AppKit side (or eliminate this >from the port of GX typography). Hit testing is implemented on the AppKit side in OpenStep today, and we're not planning to change that. The Window/DPS server is only involved once, to pass the mouse click in to the correct window. The AppKit is responsible for all glyph layout. DPS is just used as a glyph rendering engine. -- David Goldsmith Architect International, Text, and Graphics Department Apple Computer, Inc. goldsmith@apple.com
From: "Lee Byeong-ho" <bhlee@cnt.co.kr> Newsgroups: comp.sys.next.programmer Subject: Re: Creating a Compiled Application in WO Date: Thu, 19 Jun 1997 10:06:32 -0000 Organization: Hansol Telecom Message-ID: <5oavtd$4d5$1@news.hansol.net> References: <5o9fot$345$1@news.hansol.net> Lee Byeong-ho ÀÌ(°¡) <5o9fot$345$1@news.hansol.net> ±â»ç¿¡¼ ÀÛ¼ºÇß½À´Ï´Ù... >Hi all my friends, > >I searched the WO Enterprise samples created with Objective-C (or C, C++, >Perl) compiled code. > >I saw 'Register Now' samples. But it didn't help me not so much ,so If you >have any concerned samples or code, Please send me some tips, I'd love to >hear your ideas. > >Thanks. > >---------------------- >Lee, Byeong-ho. >Yuhan C&T Korea, R&D >mail: bhlee@cnt.co.kr >---------------------- > > > Oh... Is there any help....
From: jon@clarke.exnext.com (Jonathan Hendry) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] Main executable in application folder Date: 19 Jun 1997 02:24:22 GMT Organization: Internet Access Cincinnati 513-887-8877 Message-ID: <5oa58m$6d6$1@ocoee.iac.net> References: <866191282.19621@dejanews.com> <EBpyqL.61n@novice.uwaterloo.ca> <5o9gr9$e9s@cwis-20.wayne.edu> Avi Drissman (adrissmn@gopher.science.wayne.edu) wrote: : David Evans (dfevans@bbcr.uwaterloo.ca) wrote: : : Yes. If you have an app wrapper called "Blah.app" then "Blah.app/Blah" will : : be run. : Does that mean that if I rename an app it stops working? No, just that 'renaming an app' means renaming the appwrapper and the binary contained withing. If you change Foo.app/Foo to Bar.app/Bar, it should work. : (Um, that would be bad.) Um, why? -- Jonathan W. Hendry jon@exnext.com
From: yuyilkhj@trefseewy.com Newsgroups: comp.sys.next.programmer Subject: All New Service for your "Pager".... Date: 19 Jun 1997 03:05:35 GMT Organization: AT&T WorldNet Services Message-ID: <5oa7lv$ms9@mtinsc04.worldnet.att.net> ONLY AVAILABLE IN THE: U.S.A, CANADA VIRGIN ISLANDS, AND PUERTO RICO BE SURE TO GO TO OUR SITE AND CHECK OUT OUR FREE TRIAL. Call everyone and tell them to throw away all of your old home, office, fax, pager, voice-mail, and cellular numbers and give them your New "Virtual Office" 800/888 number! The only number any one will ever need. Are you tired of giving out all of your different phone numbers to everyone? Wouldn’t it be nice to be able to give everyone just ONE phone number that will find you anywhere you are? Even out of town, or in a restaurant, or even on the golf course. Now you can. We have the answer to all your communication needs. It’s called the "Virtual Office". And this new service is loaded. It comes with features like: An automated Call Attendant, Live Call Connect (in real time), Fax Sending, Fax Receiving, Even without a fax machine! E-Mail Notification and Delivery, Without a PC!, Outbound calling, Low Cost Long Distance Service, Inbound/Outbound 800/888 Number, Worldwide Call Transfer, Call Forwarding, Call Screening, Full Service Voice Mail, Nation Wide Pager Notification, Conference Calling, Speed Dialing, Auto Dial, Auto Messaging, Temporary Greeting, Unavailable Greeting, Password Protection, Pager Notification, and No Equipment or Software to buy, "Ever". Priced from $9.95* per month, Plus 10.9 cents per minute per event. That’s less than one phone line would cost per month. For more information please visit our web sight: http://www.mynumber.com * Based on our best priced plan.
From: chris@polaris.scicntr.ortn.edu Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs Subject: IB: Missing Main Menu Date: 19 Jun 1997 02:59:37 GMT Organization: University of Tennessee Message-ID: <chris-ya023580001806972300320001@news.utk.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit I've been working on an app; Its a rather large app and of course has a Main Menu. But the strangest thing is going on: In the app's main nib, there is the icon for the mainMenu. The weird thing is that I can't seem to find this actual menu anywhere, i.e I double-click on the icon, the Inspector switches to Menu Inspector, the name of the app/menu title appears in the 'Title' text field, but the 'physical' list of menu cells appears nowhere. I compile/run the app and it shows up as the app's main menu, just as you'd expect; I switch back to IB and it's still gone. (It's been missing for days, and frankly I'm worried.) Has anyone seen this before ? CB
From: chris@polaris.scicntr.ortn.edu Newsgroups: comp.sys.next.programmer Subject: IBDid/Will errors Date: 19 Jun 1997 03:17:37 GMT Organization: University of Tennessee Message-ID: <chris-ya023580001806972318310001@news.utk.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit I'm trying to get an NS 2.1 palette to work on 3.3, and I run into an 'implicit declaration of function ...' error for two IB functions. Consequently the palette gets built but doesn't work. The code and errors are as follows: (The class being compiled is a subclass of IBInspector. There are only these two methods in it.) + finishLoading:(struct mach_header *)header { NIBDidLoadClass(self, header); return nil; } + startUnloading { NIBWillUnloadClass(self); return nil; } (The errors: ) warning: implicit declaration of function `NIBDidLoadClass' warning: implicit declaration of function `NIBWillUnloadClass' I am not familiar with what is going on here, and haven't found any documentation on these functions. Anyone ? Thanks -- CB
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: yuyilkhj@trefseewy.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5oa7lv$ms9@mtinsc04.worldnet.att.net> Control: cancel <5oa7lv$ms9@mtinsc04.worldnet.att.net> Date: 19 Jun 1997 03:05:38 GMT Message-ID: <cancel.5oa7lv$ms9@mtinsc04.worldnet.att.net> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: d89cb@efd.lth.se (Christian Brunschen) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 19 Jun 1997 08:44:50 GMT Organization: Lund Institute of Technology, Sweden Message-ID: <5oari2$rbb$1@news.lth.se> References: <5o8oia$c4m$1@news.lth.se> <AFCD4CFC-3223A@206.165.44.53> NNTP-Posting-User: d89cb In article <AFCD4CFC-3223A@206.165.44.53>, Lawson English <english@primenet.com> wrote: > > >> In article <AFCC64EB-738DD@206.165.44.18>, >> Lawson English <english@primenet.com> wrote: >> >Christian Brunschen <d89cb@efd.lth.se> said: >> > >> >> >> >> Um, in case you hadn't noticed, Apple has said that GX' typography >> >> features will be added to DPS. >> > >> > >> > >> >Some of those typographical features use the 3x3 matrix transform. >> >> Which _typography_ features require non-affine tranforms ? I am not >> talking about the ability to apply arbitrary 3x3 transform matrices to >> everythin, including text -- that has nothing to do with >> _typography_. >> > >The ability to apply a 3x3 transform matrix to each glyph in a text shape >in layers called a "textface." Which will be _neede_ for what typographic purposes ... ? >> >Others >> >require hit-testing to be done for every glyph drawn on the screen, >> causing >> >a round-trip from the AppKit to DPS and back with every glyph drawn >> unless >> >you implement hit-testing of glyphs on the AppKit side (or eliminate >this >> >from the port of GX typography). >> >> Firstly, since the AppKit is the OO API that shields programmers from >> bare-bones PostScript programming, I would suggest that this is indeed >> a rather good place to put the typography functionality. >> >> Secondly, even if it is implemented on the server side, hit-testing >> would not have to include a round-trip for every glyph drawn, since >> hit-tests for multiple shapes could be easily coalesced, thus >> generating a round-trip for every hit-testing done, not one for every >> glyph. >> > >OK, so how do you do this with text? Every adjacent character might be >overlapping, so how do you manage to reduce the number of round >> trips? Um, by sending the server a) a bunch of chapes, b) a bunch of coordinates to test for, and then tell it 'ok, give me, for each pair of coordinates, the possible hit(s), and which shapes they're in' ... ? Sending data + (possibly lots of) work in the server + receiving results. One round trip. > > >> Thirdly, you seem to be determined to think that, since Apple decided >> not to go with Pure GX, they must have lost their brains completely. I >> am quite confident that >> > >You keep on trailing off... > >And yes, given remarks that Hancock and AMelio have made (e.g. Java can >replace OpenDoc's functionality, we don't need no stinkin' TV ads -they >cost too much, etc), I generally agree with your characterization about >Apple management having lost their brains. I on the other hand beleive that >Or do you think that their performance in overseeing the fall of Apple has >been an example of *competence*? 'The fall of Apple' ? In case you hadn't noticed, Apple is slowly but certainly climbing back out of the red and into the black. MacOS, which had been more or less mothballed due to the promise of Copland, is now once more being developed -- and in its new version is more stable than ever. Also, Rhapsody will give Apple an OS that is both modern, stable, and widely deployed. Yes, _widely_ -- not as widely as MacOS, or Miscrosoft's various OSes, but widely nonetheless. And >> >And if NSImage is a first class graphical object, howcome there's no >> >display list mechanism ala a GX picture shape? >> >> Um, because DPS isn't a retained-mode graphics engine ... ? >> >> >Or is there, and does it include bitmap images? >> >> Bitmap images can be told to draw themselves to a PostScript view with >> one (1) Objective-C method call. This will typically be done inside >> the view's '- drawSelf:' method, which is called when the view needs >> updating. So far, programmers under NEXTSTEP and OPENSTEP have managed >> to keep track of whatever objects, or PostScript calls, they need to >> for this to happen, without the help of a Display List ala GX. >> >> Immediate-mode graphics work _very well_, and in a networked context >> they avoid a _number_ of problems that retained-mode graphics systems >> have when the machine which composes the image, and the machine that >> blits the pixels to the screen, are not the same. >> > >Who said that this was necessary? Who said what was necessary -- being able to run an application and its display on different machines ? Excuse me, but are you completely out of synch with reality ? Microsoft is working on what they call 'Windows Terminals' -- like an X terminal, only using Windows' graphics and GUI APIs. Such products as NTrigue, which allows you to run a Windows NT application on one machine, and use another machine for display, already provide some of that functionality. X has always had it, and that is one of the reasons why some products have traditionally been available only in Unix environments -- where you could have a _large_ server in the middle, which could cope with running the application ... and a bunch of inexpensive X-terminals -- which might simply be now-obsolete workstations which have been relegated to this simpler, less taxing task -- in front of which the users would sit and do their work. DPS, under NEXTSTEP/OPENSTEP, has always done this as well, transparently to both user, application, and display server. Are you advocating that this functionality be removed ? > > >> Giving explicits commands that are executed one-at-a-time is always >> going to be simpler, and more efficient, to implement _correctly_ than >> any solution where a database has to be transported, or two databases >> have to be synchronized, over a network. >> > >Who said that this was necessary? In a retained-mode graphics environment, the application builds a database of things it wants to be drawn on-screen, and the display server draws things from that database on-screen. _Iff_ both application and display server are on the same machine, this can be done by _sharing_ the database between application and discplay server. If this is _not_ the case, then we can either store the database _once_, in the aplplication or in the display server; in either case, the party which does _not_ have the database will need to access the database through the network. Or we can keep a copy of the database in both the application and in the server, in which case we need to keep both copies synchronized with each other, lest the display server draw things that are not what the application intended. > >> >> Besides, if you want to have an explicit display list API, noone is >> going to stop you from writing a 'DisplayListView' which holds a >> display list, and in its 'drawSelf' method draws each item in the >> display list to the screen. You could then make all _your_ views >> subclasses of DisplayListView, and you'd never have to touch a >> PostScript operator again. Remeber, Rhapsody is going to include >> classes for Bezier paths, Graphics contexts, etc. No need for you to >> touch those disgusting PostScript operators. >> > >But for some of the more interesting features of GX Printing to be >workable, you need to have some kind of display-list mechanism that is >universally defined. Which more interesting features would that be ? I have already made what I beleive to be a rather good argument against placing functionality to modify the printing output on its way from the application, to the printer, arguing that this functionality would be better offered in the application itself. > >> > >> >BTW, having 3D perspective available for vector shapes and text >> >> 2D is 2D, 3D is 3D. GX offers 2D. Saying '3D perspective' is spreading >> a _falsehood_. >> > >Calling me a liar? Please explain why Loren Petrich's analysis of GX's 3D >perspective mapping is invalid. Please give me a pointer to this analysis, then. >GX implements a 3x3 matrix that allows one >to apply a 3D perspective to 2D textures such as vector graphics, bitmaps >and text. Perhaps you would explain, or even better give an example (ie, images) of, exactly what you mean by this '3D perspective' ? > > >> >is >> >considered a big plus for high-end DTP apps, even though it isn't "real" >> >3D, or so the reviewers of the latest Adobe apps have indicated. >> >> Perhaps you would care to include some pointers to these reviews so >> that we might read them for ourselves, rather than hearing your >> interpretation ? > >Every high-end DTP person that *I* know wants 3D perspective capabilties. I asked for pointers, not assertations. > >In fact, Level 3 PostScript supports "3D effects," so obviously Adobe felt >that this was important to support. BTW, is it a _falsehood_ for Adobe to >call 3D perspective "3D effects?" You are making the mistake of assuming that Adobe will be using the same techniques used by GX. > >> >> (same hideously long advertising .sig snipped) >> > (quite nice .sig, left in place :) >------------------------------------------------------------------------- >Build a better mousetrap, yes, but then market it like crazy, or instead of > >catching mice you'll only collect dust. -David Yeargin >------------------------------------------------------------------------- > > Best regards, // Christian Brunschen
From: d89cb@efd.lth.se (Christian Brunschen) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX remote drawing (Was Re: GX OOP class Date: 19 Jun 1997 09:32:11 GMT Organization: Lund Institute of Technology, Sweden Message-ID: <5oauar$ruv$1@news.lth.se> References: <5o5jbg$gra$1@news.lth.se> <AFCD4284-ACA1@206.165.44.53> NNTP-Posting-User: d89cb In article <AFCD4284-ACA1@206.165.44.53>, Lawson English <english@primenet.com> wrote: >Christian Brunschen <d89cb@efd.lth.se> said: > >> Also, the NSBezier class, which has been announced, looks suspiciously >> similar to what you claim Rhapsody won't have .... >> > >Are you so familiar with GX's capabilities, as described in about 1500 >pages of manuals that you can make that claim? Note that I didn't claim that NSBezier would do everything that GX does. It will, however, provide some of the functionality you have condemned AppKit for not having. > >With GX, I can do a "GXSImplifyShape(myBitmapShape)", and if myBitmapShape >is only one color, it will turn it into a GXRectangleShape. Will NSImage >and NSBezier support such a thing? I don't know. Ask Apple. Lobby them, even. > >With GX, I can turn off dithering/halftoning on a per-shape basis. How's >NSImage and NSBezier do for that level of coordination? With DPS, you can change the dithering/halftoning behaviour as you see fit ... >There is a LOT of stuff you can do with GX. A NeXT developer told me that >GX is "over-engineered" but a NeXTStep hobbyist programmer who was hoping >to use GX in Rhapsody snorted at that and commented that it was "powerful, >not 'over-engineered.'" >> Even further, a retained-mode interface gets really useless when you >> are displaying over a network rather than just blitting to a >> locally-connected screen. > >Who says? > >Where shall we put the database ? > >Where it has always been: in the GX Graphics Heap. And where, pray tell, is 'the GX Graphics Heap' in a networked environment where Application and Display are on physically different machines, connected through a network ? > > In the >> application ? > >No. Where it has always been: in the GX Graphics Heap. And where, pray tell, is 'the GX Graphics Heap' in a networked environment where Application and Display are on physically different machines, connected through a network ? > > Then the graphics subsystem has to talk to the display >> somehow -- either by blitting large images (which will suck up >> bandwidth, be highly sensitive to latency, and generally suck), or by >> sending some sort of commands to a display server -- you know, like X >> or DPS already do it .. ie, no gain over an immediate-mode interface, >> but rather a loss, since retained-mode has to be emulated over >> immediate-mode. > >Obviously, you don't know squat about GX. Actually, this is a quite correct statement. But it still doesn't address my question. My questions are not, as you might think, rhethorical. If you have a good answer, I am more than willing to listen. > >> Or should we put the database in the display ? > >Obviously, you don't know squat about GX. Obviously, you're not interested in a debate, but only in getting your own way. > >That way, whenever we >> want to change something in the database, we have to transmit >> whatever changes we make to the database, across the network, making >> all database changes susceptible to network bandwidth limitations, and >> latency. > >Obviously, you don't know squat about GX. Well, your not knowing anything about PostScript didn't stop you from making comments about its performance. > >> Or perhaps we should have a copy of the database in the application, >> and one in the display .. ? Thus cutting the apllication some away >> from the latency and bandwidth limits in the previous case, but >> wasting memory for two identical databases ... and we still have to >> keep them synchronized, resulting in lots of fun network traffic. >> > >Obviously, you don't know squat about GX. I beleive your record is broken. > > >If you want to draw into an application's window over a network by using >GX, you need to do a few things: > >1) Establish some kind of communications channel over the network to send >function calls to the copy of GX that exists on that MacOS machine. >2) procure the appropriate GX Graphics Heap reference from an existing >application. >3) procure the appropriate GX viewport >4) set the remote GX heap to the appropriate heap >5) make the calls via the communications channel to the copy of the GX >library on the other machine, with reference to the window's viewport >object, when needed (which might be only once -to set the default viewport, >if it isn't already set). > Basically, you would place the database in the display server, then. _Say so_ instead of insulting me needlessly. All my comments apply: The application needs to talk to the database (as managed by GX), over the network (through the communications channel) for everything it wanted to do to the database. Thus incuring the usual penalty for talking over a network. Same way as commands to the DPS server are sent over the network. > >The communications channel would be AppleSCript for convenience' sake. Probably there would be an alternate, binary streamed format, much like the X protocol ... or the DPS stream (but affecting the GX Shape database, rather than being immediate-mode-drawing instructions). > >Any app that wanted to allow remote drawing would simply respond to the >query for the appropriate graphics heap, windows viewport, etc. Or, even better, we can make _all_ graphics go through a given communications channel; that way, we need just point that communications channel at whatever display server we want to draw on, and everything will work transparently. Like in X, or DPS. >The reason why I say that you don't know squat about GX is that GX >*ALREADY* maintains the database in its own private segment of memory that >no application has access to (leaving aside MacOS's non-protected memory >space, of course). Well, that is what I was _asking_. You just, apparently, didn't read my question. Which says more about you than it does about me. > >Any modifications made to the retained-mode shapes would be made in the >private memory space created by GX on the remote machine. This is what I meant by 'accessingn the database', yes. > >In fact, it is funny that you should mention this remote drawing issue. >ONce I get the GXFCN debugged and shipping, my NeXT project is to create a >GX OSA that uses FaceSpan What is this ? sounds interesting. Please do elaborate. >for the windows. Developers can actually create >faceless apps that draw via FaceSpan, or they can create quick and dirty, >drawing scripts, or pictures, or whatevers from within any app that can >attach an AppleScript and print them out using GX Printing (or whatever >kludge Apple comes up with). Sounds interesting, but without a pointer to some more info this is, unfortunately, just a bunch of buzzwords tow me. > >Since AppleScript is already networkable, this would make my OSA a >network-server for GX. > >[thanks for the idea, Eric King!] > >Of course, since AppleScript is slow, this solution would be slow, but >obviously the same principle could be used for a speedier communications >channel... > >... like the one used in Rhapsody, maybe? > >And of course, one could always replace DPS's forth-like language with Java >and use Java to call GX or Taligent graphics... > >[thanks to my brother for this idea!!!] Or you could write the 'network server' for it so that it accepts commands via NeXT^H^H^H^HApple's Distributed Objects and/or CORBA, so that your server could be used from all manners of languages. Yes, very nice. Go ahead and do it, noone is trying to stop you. If you are so _extremely_ attached to GX, why don't you try to scratch together some cash and _buy_ it from Apple ? You could then add a whole View Subclass hierarchy to Rhapsody. Remember, there's going to be a nice and easy way to draw directly to the screen, or to a window, bypassing DPS -- NSDirectScreen (NeXT's Interceptor technology). Within this 'GXKit' you could do _whatever_ you wanted -- including on the printing side, as long as what comes out on the end is PostScript. However, Apple has to get a product to market, _fast_. DPS is a fast, well-integrated graphics engine that delivers WYSYWIG from screen to printer, something that people in the graphics and publishing businesses appear to find useful. Yes, I know GX can generate fine PostScript, but if a large part of your input is (encapsulated) PostScript, and your ourput is PostScript, then having PostScript in the middle, too, makes an awful lot of sense. Insisting that Apple dump DPS in favor of GX is not going to change anything. > >Hey! What an idea: make Rhapsody the premier internet graphics system by >making its main graphics engine the same 2D graphics engine that will be >used in Java applets starting in Java 2.0... To quote again from Sun's press release: --- snip --- Mountain View, Calif. - May 27, 1997 - - Sun Microsystems, Inc. today announced that it has licensed technologies from Taligent, Inc., a wholly-owned subsidiary of IBM Corporation. These technologies include bi-directional line layout for internationalization of text and high-level graphic capabilities for combining and manipulating geometric shapes. Both will be incorporated into Sun's implementation of the Javatm 2D API. --- snip --- Sun has licensed a few important technologies from Taligent, but that most certainly does _not_ make the Java(TM) 2D API use Taligent's graphics engine. Sorry to burst your bubble. While Sun has also said they will not be using Adobe's Bravo, I beleive they are more likely to go with a graphics rendering engine of their own. > >Call it Apple Internet Graphics, or some such... > >BTW, I don't know how Taligent handles allowing other graphics systems to >draw directly to the screen, but GX allows one to specify a clip shape for >a view device, which means that an app that wanted to draw directly to the >screen could reserve a section of the screen for its own use by supplying a >clip-shape, and not even window-frames drawn using GX would obscure it. >Huh! Sounds like Interceptor, but better since it would never interfere >with anything GX does and could still track the cursor and so on... In what way would Interceptor interfere with what DPS does ? How would it stop the cursor from tracking ? Remember, events are handled in their own separate way. They are _not_ lumped in with DPS the way you seem to think they are. > >But you can do all of this using Rhapsody graphics based on DPS, despite >the kludgeyness of it all, so therefore DPS + AppKit is "just as good" as >GX... Yes, you _can_, and it's not even a kludge like you try to make it look. Furthermore, not all of Rhapsody's graphics will be based on DPS. Interceptor isn't part of DPS, for instance. Neither is QuickDraw3D, nor QuickTime. > >Yeparoonies, them folks at Apple shore know what they're doing with this GX >vs DPS vs Taligent vs whatever issue... Actually, they do. They are using a mature product with wide industry aceptance, which also happens to be somewhat of a de-facto standard in the publishing industry (PostScript); rather than choosing one of two largely unproven graphics systems (GX and Taligent). (hideously long signature once more snipped) Best regards // Christian Brunschen
From: martin@wise-02.wiwi.tu-dresden.de (Martin Rose) Newsgroups: comp.sys.next.programmer Subject: Identifying Directories with Objective C Date: 19 Jun 1997 12:43:41 GMT Organization: TU Dresden (URZ) Message-ID: <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> Hi NeXTSTEP-Experts, Does anybody know how to find out whether a directory-entry is a subdirectory or a "real" file?? There would be a solution to check if a dot '.' appears in the filename, but this criteria would not be enough, because there might be filenames without any extension... Do directory-entries own a certain flag that determines being a directory XOR a file? I'd be glad to hear from you soon & much thanks in advance! Martin Rose martin@wise.wiwi.tu-dresden.de mrose@rcs.urz.tu-dresden.de
From: michael@rumah.pc.my (Michael Olan) Newsgroups: comp.sys.next.programmer Subject: gcc.2.7.2 & PB Date: 19 Jun 1997 13:01:21 GMT Organization: Personal Message-ID: <5obaj1$27g@rumah.pc.my> I've had gcc.2.7.2 installed in the usual /usr/local for quite awhile. I have quite a few template classes I'd like to use in some PB projects, but the Next gcc is from the pre-template days. What's the most painless way to make this work? Replace the old gcc, muck around with makefiles, or what? If I use gcc-2.7.2 in PB it doesn't know where all the Next libraries are. Ok, so I'm pretty ignorant about makefiles & all, so maybe I just need to tell it where they all are located? TIA, Mike
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 19 Jun 1997 09:19:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFCEAE9A-9F39@206.165.44.88> References: <5oari2$rbb$1@news.lth.se> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.mac.programmer.misc, nntp://news.primenet.com/comp.sys.next.programmer MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Christian Brunschen <d89cb@efd.lth.se> said: > >But for some of the more interesting features of GX Printing to be > >workable, you need to have some kind of display-list mechanism that is > >universally defined. > > Which more interesting features would that be ? > > I have already made what I beleive to be a rather good argument > against placing functionality to modify the printing output on its way > from the application, to the printer, arguing that this functionality > would be better offered in the application itself. > And I would argue that you are wrong. At least my DTP-ing friends like the design of GX, where you can pass the output through GX extensions and modify a print job automatically in ways that the developers of the original application never intended. Or are you an advocate of one-size-fits-all apps? Most NeXT-advocates appear to me, which is why OpenDoc doesn't appeal to them, I guess. ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX remote drawing (Was Re: GX OOP class Date: 19 Jun 1997 09:41:01 -0700 Organization: Primenet Services for the Internet Message-ID: <AFCEB3B5-1D245@206.165.44.88> References: <5oauar$ruv$1@news.lth.se> 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 Christian Brunschen <d89cb@efd.lth.se> said: > Basically, you would place the database in the display server, > then. _Say so_ instead of insulting me needlessly. > *I* wouldn't place the database anywhere. The GX graphics heap resides where it always does, in a private memory area that reveals itself to the rest of the world as a database. You don't use pointers to access things with GX, you use object references, like "1" or "2" -pointers to "1" or "2" actually. > All my comments apply: The application needs to talk to the database > (as managed by GX), over the network (through the communications > channel) for everything it wanted to do to the database. Thus incuring > the usual penalty for talking over a network. Same way as commands to > the DPS server are sent over the network. > And the whole idea of a retained mode engine, whether it is evoked via a communications channel or via a standard function call, is to *reduce* the number of times that the engine needs to be evoked per shape drawn. A GX shape encapsulates info about color, transfer mode, transform, view destination, etc. Once these are set, a single call, GXDrawShape(myShape), will draw that shape, whether it be a rectangle or curve or bitmap. You can even construct an arbitrarily complex list of shapes called a "picture shape," and draw ALL of those, regardless of which monitor, off-screen destination, window, or combination thereof, that this picture is to be drawn to, merely by saying "GXDrawShape(myPicture)" Lessee... A picture might include shapes with myriad colors, transformations, transfer modes (composite modes), and so on. Under the DPS model, each component of each shape would need to be created on one machine and the individual elements of the shape would need to be sent over the communications channel, one-at-a-time... You can "retain" SOME information via dictionaries and so on, but the DPS model says that you can't store global-state-modifying information this way, which means that colors, transforms, transfer modes, dithering options, clip shapes, etc., all need to be sent, one-at-a-time over the communications channel. The reason why I was "insulting" you was that THIS IS THE WHOLE POINT OF A RETAINED MODE graphics engine: reduce the number of times the system is evoked by the user, whether via an API or via a communications channel. If you are going to use the term, at least make some pretense of understanding it. Or read a book. -------------------------------------------------------------------------- Any sufficiently advanced magic will be indistinguishable from technology -my corollary to Clarke's Law (AFAIK Mercedes Lackey got it from me at a World Fantasy Convention) --------------------------------------------------------------------------
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: Identifying Directories with Objective C Date: 19 Jun 1997 17:17:24 GMT Organization: 21st Century Software, New York City Sender: daver@ts1-9.nj.cnct.com Message-ID: <5obpj4$574$1@darla.visi.com> References: <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 19 Jun 1997 12:17:24 CDT Cc: martin@wise-02.wiwi.tu-dresden.de In <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> Martin Rose wrote: > Does anybody know how to find out whether a directory-entry is a subdirectory > or a "real" file?? There would be a solution to check if a dot '.' appears in > the filename, but this criteria would not be enough, because there might be > filenames without any extension... If you're on UNIX and not using OpenStep, see the stat(2) system call. man stat from a shell prompt. If you are using OpenStep, see the NSFileManager class. -- :: 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: d89cb@efd.lth.se (Christian Brunschen) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX OOP class proposal & contest (was Re: GX game graphics Date: 19 Jun 1997 18:01:35 GMT Organization: Lund Institute of Technology, Sweden Message-ID: <5obs5v$5mr$1@news.lth.se> References: <5oari2$rbb$1@news.lth.se> <AFCEAE9A-9F39@206.165.44.88> NNTP-Posting-User: d89cb In article <AFCEAE9A-9F39@206.165.44.88>, Lawson English <english@primenet.com> wrote: >Christian Brunschen <d89cb@efd.lth.se> said: > >> >But for some of the more interesting features of GX Printing to be >> >workable, you need to have some kind of display-list mechanism that is >> >universally defined. >> >> Which more interesting features would that be ? >> >> I have already made what I beleive to be a rather good argument >> against placing functionality to modify the printing output on its way >> from the application, to the printer, arguing that this functionality >> would be better offered in the application itself. >> >And I would argue that you are wrong. Of course you would. >At least my DTP-ing friends like the >design of GX, where you can pass the output through GX extensions and >modify a print job automatically in ways that the developers of the >original application never intended. One word: Services. If you have a) your data in a format that can be copied&pasted to/from the pasteboard, and b) an application that supports Services for that type of data, you can write services which do exactly _anything_ with your data, in ways the developers of the original application never intended or thought possible. Of course, the application writer wouldn't actually have to write any code for this -- the hypothetical GXKit would simply define the necessary Pasteboard data type, and the GXView class would implement the necessary methods for providing and/or receiving data from services. This is obviously much more flexible than putting filters on the way to the printer, because I can now perform whatever transformations I want on my data not only when it goes to the printer, but also if I just want to view the result on-screen. Or if I want to email the result to a co-worker. To pick up your 'translate japanese glyphs to english words' example again, why would I want to limit that translation to be made only on the way to the printer ? Perhaps I am sinply providing the conversion as a favour to my friend, Alex, who wrote the original document and wants to proof-read. But, alas, he only has access to a PC running Windows where he currently is, so he mails me the document, I convert it for him, and mail him the result back ... not a single printer in sight. > >Or are you an advocate of one-size-fits-all apps? Never. A bunch of applications that provide services to each other, that are scriptable, will work _wonders_ together in ways that no monolithic app could. You keep making the rather large mistake that you consider everyone who doesn't advocate your particular technology, to be anti-cool-technology, or not having any grasp on what is useful. Services, which NeXT has had for _years_, is a more general, more flexible, solution than GX's ability to insert filters on the way from application to printer. > >Most NeXT-advocates appear to me, which is why OpenDoc doesn't appeal to >them, I guess. No, OpenDoc doesn't appear to most NeXT advocates because they already _have_ a system that works very well for making applications work _together_, without the need for embedding them within each other: Services. Personally, I think OpenDoc sounded cool -- idea-wise. However, the fact that the implementation was C++-based immediately gave me a headache, for an Open- Doc like system could not possibly work without a highly dynamic runtime system to support it -- like Objective-C has .. or Java. >------------------------------------------------------------------------- >Build a better mousetrap, yes, but then market it like crazy, or instead of > >catching mice you'll only collect dust. -David Yeargin >------------------------------------------------------------------------- > > Best regards // Christian Brunschen
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: Identifying Directories with Objective C Date: 19 Jun 1997 18:17:08 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5obt34$rrk$2@owl.slip.net> References: <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit Martin Rose < martin@wise-02.wiwi.tu-dresden.de> wrote: > Hi NeXTSTEP-Experts, > Does anybody know how to find out whether a directory-entry is a subdirectory > or a "real" file?? There would be a solution to check if a dot '.' appears in > the filename, but this criteria would not be enough, because there might be > filenames without any extension... > Do directory-entries own a certain flag that determines being a directory XOR > a file? > I'd be glad to hear from you soon & much thanks in advance! This is OS specific. See your OS's man pages. I think it stat() or dir() or something under BSD. -- Steve Dekorte - OpenStep consultant - San Francisco
From: d89cb@efd.lth.se (Christian Brunschen) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX remote drawing (Was Re: GX OOP class Date: 19 Jun 1997 18:20:40 GMT Organization: Lund Institute of Technology, Sweden Message-ID: <5obt9o$5va$1@news.lth.se> References: <5oauar$ruv$1@news.lth.se> <AFCEB3B5-1D245@206.165.44.88> NNTP-Posting-User: d89cb In article <AFCEB3B5-1D245@206.165.44.88>, Lawson English <english@primenet.com> wrote: >Christian Brunschen <d89cb@efd.lth.se> said: > >> Basically, you would place the database in the display server, >> then. _Say so_ instead of insulting me needlessly. >> > >*I* wouldn't place the database anywhere. The GX graphics heap resides >where it always does, in a private memory area that reveals itself to the >rest of the world as a database. You don't use pointers to access things >with GX, you use object references, like "1" or "2" -pointers to "1" or "2" >actually. Who mentioned pointers ? > > >> All my comments apply: The application needs to talk to the database >> (as managed by GX), over the network (through the communications >> channel) for everything it wanted to do to the database. Thus incuring >> the usual penalty for talking over a network. Same way as commands to >> the DPS server are sent over the network. >> > >And the whole idea of a retained mode engine, whether it is evoked via a >communications channel or via a standard function call, is to *reduce* the >number of times that the engine needs to be evoked per shape drawn. > >A GX shape encapsulates info about color, transfer mode, transform, view >destination, etc. Once these are set, a single call, GXDrawShape(myShape), >will draw that shape, whether it be a rectangle or curve or bitmap. > >You can even construct an arbitrarily complex list of shapes called a >"picture shape," and draw ALL of those, regardless of which monitor, >off-screen destination, window, or combination thereof, that this picture >is to be drawn to, merely by saying "GXDrawShape(myPicture)" Yes, but first you have to _construct_ your database. And then, if you are changing anything, you have to modify the contents of your database. Yes, the 'GX Heap'. Which means communicating with the server over the wire. > >Lessee... > >A picture might include shapes with myriad colors, transformations, >transfer modes (composite modes), and so on. Under the DPS model, each >component of each shape would need to be created on one machine and the >individual elements of the shape would need to be sent over the >communications channel, one-at-a-time... > >You can "retain" SOME information via dictionaries and so on, but the DPS >model says that you can't store global-state-modifying information this >way, which means that colors, transforms, transfer modes, dithering >options, clip shapes, etc., all need to be sent, one-at-a-time over the >communications channel. You can 'retain' any kind of information you like in PostScript dictionaries. A Path in itself doesn't store the global-state-modifying-info, no, but that's because the same path can be reused with different global state info. In a PostScript dictionary -- or any other kind of PostScript variable -- I can store any kind of data PostScript can handle. Want to store line width, dashing, transformation matrix, dithering options ? Yes, you can. Easily. Fact is, you can write a full-blown retained graphics system _in PostScript_ if you want to. Furthermore, yes, there are cases when a retained-mode graphics system will generate less network traffic then an immediate-mode system. However, the inverse is _also_ true. A restructuring of your database, which might hardly affect the on-screen display at all, will generate more traffic than if an immediate-mode system were used. You see, in an immediate-mode system noone tries to second-guess the application programmer's needs of possible storage of graphics. > >The reason why I was "insulting" you was that THIS IS THE WHOLE POINT OF A >RETAINED MODE graphics engine: reduce the number of times the system is >evoked by the user, whether via an API or via a communications channel. > And of course, this WHOLE POINT is not appropriate in all cases. >If you are going to use the term, at least make some pretense of >understanding it. > >Or read a book. *shaking my head* Your factual arguments really must be running out for you to have to stoop to insults. > >-------------------------------------------------------------------------- >Any sufficiently advanced magic will be indistinguishable from technology >-my corollary to Clarke's Law >(AFAIK Mercedes Lackey got it from me at a World Fantasy Convention) >-------------------------------------------------------------------------- > When would this event have taken place ... ?
From: boehring@biomed.ruhr-uni-bochum.de (Daniel Boehringer) Newsgroups: comp.sys.next.programmer Subject: Re: Identifying Directories with Objective C Date: 19 Jun 1997 17:50:30 GMT Organization: Ruhr-Universitaet Bochum, Rechenzentrum Message-ID: <5obrh6$pvv$1@sun579.rz.ruhr-uni-bochum.de> References: <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> hi martin, try: str=[[MiscString alloc] initString:"/der/pfad/der/datei"] isDir=[str isFileOfType: Misc_Directory]; if you have the misckit installed, though :-)
From: rbarris@quicksilver.com (Rob Barris) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.programmer Subject: Re: Prelude2Rhapsody on black hardware! Date: Thu, 19 Jun 1997 11:52:53 -0700 Organization: Quicksilver Software, Inc. Message-ID: <rbarris-ya023280001906971152530001@news.intelenet.com> References: <rbarris-ya023280001106971149270001@news.intelenet.com> <5o3mv4$rh7$1@news.platinum.com> <rbarris-ya023280001606971056060001@news.intelenet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Finally got to the point where I can install and boot "Prelude To Rhapsody" OpenStep 4.2 from my Mac's old Conner 1GB hard drive, on the NextStation-Turbo. The successful sequence (I'll spare you the un-successful attempts) a. detach internal Seagate 240MB drive b. re-attach termination resistors, and install Conner 1080S at ID#0 c. attach CDROM drive at ID#1 d. insert NextStep 3.3 User CD (only bootable CD, it seems) e. power on machine, enter low level monitor (command-backquote) f. type "bsd(1,0,0)rootdev=sd1a" as in the "installing NextStep" book g. boot from CD and install NS3.3 h. reboot from HD i. insert OpenStep 4.2 CDROM and run installer (but make sure you are logged in as root first, not as 'me') Whew! The main hurdles that I ran into were 0) the Prelude 4.2 CD is not bootable on black hardware it seems. 1) the Conner wouldn't play friendly on the SCSI bus when cabled up externally. Getting the old Seagate off the bus and hooking the Conner up internally made all the difference. 2) RTFM, it is crucial when doing this to have the CDROM at a higher SCSI ID, and to issue the bsd command as above (from memory, pretty sure that is the correct syntax for my example of HD=0, CD=1) Now that I have 4.2 User installed (and still several hundred MB's left on this drive) I can go ahead with installing 4.2 Developer and write some programs! Thanks to all who responded and helped out. Rob Barris Quicksilver Software Inc. rbarris@quicksilver.com * Opinions expressed not necessarily those of my employer *
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX remote drawing (Was Re: GX OOP class Date: Thu, 19 Jun 1997 16:04:51 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <oneN3XW00iVD069ts0@andrew.cmu.edu> References: <5oauar$ruv$1@news.lth.se> <AFCEB3B5-1D245@206.165.44.88> In-Reply-To: <AFCEB3B5-1D245@206.165.44.88> Excerpts from netnews.comp.sys.next.advocacy: 19-Jun-97 Re: GX remote drawing (Was .. by "Lawson English"@primene >> All my comments apply: The application needs to talk to the database >> (as managed by GX), over the network (through the communications >> channel) for everything it wanted to do to the database. Thus incuring >> the usual penalty for talking over a network. Same way as commands to >> the DPS server are sent over the network. > > And the whole idea of a retained mode engine, whether it is evoked via a > communications channel or via a standard function call, is to *reduce* the > number of times that the engine needs to be evoked per shape drawn. They have nothing particularly to do with each other. Reducing communications latency by eliminating unnecessary round-trips between a client process which wants drawing to be done on it's behalf and the server process which performs the drawing is a good idea in general, and you can find both immediate-mode and retained-mode graphic engines which try to optimize communications. Of course, the problem is that retained-mode graphic engines run into problems trying to run in a distributed fashion due to the reasons already explained. [ ... ] > A picture might include shapes with myriad colors, transformations, > transfer modes (composite modes), and so on. Under the DPS model, each > component of each shape would need to be created on one machine and the > individual elements of the shape would need to be sent over the > communications channel, one-at-a-time... > > You can "retain" SOME information via dictionaries and so on, but the DPS > model says that you can't store global-state-modifying information this > way, which means that colors, transforms, transfer modes, dithering > options, clip shapes, etc., all need to be sent, one-at-a-time over the > communications channel. You completely don't get it, do you? The whole reason DPS _doesn't_ put that kind of information in the drawing engine itself is to avoid round-trip communication latency between the client and server. That information (or much of it, away) is kept in the client as state within various AppKit objects like (NS)Views, and (NS)Text objects, and so forth. That means your client application can access that information directly without having to get a response back from DPS. In the event that GX ever becomes network remoteable (ie, capable of drawing to other machines on the network), the fact that most of the "interesting" information is kept by GX means that you will have _more_ communication latency when your client wants to access things like clipping regions, glyph characteristics, colors, and so forth. > The reason why I was "insulting" you was that THIS IS THE WHOLE POINT OF A > RETAINED MODE graphics engine: reduce the number of times the system is > evoked by the user, whether via an API or via a communications channel. And you're mistaken if you believe that a retained mode graphic engine does a better job of reducing communications latency when remotely displaying graphics than an immediate mode graphic engine in the event that the client wishes to examine and/or manipulate the state of the various graphic objects present in the system. > If you are going to use the term, at least make some pretense of > understanding it. > > Or read a book. "Do as I say, not as I do?", huh? Gods, what hypocracy.... -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: Identifying Directories with Objective C Date: Thu, 19 Jun 1997 16:23:27 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Message-ID: <IneNIzu00iVD069lY0@andrew.cmu.edu> References: <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> In-Reply-To: <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> Excerpts from netnews.comp.sys.next.programmer: 19-Jun-97 Identifying Directories wit.. by Martin Rose) > Does anybody know how to find out whether a directory-entry is a > subdirectory or a "real" file?? Take a look at 'man stat' for the stat() and/or fstat() functions, and the S_IFDIR bit of st_mode field of a (struct stat *), at least under Unix and/or the POSIX environment of Windows NT. > There would be a solution to check if a dot '.' appears in the filename, but > this criteria would not be enough, because there might be filenames > without any extension... Hint: the special entries '.' and '..' have periods in them, and those are directories. Looking at a filename in order to guess whether it's a directory is the wrong approach, and is guaranteed to be non-portable. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: bozack@blkbox.com Newsgroups: comp.sys.next.programmer Subject: posix libraries Date: 19 Jun 1997 21:48:50 GMT Organization: The Black Box, Houston, Tx (713) 480-2686 Message-ID: <5oc9g2$ca0@news.blkbox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii I noticed OPENSTEP 4.1 comes with no uname binary, so I thought I'd write one. So I looked at the man page for the uname() system call and slapped out some code that would theoretically work .. until I found that the uname() isn't present in the libraries that are linked automatically. I also tried compiling tcl-7.6p2 (so I could compile eggdrop). the tcl library is archived fine, but tclsh fails during linkage since it can't find mkfifo() .. fine, I say. I didn't need tcslsh ANYWAY. All I needed was the library. So I go to compile eggdrop, but it fails during linkage because references to the mkifo() call are present in the library, but the call itself is nowhere to be found. I could theoretically remove all references to mkfifo() in tcl (it's only referenced twice), but you never know when a tcl app will pop up that needs to use fifos! Both of these (uname and mkfifo) are posix only system calls. OPENSTEP 4.1 doesn't come with a posix library, and simply feeding the compiler a -posix flag isn't going to do much good. So where do I go from here to find these magic system calls? Thanks in advance! Dan
From: robertg@trg.saic.com (Robert Gottlieb) Newsgroups: comp.sys.next.programmer Subject: PB.project parsing solved MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Message-ID: <12722230@NEWS.SAIC.COM> Date: Thu, 19 Jun 1997 22:08:08 GMT Hi ya'all, I figured out how to parse the PB.project in NS 3.3. Read the file into an NSString. Ask NSString for propertyListFromStringsFileFormat. This will return an NSDictionary. You're done! L8r, Robert -- robertg@trg.saic.com NeXTMail preferred, others accepted "I believe that what I'm feeling Changes how the world appears" - Neil Peart (Rush) "The above comments/opinions are not that of SAIC"
From: macintsh@bu.edu (John Siracusa) Newsgroups: comp.sys.next.programmer Subject: NSClassNames -> ??ClassNames Date: 19 Jun 1997 23:30:12 GMT Organization: Boston University Message-ID: <5ocfe4$h5q@news.bu.edu> I was told by a former NeXT (now Apple) employee at PC Expo this week that all the class names in OpenStep like NSButton, etc. are going to be renamed now that "NS" no longer has any bearing on the OS (Rhapsody). My question is, what nomenclature will they use? "CClassName" like Microsoft? Class names based on libararies like the EOF classes seem to have? -----------------+---------------------------------------- John Siracusa | If you only have a hammer, you tend to macintsh@bu.edu | see every problem as a nail. -- Maslow
From: "Lawson English" <english@primenet.com> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX remote drawing (Was Re: GX OOP class Date: 20 Jun 1997 00:13:00 -0700 Organization: Primenet Services for the Internet Message-ID: <AFCF8013-212B2@206.165.44.38> References: <oneN3XW00iVD069ts0@andrew.cmu.edu> nntp://news.primenet.com/comp.sys.next.advocacy, nntp://news.primenet.com/comp.sys.next.programmer, nntp://news.primenet.com/comp.sys.mac.programmer.misc MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Charles William Swiger <cs4w+@andrew.cmu.edu> said: > > And you're mistaken if you believe that a retained mode graphic engine > does a better job of reducing communications latency when remotely > displaying graphics than an immediate mode graphic engine in the event > that the client wishes to examine and/or manipulate the state of the > various graphic objects present in the system. If the client app has to examine and/or manipulate the state of various graphics objects present in the system often enough to make this an issue, than the retained graphics model isn't a very good fit for that purpose, even if running on the same system. How often do you need to examine the color of a rectangle? ------------------------------------------------------------------------- Build a better mousetrap, yes, but then market it like crazy, or instead of catching mice you'll only collect dust. -David Yeargin -------------------------------------------------------------------------
From: chris@polaris.scicntr.ortn.edu Newsgroups: comp.sys.next.programmer,comp.sys.next.bugs Subject: Re: IB: Missing Main Menu Date: 20 Jun 1997 08:31:24 GMT Organization: University of Tennessee Message-ID: <chris-ya023580002006970432230001@news.utk.edu> References: <chris-ya023580001806972300320001@news.utk.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit In article <chris-ya023580001806972300320001@news.utk.edu>, chris@polaris.scicntr.ortn.edu wrote: > I've been working on an app; Its a rather large app and of course has a > Main Menu. But the strangest thing is going on: > > In the app's main nib, there is the icon for the mainMenu. The weird thing > is that I can't seem to find this actual menu anywhere, i.e I double-click > on the icon, the Inspector switches to Menu Inspector, the name of the > app/menu title appears in the 'Title' text field, but the 'physical' list > of menu cells appears nowhere. > > I compile/run the app and it shows up as the app's main menu, just as you'd > expect; I switch back to IB and it's still gone. > > (It's been missing for days, and frankly I'm worried.) > > Has anyone seen this before ? > > > CB Sorry... New e-mail address : beauvois@usa.net Regards -- CB
From: dirk@object-factory.com (Dirk Olmes) Newsgroups: comp.sys.next.programmer Subject: no genstrings on Solaris Date: 20 Jun 1997 07:53:05 GMT Organization: Object Factory GmbH (Germany) Message-ID: <5odct1$q45$1@leonie.object-factory.com> Hi, I just tried to port a small WebObjects test to Solaris. Unfortunately it uses the MOKit. So I built a version of MOKit on Solaris (without the AppKit Stuff, of course) and tried to install the Framework. Unfortunately, MOKit wants to run a utility named "genstrings" when installing the framework. On Intel it is located in /usr/bin but I can't find it on Solaris. Am I missing something? -dirk --- ______________________________________________________________________ Dirk Olmes OBJECT FACTORY Gesellschaft fuer Informatik und Datenverarbeitung mbH Otto-Hahn Str. 18, 44227 Dortmund, Germany Telephon +49 (0) 231 975 137 0 Telefax +49 (0) 231 975 137 99 dirk@object-factory.com http://www.object-factory.com/ Hiroshima 45, Tschernobyl 86, Windows 95
From: beauvois@usa.net Newsgroups: comp.sys.next.programmer Subject: IB Did/Will Date: 20 Jun 1997 08:52:34 GMT Organization: University of Tennessee Message-ID: <beauvois-ya023580002006970453330001@news.utk.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sorry... I posted this earlier but left my old e-mail address. I'm trying to get an NS 2.1 palette to work on 3.3, and I run into an 'implicit declaration of function ...' error for two IB functions. Consequently the palette gets built but doesn't work. The code and errors are as follows: (The class being compiled is a subclass of IBInspector. There are only these two methods in it.) + finishLoading:(struct mach_header *)header { NIBDidLoadClass(self, header); return nil; } + startUnloading { NIBWillUnloadClass(self); return nil; } (The errors: ) warning: implicit declaration of function `NIBDidLoadClass' warning: implicit declaration of function `NIBWillUnloadClass' I am not familiar with what is going on here, and haven't found any documentation on these functions. Anyone ? Thanks -- CB
From: far@ix.netcom.com(Felipe A. Rodriguez) Newsgroups: comp.sys.next.programmer Subject: Re: posix libraries Date: 20 Jun 1997 08:55:33 GMT Organization: Netcom Message-ID: <5odgi5$8qd@sjx-ixn10.ix.netcom.com> References: <5oc9g2$ca0@news.blkbox.com> In article <5oc9g2$ca0@news.blkbox.com> bozack@blkbox.com writes: >I noticed OPENSTEP 4.1 comes with no uname binary, so I thought I'd write >one. > >So I looked at the man page for the uname() system call and slapped out some >code that would theoretically work .. until I found that the uname() isn't >present in the libraries that are linked automatically. > >I also tried compiling tcl-7.6p2 (so I could compile eggdrop). the tcl >library is archived fine, but tclsh fails during linkage since it can't find >mkfifo() .. fine, I say. I didn't need tcslsh ANYWAY. All I needed was the >library. So I go to compile eggdrop, but it fails during linkage because >references to the mkifo() call are present in the library, but the call >itself is nowhere to be found. I could theoretically remove all references >to mkfifo() in tcl (it's only referenced twice), but you never know when a >tcl app will pop up that needs to use fifos! > >Both of these (uname and mkfifo) are posix only system calls. OPENSTEP 4.1 >doesn't come with a posix library, and simply feeding the compiler a -posix >flag isn't going to do much good. > >So where do I go from here to find these magic system calls? > >Thanks in advance! >Dan For mkfifo you can just add the following to your code: int mkfifo (const char *path, mode_t mode) { return mknod (path, mode | S_IFIFO, 0); } You should be able to do the same for uname by extracting the necessary info using BSD system calls. -- Felipe A. Rodriguez # Francesco Sforza became Duke of Milan from Agoura Hills, CA # being a private citizen because he was # armed; his successors, since they avoided far@ix.netcom.com # the inconveniences of arms, became private (NeXTmail preferred) # citizens after having been dukes. (MIMEmail welcome) # --Nicolo Machiavelli
From: flight@mathi.uni-heidelberg.de (Gregor Hoffleit) Newsgroups: comp.sys.next.programmer Subject: Re: GNU binutils-2.8.1 Date: 20 Jun 1997 13:02:23 GMT Organization: InterNetNews at News.BelWue.DE (Stuttgart, Germany) Message-ID: <5odv0v$905$1@news.belwue.de> References: <kevincEBz5Mq.Hn1@netcom.com> kevinc (kevinc@netcom.com) wrote: : Has anyone compiled 2.8.1 for NextStep 3.3? configure complains : thjat Intel 3.3 is not supported by BFD. : Any assistance will be most appreciated. You'll have to implement support for the mach-o format for the BFD package. It's certainly non-trivial. Gregor -- | Gregor Hoffleit Mathematisches Institut, Uni HD | | flight@mathi.uni-heidelberg.de INF 288, 69120 Heidelberg, Germany | | (NeXTmail, MIME) (49)6221 54-5771 fax 54-8312 | | "We will make windows invisible" |
From: alanf@izzy.net Newsgroups: comp.sys.next.programmer Subject: Re: [Q]Using devices Date: 20 Jun 1997 14:34:33 GMT Organization: "Comshare, Inc." Message-ID: <5oe4dp$dr6$1@inet-prime.comshare.com> References: <EBMHKE.B00@x-lan.alienor.fr> Cc: fgalot@x-lan.alienor.fr In <EBMHKE.B00@x-lan.alienor.fr> fgalot@x-lan.alienor.fr wrote: > -Which device (/dev/????) should I use to communicate whith my serial port? > -How configuring a device? > -How configuring a port for lauching an executable automatically. > (example : a user connect himself on a port via a modem et man launch a BBS program). > > Thanks for help. > You may want to look at the MiscSerialPort object in MiscKit. Regards, Alan Frabutt (alanf@izzy.net)
From: bruno@gruick.univ-lr.fr (Bruno Garnier) Newsgroups: comp.sys.next.programmer Subject: Problem EOF procedure Date: 20 Jun 1997 14:54:54 GMT Organization: Universite de La Rochelle Message-ID: <5oe5ju$asd@hpuniv.univ-lr.fr> 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*" after verification i can only retrieve stored procedures with 20 parameters or less Can you give me some help please thank you and sorry for my poor english -- Bruno Garnier Centre de Ressources Informatique Universite de La Rochelle Avenue Marillac 17042 LA ROCHELLE CEDEX 01 Tel : 46 45 82 14 Fax : 46 45 82 45 http://www.univ-lr.fr/ bruno@cri.univ-lr.fr (NeXTMail, MIME Mail) |~~~| |~~~|D Guinness C|. :|D is |: .|D good for you ! |___| \\|// || |^ _| || ( -_+ ) // /_\ O / / __ __ __ || \_/ / | | | | | | ||____/ | | | | | | ---`----- \-- ---| |-| |--| |-- v |==| |==| |==|
From: alanf@izzy.net Newsgroups: comp.sys.next.programmer Subject: Re: Impressed Date: 20 Jun 1997 15:12:38 GMT Organization: "Comshare, Inc." Message-ID: <5oe6l6$ecc$1@inet-prime.comshare.com> References: <5nhfrv$c22@argentina.earthlink.net> <5njqj7$jqi@shelob.afs.com> Cc: Greg_Anderson@afs.com In <5njqj7$jqi@shelob.afs.com> Gregory H. Anderson wrote: > "Robert Norman" <rob@neurodata.com> writes > > > > Now I've produced a demonstration package for my client that builds a > > signal (sin wave, square wave) dyamically in one view. You move sliders > > to change the amplitude, frequency, noise content, and phase of the > > waveform and the wave updates as you drag the sliders. This was to test > > the graphics and the results were impressive. The waveforms update very > > fast with no disturbing screen update garbage. I'm just using PSxxxx > > calls to do the graphics. > > Yes, but can you draw 3D perspective text along the waves in six different > transfer modes and three different languages with different orientations? > Geez... what's really scary is we all know exactly which mock-persona would have wrote that! Regards, Alan Frabutt (alanf@izzy.net)
From: mpaque@yummyspam.wco.com (Mike Paquette) Newsgroups: comp.sys.next.programmer Subject: Re: Identifying Directories with Objective C Date: 20 Jun 1997 09:04:47 -0700 Organization: Electronics Service Unit No. 16 Sender: mpaque@mpaque Distribution: world Message-ID: <5oe9mv$jr@mpaque.mpaque> References: <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> In article <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> martin@wise-02.wiwi.tu-dresden.de (Martin Rose) writes: > Hi NeXTSTEP-Experts, > > Does anybody know how to find out whether a directory-entry is a subdirectory > or a "real" file?? There would be a solution to check if a dot '.' appears in > the filename, but this criteria would not be enough, because there might be > filenames without any extension... The proper, portable way to do this under OPENSTEP is to use NSFileManager to ask if a particular file object is a directory. (Note that hard-coded paths are bad. They are used here only for brevity. Use the NSOpen panel or other UI for user settable paths, the NSDefaults system for preferences, and NSBundle for code related paths.) The fileExistsAtPath:isDirectory: method will tell you if a file exists, and if it is a directory This example gets an NSArray that identifies the fonts in /NextLibrary/Fonts: NSArray *subpaths; BOOL isDir=NO; NSString *fontPath = @"/NextLibrary/Fonts"; NSFileManager *manager = [NSFileManager defaultManager]; if ([manager fileExistsAtPath:fontPath isDirectory:&isDir] && isDir) subpaths = [manager subpathsAtPath:fontPath]; You can easily enumerate the subpaths or entries in a directory using the NSDirectoryEnumerator class. The following example enumerates the contents of a directory and processes files; if, however, it comes across RTFD file packages, it skips recursion into them: NSDirectoryEnumerator *direnum = [[NSFileManager defaultManager] enumeratorWithPath:@ /Sales/Reports ]; NSString *pname; while (pname = [direnum nextObject]) { if ([[pname pathExtension] isEqualToString:@ rtfd ]) { [direnum skipDescendents]; /* don't enumerate this directory */ } else { /* ...process file here... */ } -- Mike Paquette (mpaque AT wco.com ; Yank that yummyspam 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: kcd@babylon5.jumpgate.com (Kenneth C. Dyke) Newsgroups: comp.sys.next.programmer Subject: Re: NSClassNames -> ??ClassNames Date: 20 Jun 1997 16:41:24 GMT Message-ID: <5oebrk$g51$1@nntp2.ba.best.com> References: <5ocfe4$h5q@news.bu.edu> In-Reply-To: <5ocfe4$h5q@news.bu.edu> On 06/19/97, John Siracusa wrote: >I was told by a former NeXT (now Apple) employee at >PC Expo this week that all the class names in >OpenStep like NSButton, etc. are going to be renamed >now that "NS" no longer has any bearing on the OS >(Rhapsody). My question is, what nomenclature will >they use? "CClassName" like Microsoft? Class names >based on libararies like the EOF classes seem to >have? Hmm... this seems like a really dumb idea if true. I can't see any real practical value in changing the class prefix on already existing OpenStep classes. I can certainly see the value in using different prefixes for NEW classes, like QTxxxx for QuickTime classes, etc. Changing the AppKit and Foundation class names just for the sake of changing the class names seems like it would be far more trouble than it would be worth, would do nothing (IMHO) to help new programmers, and would only be a pain in the ass for people that already have OpenStep code written or in progress. -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: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: yhjkm@asd.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5oeelp$434@mtinsc02.worldnet.att.net> Control: cancel <5oeelp$434@mtinsc02.worldnet.att.net> Date: 20 Jun 1997 17:29:30 GMT Message-ID: <cancel.5oeelp$434@mtinsc02.worldnet.att.net> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: jabi@acsu.buffalo.edu Newsgroups: comp.sys.next.programmer Subject: Re: Does OS4.2 finally fix the printf %g bug? Date: Fri, 20 Jun 1997 13:45:44 -0400 Organization: University At Buffalo Message-ID: <33AAC1C8.1A6D@arch.buffalo.edu> References: <5npof5$jlf@tribune.usask.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-User: jabi eric@skatter.USask.Ca wrote: > > Would someone out there running OPENSTEP 4.2 try my famous printf test program? Using gcc from OPENSTEP 4.2 Enterprise on Windows NT 4.0: $ a.out Expect 0.00123: 0.00123 Expect 123: 123 Expect 123.5: 123.5 Expect 1e+03: 1e+003 $ -- Wassim Jabi, Assistant Professor Department of Architecture wjabi@arch.buffalo.edu State University of New York at Buffalo Tel: +1 716.829.3485 Ext. 323 3435 Main St. - Hayes Hall Fax: +1 716.829.3256 Buffalo, NY 14214-3087
From: fpottier@pauillac.inria.fr (Francois Pottier) Newsgroups: comp.sys.next.programmer Subject: Re: Identifying Directories with Objective C Date: 20 Jun 1997 17:53:32 GMT Organization: INRIA Rocquencourt, BP 105, 78153 Le Chesnay Cedex, France Message-ID: <5oeg2s$ka@news-rocq.inria.fr> References: <5ob9ht$cs3$1@rks1.urz.tu-dresden.de> <5obrh6$pvv$1@sun579.rz.ruhr-uni-bochum.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Cc: In article <5obrh6$pvv$1@sun579.rz.ruhr-uni-bochum.de>, Daniel Boehringer <boehring@biomed.ruhr-uni-bochum.de> wrote: >if you have the misckit installed, though :-) What is it and where can we find it? Is it portable across OPENSTEP platforms? (Right now I thought the only portable kits were basically FoundationKit and AppKit). Thanks! -- François Pottier Francois.Pottier@inria.fr http://pauillac.inria.fr/~fpottier/
From: scott@doubleu.com (Scott Hess) Newsgroups: comp.sys.next.programmer Subject: Re: Does OS4.2 finally fix the printf %g bug? Date: 19 Jun 1997 14:14:29 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jun19093800@slave.doubleu.com> References: <5npof5$jlf@tribune.usask.ca> <5o94tp$atk@news.orst.edu> In-reply-to: Tim Pugh's message of 18 Jun 1997 17:12:25 GMT In article <5o94tp$atk@news.orst.edu> Tim Pugh <tpugh@nospams.OCE.ORST.EDU> writes: In <5npof5$jlf@tribune.usask.ca> eric@skatter.USask.Ca wrote: > Would someone out there running OPENSTEP 4.2 try my famous printf > test program? > > For those of you coming in late, this program tests for a bug > that has been present since NeXTstep version 1.0 and which I > first reported in December, 1989. <...tries it...> The answer is NO. I guess we'll have to wait until Rhapsody and BSD4.4 . What if they are actively supporting the broken behaviour? I can't imagine it would be more than a couple line change, no need to go to BSD4.4. Perhaps some large NeXT-loyal corporation has been using the broken behaviour for something, and every time NeXT fixes it, all their programs break. Hopefully that corporation isn't NeXT (in the depths of FoundationKit or somesuch). 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.next.programmer Subject: Re: PB.project file format Date: 19 Jun 1997 14:16:09 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jun19095252@slave.doubleu.com> References: <12342024@NEWS.SAIC.COM> In-reply-to: robertg@trg.saic.com's message of Wed, 18 Jun 1997 01:53:37 GMT In article <12342024@NEWS.SAIC.COM> robertg@trg.saic.com (Robert Gottlieb) writes: I'm trying to parse a PB.project file (NS 3.3) and I want to put the results in a dictionary. Ultimately I just need the PROJECTTYPE and PROJECTNAME attributes. I know I could get both of those via grep or some other shell utility, but I really want to try to make this work in ObjC. I've also heard that NS 4.? changes it to be a true property list, but I'm not upgrading just yet. Anyway, if anyone knows of a way of reading the PB.project file into an NSDictionary or some similiar structure, I'd love to hear your ideas. Something like this should work: void main( void) { NSString *file=[[NSString alloc] initWithContentsOfFile:@"PB.project"]; id dict=[file propertyListFromStringsFileFormat]; printf( "dict==%s/%p\n", [[[dict class] description] cString], dict); printf( "contents==%s\n", [[dict description] cString]); } Change propertyListFromStringsFileFormat to propertyList to read an NS4.0-style PB.project, -- 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: fpottier@pauillac.inria.fr (Francois Pottier) Newsgroups: comp.sys.next.programmer Subject: OpenStep: threads vs garbage collection Date: 20 Jun 1997 18:03:06 GMT Organization: INRIA Rocquencourt, BP 105, 78153 Le Chesnay Cedex, France Distribution: world Message-ID: <5oegkq$sf@news-rocq.inria.fr> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi, I'm discovering OpenStep and have a question which I hope might be answered here... I have read about OpenStep's convention for retaining and releasing objects, using reference counts. When an object's reference count drops to zero, it becomes eligible for collecting. The question is, when will it be collected? OpenStep's documentation says "at the end of the current event loop". This explanation sounds a little too simplistic and I'd like to have a better understanding. The way I figure, the object is collected when its NSAutoreleasePool dies. OpenStep creates such a pool at the beginning of the event cycle and kills it at the end, hence the above sentence. Is this correct? Now, on to the next question: what happens in a multi-threaded application? Suppose I have a function which creates an object, autoreleases it and returns it to its caller, and the function is executed in a secondary thread. When the function exits, the object's reference count is 0. Obviously, we don't want it to be collected right now, because the caller is going to retain it. So, the object must not be collected "at the end of the current event cycle", because that might happen at any moment. Rather, we need the object to live as long as the thread lives, because we can't determine when it will no longer be necessary, can we? So, my conclusion is, each thread needs to have its own NSReleasePool which has the same life duration as the thread; and objects allocated in a thread shall be released only when the thread dies. Is all this correct? Or did I miss something altogether? To sum up, the basic question, can I follow the retain/autorelease rules blindly even in a multi-threaded application? By the way, while I'm here, I have another question about threads. OpenStep's documentation says that the AppKit is not reentrant. So, in a multi-threaded application, I guess I need to either: - allow calling the AppKit only from the main thread or: - create a lock and acquire it every time I call the AppKit. (The first method sounds simpler). What do you guys do about this? Thanks a lot for reading up to this point! -- François Pottier Francois.Pottier@inria.fr http://pauillac.inria.fr/~fpottier/
From: Timothy Luoma <luomat@peak.org> Newsgroups: comp.sys.next.programmer Subject: nuke.c Date: Fri, 20 Jun 1997 13:06:37 -0700 Organization: The PEAK FTP site for OpenStep & NeXTStep Message-ID: <Pine.SUN.3.96.970620125907.10581X-100000@kira> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Remember the 'nuke' program that was supposed to kill unkillable processes? This is the code: #include <stdio.h> #include <mach.h> main(int argc, char *argv[]){ if(argc != 2){ fprintf(stderr, "Usage: %s pid\n", argv[0]); exit(1); } else{ task_t task; task_by_unix_pid(task_self(), atoi(argv[1]), &task); task_terminate(task); exit(0); } } It caused my 3.3 m68k machine to panic after working fine with 3.2.... Just wanted to mention that it works fine with OS4.1/mach/Intel TjL
From: sams@best.com (Samuel G. Streeper) Newsgroups: comp.sys.next.programmer Subject: Re: OpenStep: threads vs garbage collection Date: 20 Jun 1997 16:58:10 -0700 Organization: BEST Internet Communications Message-ID: <sams.866850965@shellx> References: <5oegkq$sf@news-rocq.inria.fr> These were fine questions, mon ami! Before I begin let me say that the ref-counting system invites problems with under-freeing (leaking) and over-freeing (invalid pointers and smashers) due to programming bugs, incomplete documentation or design knowlege, and even design necessities, and when these problems occur they will be very tricky to track. This info is just so you know what you're in for. I understand the rationales behind the design of the freeing system, but the design is inherently flawed and unwieldy. I believe that there is by default 1 autorelease pool per thread. I don't think this is documented but I have faith that this works because I've seen no evidence to the contrary and I think the merde would quickly and very apparently show evidence of its encounter with the proverbial fan if its function were otherwise. > To sum up, the basic question, can I follow > the retain/autorelease rules blindly > even in a multi-threaded application? Similarly summarily, I suspect you shall not see memory anomalies from mixing threads and release pool stuff, though I would expect you to see them for other reasons. I could be wrong... -sam
Date: Sat, 21 Jun 1997 11:41:35 -0600 From: e1@biogate.com Subject: Free sex site password Newsgroups: comp.sys.next.programmer,comp.sys.next.software,comp.sys.northstar,comp.sys.nsc.32k,comp.sys.oric,comp.sys.palmtops Message-ID: <866910340.16097@dejanews.com> Organization: Deja News Usenet Posting Service Hi, I paid $24 dollars for this password, and then for some strange reason my check bounced, so I ended up paying almost $50 bucks. For that much money I think I should be able to share it with people in my newsgroups. username is - tom25 passcode is - dodge the site is at; http://204.191.136.6/~kiervinl/s.html -------------------==== 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: e1@biogate.com Newsgroups: comp.sys.next.programmer,comp.sys.next.software,comp.sys.northstar,comp.sys.nsc.32k,comp.sys.oric,comp.sys.palmtops Subject: cmsg cancel <866910340.16097@dejanews.com> Control: cancel <866910340.16097@dejanews.com> Date: 21 Jun 1997 16:42:32 GMT Message-ID: <cancel.866910340.16097@dejanews.com> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: mmalcolm crawford <Malcolm_Crawford@plsys.co.uk> Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.next.programmer,comp.sys.mac.programmer.misc Subject: Re: GX remote drawing (Was Re: GX OOP class Date: 21 Jun 1997 13:45:18 GMT Organization: P&L Systems Message-ID: <5oglte$9f5$1@ironhorse.plsys.co.uk> References: <oneN3XW00iVD069ts0@andrew.cmu.edu> <AFCF8013-212B2@206.165.44.38> In-Reply-To: <AFCF8013-212B2@206.165.44.38> On 06/20/97, "Lawson English" wrote: >How often do you need to examine the color of a rectangle? > Probably about as often as you have to translate every third blue word in every other paragraph into a 3D Japananese character. Best wishes, mmalc. --
Newsgroups: comp.sys.next.programmer,control From: news@news.msfc.nasa.gov Message-ID: <cancel.5ohevu$1h2$39@talia.mad.ibernet.es> Control: cancel <5ohevu$1h2$39@talia.mad.ibernet.es> Subject: cmsg cancel <5ohevu$1h2$39@talia.mad.ibernet.es> no reply ignore Organization: Semi-Automatic Lupine Remover Date: Sat, 21 Jun 1997 20:57:14 GMT Sender: HDDL9@VSEZD.V7G0X ignore Make Money Fast post canceled by J. Porter Clark.
From: MWRon@metrowerks.com (MW Ron) Newsgroups: comp.sys.next.programmer,comp.os.ms-windows.programmer.tools.mfc,comp.sys.next.misc Subject: [JOBS] Metrowerks Research and Development Date: Sat, 21 Jun 1997 19:29:20 -0400 Organization: Metrowerks Message-ID: <MWRon-2106971929200001@aumi2-a02.ccm.tds.net> Metrowerks Current Job Openings Metrowerks is the creator of CodeWarrior, a suite of software development tools used by programmers around the world. This fast-growing, Austin-based company is currently seeking candidates to fill a number of positions: RESEARCH AND DEVELOPMENT Four-year college degrees required; computer science, computer engineering and electrical engineering degrees preferred. A. Libraries/Performance Analysis Engineer, C/C++/Pascal/Java Some experience with FORTRAN, C, C++ or Java required. Duties will include various subsets of the following: * measuring compiler performance across various x86, PowerPC, 68K, MIPS implementations * analysis of object code for codgen flaws, possible areas of improvement * analysis via automated suites to determine optimal library designs, evaluating new sources * testing C/C++ libraries for conformance to ISO and emerging standards, fixing singularities * writing example code, writing technical documentation, providing high-end support for STL users B. Rapid Application Development (RAD) Tools (Both junior and senior level positions available) These positions will include the design and implementation of core technology and user interface for tools to assist in the development of object-oriented applications. Candidates should have strong Windows or Macintosh experience (experience with both Windows and Macintosh preferred) and a proven track record delivering major applications in C++. Candidates should also have played a key role in contributing to projects involving extensive user interface and core code. Preference will be given to candidates with experience with the following: previous development tool implementation experience, Java, Visual C++, MFC, ActiveX or NextStep. C. Software Engineer for CodeWarrior Latitude This person will develop portable implementations of Macintosh Toolbox functionality for CodeWarrior Latitude. Will write code in C, C++, & ObjC on non-Mac platforms including Solaris, IRIX, HP-UX and Rhapsody. UNIX experience, however, is not essential. Will develop testing suites for quality assurance. Candidates are required to have at least three years of solid Mac programming experience. Also, candidates with strengths in interprocess communications such as AppleTalk and AppleEvents, AppleScript and OpenTransport are desired. Experience with WorldScript & other Internationalization features, such as multi-byte character handling, is a bonus. D. UNIX Programmer Metrowerks is looking for a junior to intermediate level programmer for its UNIX group. Candidates must know C and C++ and have at least two years programming experience. Previous experience programming on UNIX is a must; experience in UNIX programming with gcc preferred. E. Quality Assurance Engineers Metrowerks is looking for people with strong cross platform skills and some programming and testing experience. Quality assurance engineers will work in a rapidly changing environment and will enjoy the benefits of working with state-of-the-art (bleeding edge) technologies such as Java, BeOS, PalmOS (Pilot), MacOS 8 and others. Macintosh and/or Microsoft Windows programming experience is required. Must be familiar with one or more of the following programming languages: C, C++, Java, Pascal and one or more of the following platforms: Macintosh, Microsoft Windows 95/NT or UNIX. Medical, dental, life, disability and 401K plans are available to employees. For more information, see the Metrowerks web page <http://www.metrowerks.com>. Please send resumes by email to hr@metrowerks.com, or by fax to Recruiting Coordinator at 512/873-4900. -- METROWERKS Ron Liechty http://www.metrowerks.com MWRon@metrowerks.com
From: ;odirujewo@oafosd.com Newsgroups: comp.sys.next.programmer Subject: .Teen Babes take off there school clothes teenclo.jpg Date: 22 Jun 1997 00:18:59 GMT Organization: All USENET -- http://www.Supernews.com Message-ID: <5ohr1j$9gu@usenet84.supernews.com> Looking for Dirty Schoolgirls Having Sex? Well, then look no further, Nasty-Schoolgirls is your source for the Nastyiest girls that will Fuck and Suck your Cock. Are you ready to have a Schoolgirl Suck you Cock? Then Visit: http://www.nasty-schoolgirls.com All Models Represented on Nasty-Schoolgirls are 18 years of age or older.
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: ;odirujewo@oafosd.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5ohr1j$9gu@usenet84.supernews.com> Control: cancel <5ohr1j$9gu@usenet84.supernews.com> Date: 22 Jun 1997 00:02:00 GMT Message-ID: <cancel.5ohr1j$9gu@usenet84.supernews.com> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: alsfdjs@;lakdjfalsk.com Newsgroups: comp.sys.next.programmer Subject: .Teen Pussy Closeup teenpus.jpg Date: 22 Jun 1997 02:06:01 GMT Organization: All USENET -- http://www.Supernews.com Message-ID: <5oi1a9$d7c@usenet84.supernews.com> Looking for Dirty Schoolgirls Having Sex? Well, then look no further, Nasty-Schoolgirls is your source for the Nastyiest girls that will Fuck and Suck your Cock. Are you ready to have a Schoolgirl Suck you Cock? Then Visit: http://www.nasty-schoolgirls.com All Models Represented on Nasty-Schoolgirls are 18 years of age or older.
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: alsfdjs@;lakdjfalsk.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5oi1a9$d7c@usenet84.supernews.com> Control: cancel <5oi1a9$d7c@usenet84.supernews.com> Date: 22 Jun 1997 01:39:37 GMT Message-ID: <cancel.5oi1a9$d7c@usenet84.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: 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: <3108866347221@digifix.com> Date: 22 Jun 1997 03:57:06 GMT Organization: Digital Fix Development Message-ID: <10436866952022@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.informatik.uni-muenchen.de: (Peanuts) Located in Germany. Comprehensive archive site. Very well maintained. ftp://terra.stack.urc.tue.nl (Dutch NEXTSTEP User Group) 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: jbf_see_sig@frazer.com (James B. Frazer) Newsgroups: comp.sys.next.programmer Subject: Re: posix libraries Date: Sun, 22 Jun 1997 06:36:36 -0400 Organization: The Internet Access Company, Inc. Message-ID: <jbf_see_sig-ya023580002206970636360001@news.tiac.net> References: <5oc9g2$ca0@news.blkbox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit In article <5oc9g2$ca0@news.blkbox.com>, bozack@blkbox.com wrote: > Both of these (uname and mkfifo) are posix only system calls. OPENSTEP 4.1 > doesn't come with a posix library, and simply feeding the compiler a -posix > flag isn't going to do much good. > > So where do I go from here to find these magic system calls? Back to 3.3? Barney (delete that _see_sig to email me) a Zip) only to have the Mac OS refuse to open them. Thanks in advance for any advice!! -------------------------------------------------------------- chuck@cmich.slip.netcom.com NeXT Mail and MIME Mail gladly accepted --------------------------------------------------------------
From: gtupar@ctp.com (Georg Tuparev) Newsgroups: comp.sys.next.programmer Subject: OS/NT with Citrix/WinFrame Date: 23 Jun 1997 12:59:47 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5ols03$agj$1@concorde.ctp.com> Hey Folks! The client asked us if our system will work on top of NT product, called WinFrame (from Citrix). It seams to provide a sort of functionality similar to NXHosing (I was told so...). Any ideas, rumors, experience? Thanks G.T. -- ------- /\/\ 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: "Richard Oesterreicher" <ROester@rodesign.com> Newsgroups: comp.sys.next.programmer,comp.sys.amiga.programmer,comp.sys.be.programmer Subject: Re: Free online pubication - Attainment adding Software Developers' column Date: 22 Jun 1997 18:32:15 GMT Organization: ro design Message-ID: <01bc7f3a$575a7cf0$020a0a64@batman> References: <Pine.BSF.3.95q.970615122406.4073B-100000@web2.calweb.com> attainment <abiogen@abiogenesis.com> wrote in article <Pine.BSF.3.95q.970615122406.4073B-100000@web2.calweb.com>... > > Subject: Free online pubication - Attainment adding Software Developers' column > > > I thought programmers who are software entrepreneurs would appreciate > knowing that we've added a column to our free online publication that will > deal with software issues, and how to build a business selling software > online and off. > > It's located at > > http://www.abiogenesis.com/attainment > Why does almost every page say Issue #1 - February '97? I can't help but notice that is around the same time (End of January) the Northwest Firebug Be User's group site (also at www.abiogenesis.com) dried up and stopped updating its site with the tag, "Under construction, Great things to come!" How often do you plan to update your business resources site with new articles? Thanks, Richard Oesterreicher ROester@rodesign.com "If you don't make time to do it right, when will you have time to do it over." +++++++++++++++++++++++++++++++++++++++++ + Your life is HARD, your software should be EASY! + Now it can be… http://www.rodesign.com + ro design, Inc. (425) 844-2921 +++++++++++++++++++++++++++++++++++++++++
From: "Lee Byeong-ho" <bhlee@cnt.co.kr> Newsgroups: comp.sys.next.programmer Subject: [Q]How can I save in database? Date: Mon, 23 Jun 1997 10:06:07 -0000 Organization: Hansol Telecom Message-ID: <5olhc6$pld$1@news.hansol.net> References: <1997Jun22.033358.340@cmich.slip.netcom.com> Hi All, I'm using WOF3.1, EOF2.1, and Oracle 7.3 and have problem creating records using Oracle. I'm going to make some BBS and it's necessary for me to save some contents in DB. How can I do this? Someone talked me to use 'saveObject:'method. I'm novice. Please tell me about those in Detail... Thanks in advance for any comments!!! ---------------------- Lee, Byeong-ho Yuhan C&T, Seoul, Korea mailto:bhlee@cnt.co.kr ---- God is Love-------
From: "Lee Byeong-ho" <bhlee@cnt.co.kr> Newsgroups: comp.sys.next.programmer Subject: [Q]How can I save in database? Date: Mon, 23 Jun 1997 10:06:07 -0000 Organization: Hansol Telecom Message-ID: <5olhcj$2da$1@news.hansol.net> References: <1997Jun22.033358.340@cmich.slip.netcom.com> Hi All, I'm using WOF3.1, EOF2.1, and Oracle 7.3 and have problem creating records using Oracle. I'm going to make some BBS and it's necessary for me to save some contents in DB. How can I do this? Someone talked me to use 'saveObject:'method. I'm novice. Please tell me about those in Detail... Thanks in advance for any comments!!! ---------------------- Lee, Byeong-ho Yuhan C&T, Seoul, Korea mailto:bhlee@cnt.co.kr ---- God is Love-------
From: dob@mrd3.mmm.ucar.edu (David Blanchard,,,) Newsgroups: comp.sys.next.programmer Subject: Fortran compiler Date: 22 Jun 1997 16:40:21 GMT Organization: NOAA/NSSL, Boulder, Colorado Message-ID: <5ojkhl$lll$1@ncar.ucar.edu> I've scanned the FAQs and looked through the newsgroups but cannot find the information I seek. I'm interested in a Fortran compiler for NeXTstep v3.3 running on a Sun Sparcstation. I've found information for compilers running on Motorola and Intel, but not Sparc. Are there any? Thanks. -db- -- +----------------------------------------------------------------------+ | David Blanchard NOAA/NSSL & OU/CIMMS Boulder, Colorado | | blanch@ucar.edu http://mrd3.mmm.ucar.edu/~dob/www/ | +----------------------------------------------------------------------+
Newsgroups: comp.os.os2.advocacy,comp.os.os2.apps,comp.os.os2.beta,comp.os.os2.bugs,comp.os.os2.games,comp.os.os2.misc,comp.os.os2.networking.misc,comp.os.os2.networking.tcp-ip,comp.os.os2.programmer.misc,comp.os.os2.setup.misc,comp.sys.acorn.programmer,comp.sys.amstrad.8bit,comp.sys.apple2,comp.sys.apple2.marketplace,comp.sys.cbm,comp.sys.dec,comp.sys.handhelds,comp.sys.hp.hardware,comp.sys.hp.hpux,comp.sys.hp.mpe,comp.sys.hp48,comp.sys.intel,comp.sys.laptops,comp.sys.next.advocacy,comp.sys.next.hardware,comp.sys.next.programmer,comp.sys.palmtops,comp.sys.pen,comp.sys.tandy,control From: news@news.msfc.nasa.gov Message-ID: <cancel.5om47t$cg6@newton.fgg.eur.nl> Control: cancel <5om47t$cg6@newton.fgg.eur.nl> Subject: cmsg cancel <5om47t$cg6@newton.fgg.eur.nl> no reply ignore Organization: Semi-Automatic Lupine Remover Date: Mon, 23 Jun 1997 15:44:18 GMT Sender: Ray1996@freemail.nl ignore Make Money Fast post canceled by J. Porter Clark.
From: Lars Immisch <lars@ibp.de> Newsgroups: comp.sys.next.programmer Subject: Re: Saving in Mac Format Question Date: Mon, 23 Jun 1997 15:17:15 +0200 Organization: Immisch, Becker & Partner Message-ID: <33AE775B.7A1@ibp.de> References: <1997Jun22.033358.340@cmich.slip.netcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit chuck@cmich.slip.netcom.com wrote: > > I'm trying to save .snd files into a Macintosh format (So I can open them > in Sound Designer II). I can do conversions (sample rate, etc), but even > though the data is in the right format, the Mac doesn't recognize these as > SDII files, so I can't open them on the Mac. > > I've heard there is some type of "Mac Header" either in the file or > perhaps in the directory entry itself. Is there any way to duplicate this > header info on the NeXT? It seems kind of useless to save files in Mac > format (on a Zip) only to have the Mac OS refuse to open them. > > Thanks in advance for any advice!! The "Mac Header" you talk about is called resource fork. You cannot create it on the NeXT because it's part of the Mac File system. Off the top of my head, there are three way to create the necessary Type and Creator information: you use ResEdit, the "Get Info" menu (or similar) and set Type and Creator to the values of a Sound Designer II document you peeked at before. Or you use AutoTyper, a very useful utility for exactly that. There you can can clone Type and Creator from an existing document. Or you copy the file to a DOS disk and tell PC exchange to map the ending .snd to the desired application and have it set when you copy the file from the DOS disk to the Mac. This is probably a frequently asked question in the Mac forums. Cheers, Lars -- mailto:immisch@pobox.com http://pobox.com/~immisch Yesterdays yellow yoyo can make you yawn today
From: eskimo1@apple.com (Quinn) Newsgroups: comp.sys.next.programmer Subject: Re: OpenStep: threads vs garbage collection Date: Mon, 23 Jun 1997 09:39:47 -0700 Organization: Apple Computer Inc Distribution: world Message-ID: <eskimo1-2306970939470001@guy-smiley.apple.com> References: <5oegkq$sf@news-rocq.inria.fr> In article <5oegkq$sf@news-rocq.inria.fr>, fpottier@pauillac.inria.fr (Francois Pottier) wrote: >Now, on to the next question: what happens in a multi-threaded >application? Suppose I have a function which creates an object, >autoreleases it and returns it to its caller, and the function is >executed in a secondary thread. When the function exits, the >object's reference count is 0. Obviously, we don't want it to be >collected right now, because the caller is going to retain it. So, >the object must not be collected "at the end of the current event >cycle", because that might happen at any moment. Rather, we need >the object to live as long as the thread lives, because we can't >determine when it will no longer be necessary, can we? So, my >conclusion is, each thread needs to have its own NSReleasePool >which has the same life duration as the thread; and objects >allocated in a thread shall be released only when the thread dies. > >Is all this correct? [...] Pretty much. Each thread *does* have its own autorelease pool. When that pool is released depends on the design of your thread. If your thread is modelled on some sort of periodic action, you could have your thread release the autorelease the pool at the end of that periodic action, much like the AppKit does. This is most probably a good idea. Alternatively, releasing the autorelease pool at the end of the thread is the default behaviour. For more information on this, take a look at the class documentation for NSConnection. S+E -- Quinn "The Eskimo!" <http://devworld.apple.com/dev/geeks.html> Apple Developer Technical Support * Rhapsody Networking & Low-Level Stuff ps Good to have you on board Francois (-:
Newsgroups: comp.sys.next.programmer From: hugob@tamtam.xs4all (Hugo Burm) Subject: Re: Want to program in C++ Message-ID: <EC8506.6y@tamtam.xs4all.nl> Sender: hugob@tamtam.xs4all.nl (Hugo Burm) Organization: datagram References: <Pine.SOL.3.91.970622130916.13691B-100000@expert.cc.purdue.edu> Date: Mon, 23 Jun 1997 10:26:29 GMT In article <Pine.SOL.3.91.970622130916.13691B-100000@expert.cc.purdue.edu> Adrian Silveanu <silveanu@expert.cc.purdue.edu> writes: > > Hello, > > I have NS v3.2 and the developer kit running on an Intel > computer. Would somebody please tell me how or where to find information > to setup the developer kit so that I can program in C++? > > Thanks in advance for any help! > > Adrian > silveanu@purdue.edu > Go to the docs and find the Concepts manual. There is a chapter about programming in C++. On my system the info is in: "/NextLibrary/Documentation/NextDev/Concepts/ObjectiveC/5_Programming/Prog ramming.rtfd". (If you did not install the manuals, you will find it on your NextStep Developer CD). Or have a look at the mixed Obj-C/C++ demo: "/NextDeveloper/Examples/AppKit/CalculatorLab++" hugob@tamtam.xs4all.nl
From: paul@oneclick.com (Paul Collins) Newsgroups: comp.sys.next.programmer Subject: Re: Saving in Mac Format Question Date: Mon, 23 Jun 1997 09:16:33 -0700 Organization: One Click Systems Message-ID: <paul-2306970916330001@oneclick.vip.best.com> References: <1997Jun22.033358.340@cmich.slip.netcom.com> In article <1997Jun22.033358.340@cmich.slip.netcom.com>, chuck@cmich.slip.netcom.com wrote: > I'm trying to save .snd files into a Macintosh format (So I can open them > in Sound Designer II). I can do conversions (sample rate, etc), but even > though the data is in the right format, the Mac doesn't recognize these as > SDII files, so I can't open them on the Mac. > > I've heard there is some type of "Mac Header" either in the file or > perhaps in the directory entry itself. Is there any way to duplicate this > header info on the NeXT? It seems kind of useless to save files in Mac > format (on a Zip) only to have the Mac OS refuse to open them. Chuck, it's not really a header. Mac files are stored in 2 or 3 pieces on the hard disk (invisible to the user). This format can be stored on foreign file systems (as AppleSingle or AppleDouble), but still has to be converted when put on the Mac (everything in a foreign file gets put into the Mac "data fork" when copying, unless you have a program translating the file). But I'm surprised that you're sound app can't open the file. Did you run the program and use the Open command? (as opposed to double-clicking the Finder, which doesn't know what to do with the file). Doesn't the documentation cover this? You could try a shareware sound program that can read files from other systems. Look here: http://hyperarchive.lcs.mit.edu/HyperArchive.html Good luck! -- Paul Collins, Owner/Developer One Click Systems -- -- paul@oneclick.com http://www.oneclick.com/ -- -- Now shipping! ClickMail 1.1 Internet gateway for QuickMail --
Newsgroups: comp.sys.next.programmer From: Fabien_Roy@no.spam.free.fdn.fr (Remove no.spam to reply) Subject: Re: posix libraries Message-ID: <EC8tq0.CGE@free.fdn.fr> Sender: news@free.fdn.fr Organization: Fabien Roy Consultant. References: <5oc9g2$ca0@news.blkbox.com> <5odgi5$8qd@sjx-ixn10.ix.netcom.com> Date: Mon, 23 Jun 1997 19:20:24 GMT far@ix.netcom.com(Felipe A. Rodriguez) wrote: > In article <5oc9g2$ca0@news.blkbox.com> bozack@blkbox.com writes: > >I noticed OPENSTEP 4.1 comes with no uname binary, so I thought I'd write > >one. > > > >So I looked at the man page for the uname() system call and slapped out some > >code that would theoretically work .. until I found that the uname() isn't > >present in the libraries that are linked automatically. > > > >I also tried compiling tcl-7.6p2 (so I could compile eggdrop). the tcl > >library is archived fine, but tclsh fails during linkage since it can't find > >mkfifo() .. fine, I say. I didn't need tcslsh ANYWAY. All I needed was the > >library. So I go to compile eggdrop, but it fails during linkage because > >references to the mkifo() call are present in the library, but the call > >itself is nowhere to be found. I could theoretically remove all references > >to mkfifo() in tcl (it's only referenced twice), but you never know when a > >tcl app will pop up that needs to use fifos! > > > >Both of these (uname and mkfifo) are posix only system calls. OPENSTEP 4.1 > >doesn't come with a posix library, and simply feeding the compiler a -posix > >flag isn't going to do much good. > > > >So where do I go from here to find these magic system calls? > > > >Thanks in advance! > >Dan > > > For mkfifo you can just add the following to your code: > > int mkfifo (const char *path, mode_t mode) > { > return mknod (path, mode | S_IFIFO, 0); > } > > You should be able to do the same for uname by extracting the > necessary info using BSD system calls. > > > > > -- > Felipe A. Rodriguez # Francesco Sforza became Duke of Milan from > Agoura Hills, CA # being a private citizen because he was > # armed; his successors, since they avoided > far@ix.netcom.com # the inconveniences of arms, became private > (NeXTmail preferred) # citizens after having been dukes. > (MIMEmail welcome) # --Nicolo Machiavelli FYI a shell equiv of uname fabien@p133> cat /etc/uname #!/bin/sh # # uname # # Author: Jim Vlcek, ByteWare Consulting (uunet!molly!vlcek) 3 Dec 1993 # Tweaked-By: Larry Blische (lkba!lkb@uunet.uu.net) # Warped-By: Tom Hageman (tom@basil.icce.rug.nl) 8 Dec 1993 # # An attempt to implement a SysV-ish "uname" under NeXTStep 3.x # # Options # # -s Print the operating system name # -n Print the node name (essentially, the hostname) # -v Print the operating system version # -r Print the operating system release # -p Print the host machine's processor type # -m Print the machine hardware name # -a Print all the above information # # Non-SysV options # # -i Print the host identification number (hostid) # case "$*" in "") set - -s ;; *-a*) set - -s -n -v -r -p -m ;; esac for arg do case $arg in -s) system="NEXTSTEP" ;; -n) node="`uuname -l`" ;; -r) release="`hostinfo | sed -n 's/.*NeXT Mach \([0-9\.]*\).*/\1/p'`" ;; -m) mach="`hostinfo | sed -n 's/.*Processor type: \([^ ]*\).*/\1/p'`" ;; -p) processor="`hostinfo | \ sed -n 's/.*Processor type: [^ ]* (\([^)]*\).*/\1/p'`" ;; -v) version="`tail -1 /usr/lib/NextStep/software_version`" ;; -i) hostid="`hostid`" ;; *) echo $0: Usage: $0 [-asnvrpmi] >&2 ; exit 1 ;; esac done echo $system $node $version $release $processor $mach $hostid fabien@p133> -- Fabien Roy --------------------------------------------------------------------- Fabien_Roy@free.fdn.org (NextMail/MIME accepted) Fabien Roy Consultant NEXTSTEP/OPENSTEP/EOF Consultant, SYBASE DBA 10 rue de la DEFENSE 93100 MONTREUIL, France Tel: 33 (0)1 45 28 32 23 Fax: 33 (0)1 48 55 09 90 GSM: 33 (0)6 60 46 36 83
Newsgroups: comp.sys.next.programmer From: valerie@mizar (Moukdarath Valerie) Subject: (PB) EOF NLS_LANG Content-Type: text/plain; charset=iso-8859-1 Message-ID: <EC89pM.8BA@x-lan.alienor.fr> Sender: news@x-lan.alienor.fr Content-Transfer-Encoding: 8bit Organization: x&lan Mime-Version: 1.0 Date: Mon, 23 Jun 1997 12:08:09 GMT I work on NeXTstep 3.2 and EOF 1.1 and Oracle 7.1 My oracle NLS_LANG is american_america.we8dec I don't know what NLS_LANG to use on EOF to read correctely my date ex : "Boite de crème à été" in my database is not the same when I read it with EOF. é may be replace by "Y" or "î" it depends on what caracter set i use. valérie
From: silveanu@expert.cc.purdue.edu (Adrian Silveanu) Newsgroups: comp.sys.next.programmer Subject: I want to program in C++ - Follow Up Date: 23 Jun 1997 19:44:49 GMT Organization: Purdue University Computing Center Message-ID: <5omjnh$jqv@mozo.cc.purdue.edu> Hello, I posted earlier that I wanted to program in C++. I now realize that I wasn't very clear. What I meant was that I tried to program in C++ on NS and it didn't work. I am running NS v3.2 w/ Dev Kit on an Intel computer. This is the simple code that I am trying to compile: #include <iostream.h> void main() { cout << "Coolness!" << endl; } Those of you who program in C++ will agree that this code should work. I know it works, because I have tried it on a SUN with a GNU C++ compiler and on OS/2 with a GNU C++ compiler. What I can't figure out is how to get it to compile on NS. Here are some details. I know that NS v3.2 can do C++. This info is in the documentation /NextLibrary/Documentation/GNU-libg++/Intro.rtf and /NextLibrary/Documentation/NextDev/ReleaseNotes/C++.rtf. I checked the compiler version: cc -v Reading specs from /lib/i386/specs NeXT Computer, Inc. version cc-216.obj~13, gcc version 2.2.2 And the version agrees with the documentation. The above C++ code I saved into a file called test.cc. In the documentation it says that the compiler should figure out, from the file extension, that it is a C++ file. It doesn't and this is what happens: cc test.cc test.cc:4: warning: return type for `main' changed to integer type ld: Undefined symbols: _endl__FR7ostream _cout ___ls__7ostreamPCc ___ls__7ostreamPFR7ostream_R7ostream So I have no idea what is wrong. I tried using the -lstdcpp option on the cc compiler like I do on GNU's OS/2 version of the compiler, but that option doesn't exist. By now you may be thinking why not just program C++ in OS/2. Well, I can, but NS is getting digital dust (if that is possible). I use OS/2 for WP, Inet surfing, games, and programming. When I was programming in just C I always used NS. I would like to continue using NS for programming in C++. My programming, at the moment, is for my Computer Science courses, so it isn't NS specific. If any of you can help me get the cc compiler to compile C++ code I would greatly appreciate it! Thank you, Adrian Silveanu silveanu@purdue.edu
From: ftouhi@IRO.UMontreal.CA (Majid Ftouhi) Newsgroups: comp.sys.next.programmer Subject: LEDA for NextStep/Intel? Date: 24 Jun 1997 00:15:26 GMT Organization: Universite de Montreal Distribution: world Message-ID: <5on3iu$s6i@epervier.CC.UMontreal.CA> Hi: I need to install LEDA for NextStep Intel. If someone had done the instalation, please tell me what i need for it. Thanks in advance
From: Žiboehme@abm08.abm.de (Ivo Boehme) Newsgroups: comp.sys.next.programmer Subject: Re: I want to program in C++ - Follow Up Date: 24 Jun 1997 07:50:11 GMT Organization: Nacamar Data Communications Message-ID: <5onu7j$esb$1@news.nacamar.de> References: <5omjnh$jqv@mozo.cc.purdue.edu> silveanu@expert.cc.purdue.edu (Adrian Silveanu) wrote: > > #include <iostream.h> > > void main() > { > cout << "Coolness!" << endl; > } > > [ ... ] > The above C++ code I saved into a file called test.cc. In the > documentation it says that the compiler should figure out, from the file > extension, that it is a C++ file. It doesn't and this is what happens: > cc test.cc > test.cc:4: warning: return type for `main' changed to integer type > ld: Undefined symbols: > _endl__FR7ostream > _cout > ___ls__7ostreamPCc > ___ls__7ostreamPFR7ostream_R7ostream > > So I have no idea what is wrong. I tried using the -lstdcpp option on the > cc compiler like I do on GNU's OS/2 version of the compiler, but that option > doesn't exist. The compiler complains that you didn't start the training with the `hello world' sample ... sorry -- just kidding :-)) No, no, seems you lack the c++ library (What do you think `ld: Undefined symbols' mean, BTW ;-)) Well then, try localhost> cc -Wall test.cc -lg++ -o test++ If the compiler complains about a missing g++ lib. Go and install it. After all, Install gcc >= 2.5.8 if you want to do more that three-liners, install the lastest-you-can-find-gcc, if you want to be serious. Regards, Ivo
From: huerlima@ivt.baum.ethz.ch (Daniel Huerlimann) Newsgroups: comp.sys.next.programmer Subject: NSTableView: Problem with Editing Date: 24 Jun 1997 12:23:00 GMT Organization: Swiss Federal Institute of Technology (ETHZ) Message-ID: <5ooe74$1id$1@elna.ethz.ch> I tried to end manually the editing of a tableView (NSTableView), but there was no success. After one cell has been edited, my tableView should be without a selection and without an edited cell. Has anyone an advice ? Thanks . Dani. ============================================= | Dani Huerlimann | ETH Zurich | Inst. of Transp., Traffic, Highway & Railway Engineering | CH - 8093 Zuerich, SWITZERLAND | Phone: + 41 1 633 27 38 | Fax: + 41 1 633 10 57 | E-Mail: huerlimann@ivt.baum.ethz.ch | WWW: http://www.ivt.baum.ethz.ch =============================================
Newsgroups: comp.sys.next.programmer From: gabriel@trigger.ali.bc.ca (Gabriel Musatescu) Subject: Re: Identifying Directories with Objective C Message-ID: <ECAGKA.8tz@gateway.ali.bc.ca> Sender: nobody@gateway.ali.bc.ca Organization: A.L.I. Technologies, Inc. Date: Tue, 24 Jun 1997 16:31:21 GMT References: <5oeg2s$ka@news-rocq.inria.fr> In article <5oeg2s$ka@news-rocq.inria.fr> fpottier@pauillac.inria.fr (Francois Pottier) writes: > In article <5obrh6$pvv$1@sun579.rz.ruhr-uni-bochum.de>, > Daniel Boehringer <boehring@biomed.ruhr-uni-bochum.de> wrote: > > >if you have the misckit installed, though :-) > > What is it and where can we find it? Is it portable across OPENSTEP > platforms? (Right now I thought the only portable kits were basically > FoundationKit and AppKit). > > Thanks! > MiscKit is a collection of useful objects that most of them you don't find in FoundationKit or AppKit. Check out http://www.misckit.com. It is functional on NeXTSTEP 3.3 and they just ported it to OS4.1 but as they say in the README file, the OS version is "useless". That's because the porting it's still not complete. However, if you want to identify if a file is a directory from within OPENSTEP then you better use NSFileManager from the FoundationKit. I hope this helps. -- Gabriel Musatescu, A.L.I. Technologies, gabriel@ali.bc.ca "Too much analysis leads to paralysis"
From: Chuck_Esterbrook@orcacomputer.com Newsgroups: comp.sys.next.programmer Subject: Re: NSClassNames -> ??ClassNames Date: 24 Jun 1997 18:53:11 GMT Organization: Virginia Tech, Blacksburg, Virginia Message-ID: <5op52n$lpo$1@solaris.cc.vt.edu> References: <5ocfe4$h5q@news.bu.edu> <5oebrk$g51$1@nntp2.ba.best.com> <5of0ep$1vd$1@news.digifix.com> sanguish@digifix.com (Scott Anguish) wrote: > On 06/20/97, Kenneth C. Dyke wrote: > >On 06/19/97, John Siracusa wrote: > >>I was told by a former NeXT (now Apple) employee at > >>PC Expo this week that all the class names in > >>OpenStep like NSButton, etc. are going to be renamed > >>now that "NS" no longer has any bearing on the OS > >>(Rhapsody). My question is, what nomenclature will > >>they use? "CClassName" like Microsoft? Class names > >>based on libararies like the EOF classes seem to > >>have? > > What I heard is that when accessing ObjC classes from Java, the prefixes won't be there because Java doesn't need prefixes (each package has a namespace). This sounds more reasonable. -- Chuck Esterbrook, Software Eng. http://www.orcacomputer.com/~chuck --------------------------------------------------------------------- chuck_esterbrook@orcacomputer.com / vo 540 231-3475 / fx 540 231-3480 Orca Computer, Inc. / 1800 Kraft Dr. Suite 111 / Blacksburg, VA 24060
From: Chuck_Esterbrook@orcacomputer.com Newsgroups: comp.sys.next.programmer Subject: Re: OpenStep: threads vs garbage collection Date: 24 Jun 1997 19:01:49 GMT Organization: Virginia Tech, Blacksburg, Virginia Message-ID: <5op5it$lpo$2@solaris.cc.vt.edu> References: <5oegkq$sf@news-rocq.inria.fr> <5ofmco$2jr$1@vader.wolfware.ipc.net> cwolf@wolfware.com (Christopher Wolf) wrote: > On 06/20/97, Francois Pottier wrote: > > > >Hi, > > > >I'm discovering OpenStep and have a question which I hope might be > >answered here... > > > >I have read about OpenStep's convention for retaining and releasing > >objects, using reference counts. When an object's reference count > >drops to zero, it becomes eligible for collecting. The question is, > >when will it be collected? > > > Here's the simple rules I use when explaining memory management under OpenStep > to someone: > . . . > > Hope this helps. Don't forget this very important excerpt from the Foundation release notes. This should really be in the main docs for Foundation. I do agree with Sam that retain-release is unwieldy. Its works most of the time, but when it doesn't its a real bitch. "More on Autoreleasing and Retaining The following statements are FALSE: * Returned objects are guaranteed to be valid for the scope of the current method. * Returned objects are guaranteed to be valid until the current autorelease pool is released. * Returned objects are guaranteed to be valid until the end of the current event loop. The Foundation's retain count mechanism operates via -retain and -release. A code fragment such as: id object = [collection returnObject]; creates a reference (in object) to the returned object, but does not increment the retain count of object. If you don't let the system know about your reference, by incrementing the retain count with -retain, the system can't ensure that your reference remains valid for any length of time. In practice, you can get away with such "weak references" most of the time. But the safest approach formalizes your reference to the object: id object = [[collection returnObject] retain]; /* Do operations, some of which may be on 'object' */ ... /* Don't need object any longer */ [object release]; object = nil; " -- end of excerpt -- So in other words the only way to *guarantee* that you've got the object for the duration of your method (let alone the event loop!) is to retain every return value that you use further in the method. Of course, almost no one does this. We all gamble most of the time because it's too much of a pain to do otherwise. I wish they would just add GC. -- 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: scott@doubleu.com (Scott Hess) Newsgroups: comp.sys.next.programmer Subject: Re: posix libraries Date: 24 Jun 1997 21:00:20 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jun24140612@howard.doubleu.com> References: <5oc9g2$ca0@news.blkbox.com> <5odgi5$8qd@sjx-ixn10.ix.netcom.com> In-reply-to: far@ix.netcom.com's message of 20 Jun 1997 08:55:33 GMT In article <5odgi5$8qd@sjx-ixn10.ix.netcom.com>, far@ix.netcom.com(Felipe A. Rodriguez) writes: In article <5oc9g2$ca0@news.blkbox.com> bozack@blkbox.com writes: >I noticed OPENSTEP 4.1 comes with no uname binary, so I thought >I'd write one. > >So I looked at the man page for the uname() system call and >slapped out some code that would theoretically work .. until I >found that the uname() isn't present in the libraries that are >linked automatically. > >I also tried compiling tcl-7.6p2 (so I could compile eggdrop). >the tcl library is archived fine, but tclsh fails during linkage >since it can't find mkfifo() .. fine, I say. I didn't need >tcslsh ANYWAY. All I needed was the library. For mkfifo you can just add the following to your code: int mkfifo (const char *path, mode_t mode) { return mknod (path, mode | S_IFIFO, 0); } You should be able to do the same for uname by extracting the necessary info using BSD system calls. The version of Tcl at my web site has a uname simulation. Perhaps I should apply the above mkfifo() call, too. [For now, it's just compiled out, as I've not run across many packages which _absolutely__ need fifo's. Due to various imcompatibilities, most can optionally use them.] 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: "Jean R. Moreau, Jr." <moreau@fas.harvard.edu> Newsgroups: comp.sys.next.programmer Subject: tutorial Date: Tue, 24 Jun 1997 20:01:47 -0400 Organization: Harvard University Message-ID: <33B05FEB.A71B234@fas.harvard.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, I'm using the tutorials meant for Openstep to help guide me through NextStep Dev 3.2. I have successfully compiled the application, but when I hit the convert button, CurrencyConverter just shuts down. Any suggestions? BTW, I'm running User 3.3 on a TurboStation Jean Moreau
From: jbf_see_sig@frazer.com (James B. Frazer) Newsgroups: comp.sys.next.programmer Subject: Re: Fortran compiler Date: Wed, 25 Jun 1997 02:20:48 -0400 Organization: The Internet Access Company, Inc. Message-ID: <jbf_see_sig-ya023580002506970220480001@news.tiac.net> References: <5ojkhl$lll$1@ncar.ucar.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit In article <5ojkhl$lll$1@ncar.ucar.edu>, dob@mrd3.mmm.ucar.edu (David Blanchard,,,) wrote: > I've scanned the FAQs and looked through the newsgroups but cannot > find the information I seek. I'm interested in a Fortran compiler for > NeXTstep v3.3 running on a Sun Sparcstation. I've found information > for compilers running on Motorola and Intel, but not Sparc. Are there > any? I've had no problem on Moto using f2c from netlib. You have to get the command line library flags right, or combine them. And of course, it won't be as efficient as a commercial Fortran compiler, but it gets you going. There's also a missing function or two, but they are easy to work around. Barney (delete that _see_sig to email me)
From: fpottier@pauillac.inria.fr (Francois Pottier) Newsgroups: comp.sys.next.programmer Subject: Re: OpenStep: threads vs garbage collection Date: 25 Jun 1997 12:07:31 GMT Organization: INRIA Rocquencourt, BP 105, 78153 Le Chesnay Cedex, France Message-ID: <5or1m3$1pr@news-rocq.inria.fr> References: <5oegkq$sf@news-rocq.inria.fr> <5ofmco$2jr$1@vader.wolfware.ipc.net> <5op5it$lpo$2@solaris.cc.vt.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Cc: In article <5op5it$lpo$2@solaris.cc.vt.edu>, <Chuck_Esterbrook@orcacomputer.com> wrote: >Don't forget this very important excerpt from the Foundation release notes. >This should really be in the main docs for Foundation. Thanks for posting. Can you tell me where these release notes are? I didn't know they existed. >The following statements are FALSE: > > * Returned objects are guaranteed to be valid for the scope of > the current method. > > * Returned objects are guaranteed to be valid until the current > autorelease pool is released. > > * Returned objects are guaranteed to be valid until the end of > the current event loop. What? I thought the second statement here was true! If an object has been autoreleased, it will be actually dealloc'ed when its autorelease pool dies, right? So it should be valid until the current autorelease pool goes away. (I assume that the object's autorelease pool is the current autorelease pool - I guess one could set things up so that they're different, but it shouldn't happen in a reasonable program, should it?) I'm confused now. >practice, you can get away with such "weak references" most of the time. But >the safest approach formalizes your reference to the object: > > id object = [[collection returnObject] retain]; > /* Do operations, some of which may be on 'object' */ Hah, but this is no safer, since the object might be released immediately after the returnObject message completes and before you have time to call retain. (I agree the odds are smaller, but in theory it's the same problem, and it will make the bug even harder to track). Unless we are guaranteed that the object lives as long as the current pool, and that the current pool won't die before the end of the method, we're screwed. Calling retain like above won't help (again, in theory). Can anyone shed some light on the subject? >I wish they would just add GC. So do I! (especially since I'm used to working with one, and it allows things you wouldn't have thought possible). -- François Pottier Francois.Pottier@inria.fr http://pauillac.inria.fr/~fpottier/
From: "Frank Alviani" <alviani@ix.netcom.com> Newsgroups: comp.sys.next.programmer Subject: Help with TravelAdvisor Tutorial? (Prelude2Rhapsody) Date: 25 Jun 97 11:47:55 -0500 Organization: Netcom Message-ID: <AFD6B5EE-6051A@205.184.194.171> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Hello all, Warning: I'm an experienced Mac programmer, but a novice OS programmer, so this may be an FAQ... I am working through the Travel Advisor tutorial in the Prelude to Rhapsody package, and cannot get the tabbing order to function correctly, even though examining the project with both the Inspector panel and the list-view of the NIB window show everything is connected correctly. It fails both in the test interface mode and during actual execution.It seems to be tabbing in (approximately) spatial order rather than in the linked order. What is the secret to having the next field linkages work correctly? In addition to the documentation that came with the system, are there other sources for hints about debugging techniques? Thanks in advance, Frank Alviani
From: Žoloft@hegel1.cs.chalmers.se.cs.chalmers.se (Olof Torgersson) Newsgroups: comp.sys.next.programmer Subject: Help Needed: Threads in OPENSTEP Date: 25 Jun 1997 18:46:51 GMT Organization: Chalmers University of Technology Message-ID: <5orp2r$oa3$1@nyheter.chalmers.se> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi, I need some help on threads in OPENSTEP (Mach and NT). I am about to write an application where the main should provide an interface to compose and edit querys to an evaluator running in its own thread. The user of the application should be able to provide a query to the evaluator, get the answer from the evaluator, interrupt the evaluator, and ask the evaluator for its current state. All kinds of hints and examples on how this should be done would be greatly appreciated Olof Torgersson --- Olof Torgersson oloft@cs.chalmers.se Department of Computing Science +46 31 772 54 06 Göteborg University & Chalmers University of Technology S-412 96 GÖTEBORG, SWEDEN http://www.cs.chalmers.se/~oloft/
From: thomas@zippy.sonoma.edu (Thomas Poff) Newsgroups: comp.sys.next.advocacy,comp.sys.mac.advocacy,comp.sys.next.programmer Subject: Re: Rhapsody UI, Lets go translucent Followup-To: comp.sys.next.advocacy,comp.sys.mac.advocacy Date: 26 Jun 1997 00:55:02 GMT Organization: Information Resources and Technology Message-ID: <5osel6$cvj$1@nuke.csu.net> References: <5o6sjq$kb5$1@news2.digex.net> <5o6vb9$5mb@nr1.calgary.istar.net> <maury-1806971102040001@199.166.204.230> <5o901r$bmo@nr1.calgary.istar.net> <maury-1806971220490001@199.166.204.230> <5o9c91$cqu@nr1.calgary.istar.net> <maury-1906971033110001@199.166.204.230> <5ocbfk$jhs$2@news2.digex.net> <maury-2006971044460001@199.166.204.230> <5ofq4g$b3q$4@news2.digex.net> <1997062306512023224@hoorn14.multiweb.net> I agree that this is a pretty darn good idea. There is a performance hit in rendering transparently and as well a problem with some implementations of transparency in 256 color systems. Still it's probably worth doing. Hey has anyone ever made a round window in NeXTStep/Rhapsody? It seems perfectly doable but I never got around to actually implementing it. My only question is can you open a postscript context and either _not_ associate a color with the context so that parts of it are clear or set the backing color to be a transparent color? I did an IB palette suite that rendered themselvs as 20% transparent so that you could see what was behind them in IB. That made things very convenient in terms of laying out app interfaces. If any one has and wants to talk about it, please post a thread in comp.sys.next.programmer! Thomas ---- Dennis (dennis.scp@multiweb.nl) wrote: : John Kheit <jkheit@cnj.digex.net> wrote: : > > > Then I could hide the top of the window off the edge of the : > > > screen, : > : > > Which is exactly what you're doing. : > : > No it's not, your hiding it under the menu bar, so you're getting : > say 16 extra pixels...but you could hide the window off the edge : > of the screen, which would give you 16 extra pixels plus the 21 : > extra pixels of the menubar. The menubar takes up space, period. : Here is an idea. Let's make the menubar and the menu's translucent, This : way you can clearly see trough it and still read it. As you move the : pointer over the menubar it becomes less translucent so you can clearly : read it (hey MetaTools does it quite nice). : Menu's should also be a bit translucent so you can see what is going on : underneeth them (gues, read, remember what your doing) -- <>+<> ////// __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: klui@cup.hp.com (Ken Lui) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.advocacy,comp.sys.mac.programmer.misc,comp.sys.next.programmer Subject: Re: GX remote drawing (Was Re: GX OOP class Date: 18 Jun 1997 18:56:41 GMT Organization: Hewlett-Packard Company Message-ID: <5o9b19$qi7$1@hpax.cup.hp.com> References: <5o5jbg$gra$1@news.lth.se> <AFCD4284-ACA1@206.165.44.53> In article <AFCD4284-ACA1@206.165.44.53>, Lawson English <english@primenet.com> wrote: >There is a LOT of stuff you can do with GX. I wonder if all this functionality cause GX to be slower than something like DPS. Does anyone out there have benchmarks? Maybe PS with its interpreted nature and GX's large featureset overhead would come out the same. Maintenance, codewise, would be more difficult because of excess functionality. Ken -- Ken Lui, klui@cup.hp.com 19111 Pruneridge Avenue M/S 44UR Enterprise Systems Division Cupertino, CA 95014-0795 USA Open Warehouse Team 1.408.447.3230 FAX 1.408.447.1053 Views within this message may not be those of the Hewlett-Packard Company
From: Al Sheehan <asheeh02@ctp.com> Newsgroups: comp.sys.next.programmer Subject: WOApps & Sessions Date: Wed, 25 Jun 1997 11:35:58 +0100 Organization: Cambridige Technology Partners Message-ID: <33B0F48E.38AADB32@ctp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 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 authenticate. Any ideas, thoughts or Six foot blondes looking for a good time ?? Al ...
From: neil@pobox.com (Neil Rhodes) Newsgroups: comp.sys.next.programmer Subject: Re: Help with TravelAdvisor Tutorial? (Prelude2Rhapsody) Date: Wed, 25 Jun 1997 22:49:00 -0700 Organization: Calliope Enterprises, Inc. Message-ID: <neil-2506972249020001@mail.calliopeinc.com> References: <AFD6B5EE-6051A@205.184.194.171> In article <AFD6B5EE-6051A@205.184.194.171>, "Frank Alviani" <alviani@ix.netcom.com> wrote: > >I am working through the Travel Advisor tutorial in the Prelude to Rhapsody >package, and cannot get the tabbing order to function correctly, even >though examining the project with both the Inspector panel and the >list-view of the NIB window show everything is connected correctly. It >fails both in the test interface mode and during actual execution.It seems >to be tabbing in (approximately) spatial order rather than in the linked >order. > >What is the secret to having the next field linkages work correctly? > >In addition to the documentation that came with the system, are there other >sources for hints about debugging techniques? > >Thanks in advance, > >Frank Alviani A FAQ is maintained on Rhapsody development issues at: <http://www2.stepwise.com/cgi-bin/WebObjects/Stepwise/FactBase> Currently, the one and only question is the one you raise. The simple answer is "You need to connect the "initialFirstResponder" outlet from your Window to the first field that you want selected." Apparently the documentation is out of date. -- Neil Rhodes Author of Programming for the Newton neil@pobox.com Newton Programming and Training Newton Essentials class June 9-13 <www.calliopeinc.com/training.html>
From: "Jim L. Morris" <jmorris@silcoinc.com> Subject: Openings Florida/Alabama Newsgroups: comp.sys.next.programmer Organization: SILCO Software Technology,Inc Message-ID: <01bc8243$83a0bb60$2bf58bd0@ws03.silcoinc.com> Date: 26 Jun 97 15:17:48 GMT Sorry to bother you guys, but I wanted to pass a message along to your user group. A client of ours located in Birmingham, Alabama has several permanent openings for Smalltalk Developers. Company is a solutions provider/integrator that specializes in Object-Oriented Systems. Salary is up to 80K plus full benefits. Basic requirements are a minimum of 1 year of Smalltalk experience. H-1 Visa candidates are welcome to apply!!!!!! If someone is not interested in relocating to Birmingham, then they will consider paying for your living expenses in Birmingham Mon-Fri and commuting expenses on the weekends back to Atlanta. We also have openings in Tampa, Florida for Object-Oriented Architects & Developers experienced in some of the following skills: Visual C++, JAVA, CORBA, Versant, ORBIX, BONGO, Rational Rose, ObjectStore, etc… Salaries 50K-150K based on experience. Feel free to pass my name and number around to anyone who might be interested. Thanks, Jim Morris SILCO Software Technology, Inc. 813-939-0603 888-745-2646 (Toll Free) 813-939-0703 (Fax) jmorris@silcoinc.com
Newsgroups: comp.sys.next.marketplace,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software,comp.sys.next.sysadmin,comp.sys.palmtops,comp.sys.pen,comp.sys.powerpc.advocacy,control From: news@news.msfc.nasa.gov Message-ID: <cancel.33B29E0F.757C105B@sprintmail.com> Control: cancel <33B29E0F.757C105B@sprintmail.com> Subject: cmsg cancel <33B29E0F.757C105B@sprintmail.com> no reply ignore Organization: Semi-Automatic Lupine Remover Date: Thu, 26 Jun 1997 17:00:34 GMT Sender: Tamara <tr@sprintmail.com> ignore Make Money Fast post canceled by J. Porter Clark.
From: dix.lorenz@hamburg.netsurf.de (Dix Lorenz) Newsgroups: comp.sys.next.programmer Subject: OpenStep/NT problem Date: Thu, 26 Jun 1997 18:16:58 +0200 Organization: EDV - Beratung Message-ID: <199706261816586561770@dip125-2.hamburg.netsurf.de> Hi, I have just received my Prelude to Rhapsody and installed OpenStep Enterprise for NT. That works and I can compile and run "pure" Obj-C programs. When I try to mix C++ into it, it doesn't. The problem is in one of the systemheaders: /Next/NextLibrary/FrameWorks/System.framework/Headers/redist/NEW.H the compiler complains in the line which says: #include <new> rightly so, because it doesn't exist. where can i get it and what went wrong in the first place? BTW: just deleting the line doesn't help... :-( As it is a MS-file, I also tried replacing it with new.h from my VC++-package, but that didn't work either.... Greetings, Dix
From: asdljf@;asldjf.com Newsgroups: comp.sys.next.programmer Subject: Hot Young Cock Sucking Teens *cock.jpg Date: 26 Jun 1997 18:26:49 GMT Organization: Internet MCI Message-ID: <5ouc99$j9s$217@news.internetmci.com> Are you ready to look at pictures of Teens fucking and sucking? Then cum visit: http://www.sexy-girls.com All models represented on Sexy-Girls are 18 years of Age or Older.
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: asdljf@;asldjf.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5ouc99$j9s$217@news.internetmci.com> Control: cancel <5ouc99$j9s$217@news.internetmci.com> Date: 26 Jun 1997 18:29:59 GMT Message-ID: <cancel.5ouc99$j9s$217@news.internetmci.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!
Newsgroups: comp.sys.next.programmer From: fgalot@x-lan.alienor.fr Subject: [Q]Character conversion Message-ID: <ECDyvC.LEs@x-lan.alienor.fr> Sender: news@x-lan.alienor.fr Organization: x&lan Date: Thu, 26 Jun 1997 13:59:35 GMT I'm working with EOF1.1 on NeXT3.3 and I've a character set problem. In the EOF release notes they say you've got to implement a subclass of NSString which will perform the necessary conversion to and from the NextEncoding. Has anybody ever made it ? I manage to read or insert the data with my subclass. But I am in big troubles for the update. Do you have some examples or tips? Thanks for help. -- --------------------------------------- ® ® | ® O_O ® ® | O_O -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- Fred Galot fgalot@x-lan.alienor.fr
From: scott@doubleu.com (Scott Hess) Newsgroups: comp.sys.next.programmer Subject: OS4.x, pulldown menus, and FirstResponder Date: 26 Jun 1997 23:00:32 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jun26181838@slave.doubleu.com> When I connect a pulldown menu's entry to First Responder, it doesn't call the appropriate action. Instead, I get a console message about "[NSPopupButton methodName:]: selector not recognized". Fun, fun. I've already tried -setRefusesFirstResponder:, which works well for all of the other buttons on the toolbar. [Yes, toolbar, sigh.] I can think of about three reasons for how this is happening in the AppKit, but I can't think of a reason which allows for me to _fix_ it. For now, I've just hardwired the connections. That should_ suffice for now. But long-term, that precludes leveraging off the firstResponder chain. The solution I know will work but am leery of is to create a throw-away object with the actions necessary, and have it forward them to the appropriate responder chain. But that's an _ugly_ solution (this is Objective-C, _not_ C++!). 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: jayg@fuga.com (J. Goyal) Newsgroups: comp.sys.next.sysadmin,comp.sys.next.programmer Subject: JOBS: program managers/UI leads -- handheld device - fuga corp. Date: Thu, 26 Jun 1997 16:08:09 +1000 Organization: Stanford University Message-ID: <jayg.32.00B0467D@fuga.com> Keywords: prgram managers, UI, software Fuga Corporation (www.fuga.com), a Palo-Alto startup in consumer products, is looking for Software Engineers. Working with NexT/UNIX is a plus. If you have the following credentials, please contact Jay at jayg@fuga.com or 415 691 9424. see link--> www.fuga.com\www\jobs.html Job Role 1 - Program Manager (Software Engineering) Requires Project Leadership. Responsible for the entire product-line. Requirements/Roles (Experience in shipped "consumer retail" products a plus) -Product Architect (includes product strategy; prioritizing features/enhancements) -Leading debugging sessions -Defining processes (Design, Code reviews, Doc. guidelines) -Release Coordination (Development,Testing, Bug Tracking, Backups) -C programming (solid experience required) -BS or MS. Job Roles 2 & 3 - User Interface Project Leads (Software Engineering) Should have the potential to take on major responsibility, especially project leadership (includes defining product architecture & leading debugging/testing sessions) for a product-line. Requirements (Experience in shipped "consumer retail" products a plus) -C programming (LINUX/UNIX or DOS; solid experience) -UI experience (a plus) -BS or MS. These are very important job roles where the engineer is involved in programming & testing & being an integral part of a "start-up team" to release a globally available mass consumer product. The product, an exciting consumer handheld device, will be available in retail stores (circuit city, office depot, fry's). You get an opportunity to exercise creativity and to apply technology such that it solves problems of an everyday mass consumer. We are looking for exceptionally bright and driven people. Company offers very attractive compensation and stock-options. ********** End of Message *****************
From: Jason<jlsb@ix.netcom.com> Newsgroups: comp.sys.next.programmer Subject: Don't read this message! Date: 27 Jun 1997 02:43:17 GMT Organization: Netcom Message-ID: <5ov9c5$g16@dfw-ixnews11.ix.netcom.com> CALLING ALL LITERARY MINDS! I am trying to compose a coffee table book on the topic of EXCUSES. To add an original flavor to my project I am attempting to derive my research entirely from people I come in contact with on the Net. If you have, or know of, any interesting, unique, or even bizarre excuses, and would like to participate in this experiment in creative literature, please reply to this message with your excuse. Any excuse that is used in the published version will receive full literary credit. Thank you in advance for helping a fellow netizen. Long live free speech! Sincerely yours, Jason
Newsgroups: comp.sys.next.programmer From: "Eric K. Ringger" <ringger@cs.rochester.edu> Subject: Re: [Q] Main executable in application folder In-Reply-To: Your message of "17 Jun 1997 06:08:29 GMT." <5o59kt$brt$1@leonie.object-factory.com> Message-ID: <199706270154.VAA21741@milli.cs.rochester.edu> Sender: ringger@cs.rochester.edu (Eric K. Ringger) Cc: comp.sys.next.programmer Organization: University of Rochester Computer Science Dept Date: Thu, 26 Jun 1997 21:54:11 -0400 Somebody asked: > Also, if I wish to make the main code say a Perl script or csh > script does Open Step support the possibility and if it does, how > does it know which execution environment to start? David Evans hoped: > That should work fine, provided your script starts with the > "#!/bin/csh" (or whatever) magic cookie. The kernel understands > such things. However, the utility of this inside an app wrapper is > questionable, since stdout will go to the console and there's no > stdin (as far as I know). Christopher Wolf replied: > I thought it would work fine too -- until I tried it. It didn't > work under NeXTSTEP 3.3 for me. Dirk Olmes wrote: > It never worked for me. Neither does it in 4.1. If I double click > the "app" the console shows > > Jun 17 08:06:30 Workspace: Cannot exec /tmp/foo.app: (not a valid program) [...] > I would have loved to have such a scripted app. As far as I can tell, Workspace.app expects a true Mach-O format executable file with the right name to inhabit the app wrapper. Workspace actually examines the segments of that Mach-O and finds the app icon, the icons and extensions for supported document types, services, etc. when it builds its database at login time. A Perl script (or any other script) just doesn't fit the bill, unfortunately. It would certainly seem feasible (for Apple) to change this: *all* resources could live outside the executable (but still within the app wrapper) as separate files, as most resources do nowadays. For example, even now the declaration of services provided by the app doesn't have to be stored in the Mach-O but can be saved in the app wrapper as a separate file. I'm not on my NeXTstation right now, so I can't get at the exact name of that file -- look in Webster.app to see an example. The same could be done for the icons, document extensions, etc. Perhaps that's already possible for *every* resource and I just haven't noticed. Right now, I suppose you could build a simple app that has no other purpose but to fork(2), exec(2) the desired script, and perhaps even pass a few arguments on the sly to the script (e.g., the -NXOpen args). That simple app would itself be a Mach-O with icon, etc. Does this fill the bill of the person who originally posed the question? If such a thing exists, I'd like to try it out. If this wouldn't work, I'd be interested in hearing the reasons. Have fun. --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: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Shipping Frameworks Date: 27 Jun 1997 05:41:40 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5ovjqk$75f$1@owl.slip.net> I have an app that I'd like to distribute but it uses a number of frameworks which I don't want the end user to have to install. Is there a way to statically link selected frameworks or to stick them in the app wrapper like bundles? -- Steve Dekorte - OpenStep consultant - San Francisco
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5ov9c5$g16@dfw-ixnews11.ix.netcom.com> Date: 27 Jun 1997 05:06:23 GMT Control: cancel <5ov9c5$g16@dfw-ixnews11.ix.netcom.com> Message-ID: <cancel.5ov9c5$g16@dfw-ixnews11.ix.netcom.com> Sender: Jason<jlsb@ix.netcom.com> Spam cancelled by sef@kithrup.com
From: erich@powerwareintl.com (Eric Harley) Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software Subject: NeXT Newbie Questions Date: Thu, 26 Jun 1997 22:51:15 -0800 Organization: EdgeMedia Networks Message-ID: <erich-2606972251150001@ppp-207-105-88-12.snrf01.pacbell.net> Hello NeXT Developers, I was wondering if anybody out there could answer some questions for me. 1) If I develop a program under OpenStep for Mach on Intel, can I run that application on Windows NT/95 without installing OpenStep on the Windows machine? 2) Does anybody know about an OpenStep API for the Macintosh? I dont mean Rhapsody. 3) Will OpenStep 4.2 run on a Turbo Slab? Thanks alot for any and all help! erich@powerwareintl.com
From: ulkjhlk@poikjge.com Newsgroups: comp.sys.next.programmer Subject: New Service for your "Pager"...... Date: 27 Jun 1997 09:03:51 GMT Organization: AT&T WorldNet Services Message-ID: <5ovvln$h4k@mtinsc04.worldnet.att.net> ONLY AVAILABLE IN THE: U.S.A, VIRGIN ISLANDS, AND PUERTO RICO CANADA* BE SURE TO GO TO OUR SITE AND CHECK OUT OUR FREE TRIAL. Call everyone and tell them to throw away all of your old home, office, fax, pager, voice-mail, and cellular numbers and give them your New "Virtual Office" 800/888 number! The only number any one will ever need. Are you tired of giving out all of your different phone numbers to everyone? Wouldn’t it be nice to be able to give everyone just ONE phone number that will find you anywhere you are? Even out of town, or in a restaurant, or even on the golf course. Now you can. We have the answer to all your communication needs. It’s called the "Virtual Office". And this new service is loaded. It comes with features like: An automated Call Attendant, Live Call Connect (in real time), Fax Sending, Fax Receiving, Even without a fax machine! E-Mail Notification and Delivery, Without a PC!, Outbound calling, Low Cost Long Distance Service, Inbound/Outbound 800/888 Number, Worldwide Call Transfer, Call Forwarding, Call Screening, Full Service Voice Mail, Nation Wide Pager Notification, Conference Calling, Speed Dialing, Auto Dial, Auto Messaging, Temporary Greeting, Unavailable Greeting, Password Protection, Pager Notification, and No Equipment or Software to buy, "Ever". Priced from $9.95* per month, Plus 10.9 cents per minute per event. That’s less than one phone line would cost per month. For more information please visit our web site: http://www.mynumber.com * Based on our best priced plan. * Canada has a higher per minute rate.
Newsgroups: comp.sys.next.programmer From: bofh@keltia.freenix.fr (Ollivier Robert) Sender: ulkjhlk@poikjge.com Message-ID: <cancel.5ovvln$h4k@mtinsc04.worldnet.att.net> Subject: cmsg cancel <5ovvln$h4k@mtinsc04.worldnet.att.net> Control: cancel <5ovvln$h4k@mtinsc04.worldnet.att.net> Organization: Usenet Canal Historique Date: 27 Jun 1997 09:03:51 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 Jun 27 11:53:52 1997 Original subject was: New Service for your "Pager"......
From: beauvois@usa.net Newsgroups: comp.sys.next.programmer Subject: Accessing web page data Date: 27 Jun 1997 10:51:54 GMT Organization: University of Tennessee Message-ID: <beauvois-ya023580002706970653030001@news.utk.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit I'm working on an app and would like to include a feature that would access a web site and strip it of some of its available data for use in updating parts of my app. Here's an example: I'd like my app to periodically check with a site that maintains a local weather report, grab this weather data (text/images) and have use it as a reference. I have a few ideas on how to do this, but would greatly welcome any others and/or known existing examples of anything similar. Thanks for your help -- CB
From: Barry Leslie <Barry.Leslie@snap.de> Newsgroups: comp.sys.next.programmer Subject: exeutable path name from file viewer on OIpenStep 4.2 Date: Fri, 27 Jun 1997 13:42:10 +0200 Organization: snap.de Message-ID: <33B3A6FD.6727@snap.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Can any body help me or direct me to where I can find help for the following problem. I have a standard unix type executable that when started looks for certain files in its current directory. I would like this application to work correctly when started from the File Viewer but to do this it needs to know the location of the executable. Normally this is easy because you can use the current working directory and the path name used to execute the file from argv[0]. But when started from the File Viewer the current directory is always the HOME directory and the full path name is not in argv[0]. Someone told me this is a known problem and that there is a way to get around it put they didn't know what it was. The executable can be moved around or there may be multiple copies of it on the machine so storing the installation info some where will not work. Thanks for your time, Barry -- Barry Leslie Software Engineer SNAP Innovation Grosse Elbstrasse 39 22767 Hamburg Germany E-mail barry.leslie@snap.de Phone 040/306 36 403 Fax 040/306 36 333 http://www.snap.de
From: rtr@user.transit.ru (transit shell) Newsgroups: comp.sys.next.programmer,comp.sys.next.software Subject: CVS and .nib files Date: 27 Jun 1997 13:44:57 GMT Organization: Transit Newsgate Message-ID: <5p0g4q$dlh@fido.transit.ru> Hi, Could anybody suggest me how to wrap/unwrap .nib files to store them into cvs repository? Thank you in advance. Vladimir
Newsgroups: comp.sys.next.programmer From: allan@ali.bc.ca (Allan Noordvyk) Subject: Re: OpenStep: threads vs garbage collection Message-ID: <ECFssA.FF4@gateway.ali.bc.ca> Sender: nobody@gateway.ali.bc.ca Mime-Version: 1.0 Cc: fpottier@pauillac.inria.fr, software Content-Type: text/plain; charset=us-ascii Organization: ALI Technologies Date: Fri, 27 Jun 1997 13:43:22 GMT References: <5oegkq$sf@news-rocq.inria.fr> <5ofmco$2jr$1@vader.wolfware.ipc.net> <5op5it$lpo$2@solaris.cc.vt.edu> <5or1m3$1pr@news-rocq.inria.fr> In comp.sys.next.programmer Francois Pottier wrote: > In article <5op5it$lpo$2@solaris.cc.vt.edu>, > <Chuck_Esterbrook@orcacomputer.com> wrote: > > ... > >The following statements are FALSE: > > > > * Returned objects are guaranteed to be valid for the scope of > > the current method. > > > > * Returned objects are guaranteed to be valid until the current > > autorelease pool is released. > > > > * Returned objects are guaranteed to be valid until the end of > > the current event loop. > > What? I thought the second statement here was true! If an object has > been autoreleased, it will be actually dealloc'ed when its autorelease > pool dies, right? So it should be valid until the current autorelease > pool goes away. (I assume that the object's autorelease pool is the > current autorelease pool - I guess one could set things up so that > they're different, but it shouldn't happen in a reasonable program, > should it?) The reason is that *release* not *autorelease* might be used within the code which is returning the object. For example say I have a class Randomizer with the following methods: @implementation Randomizer - number { return myNumber; } - (void)rollTheBones { // Release the old number and get a new one. [myNumber release]; myNumber = [[NSNumber numberWithInt:random_func()] retain]; } @end Now say that you have code somewhere else which does this: NSNumber *firstRoll, *secondRoll; [myRandomizer rollTheBones]; firstRoll = [myRandomizer number]; [myRandomizer rollTheBones]; secondRoll = [myRandomizer number]; printf( "Roll 1 = %d, Roll 2 = %d", [firstRoll intValue], [secondRoll intValue] ); This code will crash in the printf statement under most circumstances due to the fact that the Randomizer class uses release not autorelease in its rollTheBones method. The object pointed to by firstRoll will be deallocated during the second call to rollTheBones unless some object other than aRandomizer has retained it. That is why changing the calling code to: NSNumber *firstRoll, *secondRoll; [myRandomizer rollTheBones]; firstRoll = [[myRandomizer number] retain]; [myRandomizer rollTheBones]; secondRoll = [[myRandomizer number] retain]; printf( "Roll 1 = %d, Roll 2 = %d", [firstRoll intValue], [secondRoll intValue] ); [firstRoll release]; [secondRoll release]; will fix the problem. You could also modify the Randomizer class to use autorelease instead of release, but you have no guarantee that this might not be changed in the future. Furthermore, if you do not have source to the Randomizer class (eg. if it were a NeXT or 3rd party library) you don't have that option. Thus a returned object is safe for immediate use, but may legally be deallocated by subsequent actions. In complex systems it may not be obvious what actions may cause deallocation (eg. object A calls object B to get object C, object A then calls object D which calls object B and causes the release of object C, causing C to be deallocated, A calls object C ... crash). Therefore, if you are retaining a reference to the object (even temporarily) you should retain it during your use. Note that we could have also written the calling code to use autorelease instead of release, eg. NSNumber *firstRoll, *secondRoll; [myRandomizer rollTheBones]; firstRoll = [[[myRandomizer number] retain] autorelease]; [myRandomizer rollTheBones]; secondRoll = [[[myRandomizer number] retain] autorelease]; printf( "Roll 1 = %d, Roll 2 = %d", [firstRoll intValue], [secondRoll intValue] ); This is useful if there are a number of distinct exit points from the routine and we don't want to have to remember to call the release methods of your temporary objects at each juncture. Using autorelease has the downside that your temporary objects may hang around in the current autorelease pool longer than absolutely necessary, thus causing unnecessary memory bloat. This can usually be managed through the strategic use of local autorelease pools, however. There are also certain situations where it makes sense to use copy instead of simply retaining the returned object. For example if an object returns an NSString, you actually have no guarantee that it isn't really an instance of the NSMutableString subclass and thus may be changed later on without your knowledge. If your code depends on the returned object being immutable, then you should use the copy method to ensure that you have your own copy. If the object is an immutable object it will implement the copy method as a simple call to retain anyhow, so it has no extra memory overhead. -- Allan Noordvyk, Software Artisan e-mail: allan@ali.bc.ca ALI Technologies Voice: 604.821.6317 Richmond, Canada Fax: 604.279.5468 * NeXT and MIME mail welcome *
From: Laurent Vinet <vinet@ina.fr> Newsgroups: comp.sys.next.programmer Subject: OpenStep 4.2 debugging ? and NEXTIME ? Date: Fri, 27 Jun 1997 16:28:49 +0200 Organization: INA - France Message-ID: <33B3CE21.41C6@ina.fr> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit From a new programmer with 4.2 2 questions : - where is the browser of the class under gdb (like under the 3.3 "Gdb..." menu of edit) ? - how I can make a NSImage of the current frame of a NTMovieDocument (how build a NSBitmapImageRep) ? Any ideas? -- Laurent VINET INA Phone (33) 01.49.83.22.63 4 avenue de l'Europe Fax (33) 01.49.83.25.82 94366 Bry sur Marne Cedex Mailto:vinet@ina.fr France http://www.ina.fr/People/Laurent.VINET/
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] Main executable in application folder Date: 27 Jun 1997 17:49:03 GMT Organization: WolfWare Message-ID: <5p0uef$7f4$1@vader.wolfware.ipc.net> References: <199706270154.VAA21741@milli.cs.rochester.edu> In-Reply-To: <199706270154.VAA21741@milli.cs.rochester.edu> >Christopher Wolf replied: >> I thought it would work fine too -- until I tried it. It didn't >> work under NeXTSTEP 3.3 for me. > >Right now, I suppose you could build a simple app that has no other >purpose but to fork(2), exec(2) the desired script, and perhaps even >pass a few arguments on the sly to the script (e.g., the -NXOpen >args). That simple app would itself be a Mach-O with icon, etc. Does >this fill the bill of the person who originally posed the question? >If such a thing exists, I'd like to try it out. If this wouldn't >work, I'd be interested in hearing the reasons. This works fine - I actually resorted to doing this. The requirement was that the customer wanted a distinct icon to launch a script. So I built an app (including the icon) which had basically one line of code which was system("scriptname") to launch a script which lived within the app wrapper. An awful kludge but it worked just fine. - Chris -- _______________________________________________________________________ Christopher A. Wolf -- WolfWare -- NeXTSTEP/OpenStep/Rhapsody Developer For info about NewsFlash the lightning fast NeXTSTEP news-reader visit our newly revised web site at: http://www.wolfware.com _______________________________________________________________________
From: scott@doubleu.com (Scott Hess) Newsgroups: comp.sys.next.programmer Subject: Re: OpenStep: threads vs garbage collection Date: 27 Jun 1997 17:00:45 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jun27124728@slave.doubleu.com> References: <5oegkq$sf@news-rocq.inria.fr> <5ofmco$2jr$1@vader.wolfware.ipc.net> <5op5it$lpo$2@solaris.cc.vt.edu> In-reply-to: Chuck_Esterbrook@orcacomputer.com's message of 24 Jun 1997 19:01:49 GMT In article <5op5it$lpo$2@solaris.cc.vt.edu>, Chuck_Esterbrook@orcacomputer.com writes: Don't forget this very important excerpt from the Foundation release notes. This should really be in the main docs for Foundation. I do agree with Sam that retain-release is unwieldy. Its works most of the time, but when it doesn't its a real bitch. "More on Autoreleasing and Retaining The following statements are FALSE: * Returned objects are guaranteed to be valid for the scope of the current method. * Returned objects are guaranteed to be valid until the current autorelease pool is released. * Returned objects are guaranteed to be valid until the end of the current event loop. So in other words the only way to *guarantee* that you've got the object for the duration of your method (let alone the event loop!) is to retain every return value that you use further in the method. Of course, almost no one does this. We all gamble most of the time because it's too much of a pain to do otherwise. It's easier to just do: id object = [[collection returnObject] autorelease]; That way the object _will_ exist for the scope of the current method (unless you're doing odd autorelease pool stuff). Even cooler, say you have a -dealloc method like: -(void)dealloc { [oneObject release]; [otherObject release]; [super dealloc]; } You can change the -release calls to -autorelease, and they will still be released during the current pool's release. On the one hand, this means that objects might live longer than otherwise (for instance, if the -dealloc is from a non-autorelease -release call), on the other hand it means that you can continue to reference the object throughout the -dealloc method with no averse consequences. Of course, if you generally use -autorelease instead of -release, the object's longer lifespan isn't really a problem, because it will always be released soon after the owner was deallocated. Personally, I tend to use -retain only when I want to take ownership of an object (put it in an instance variable), and -autorelease to release ownership, rather than -release. 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.next.programmer Subject: Re: OpenStep: threads vs garbage collection Date: 27 Jun 1997 17:00:43 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jun27124619@slave.doubleu.com> References: <5oegkq$sf@news-rocq.inria.fr> <5ofmco$2jr$1@vader.wolfware.ipc.net> <5op5it$lpo$2@solaris.cc.vt.edu> In-reply-to: Chuck_Esterbrook@orcacomputer.com's message of 24 Jun 1997 19:01:49 GMT In article <5op5it$lpo$2@solaris.cc.vt.edu>, Chuck_Esterbrook@orcacomputer.com writes: Don't forget this very important excerpt from the Foundation release notes. This should really be in the main docs for Foundation. I do agree with Sam that retain-release is unwieldy. Its works most of the time, but when it doesn't its a real bitch. "More on Autoreleasing and Retaining The following statements are FALSE: * Returned objects are guaranteed to be valid for the scope of the current method. * Returned objects are guaranteed to be valid until the current autorelease pool is released. * Returned objects are guaranteed to be valid until the end of the current event loop. So in other words the only way to *guarantee* that you've got the object for the duration of your method (let alone the event loop!) is to retain every return value that you use further in the method. Of course, almost no one does this. We all gamble most of the time because it's too much of a pain to do otherwise. It's easier to just do: id object = [[collection returnObject] autorelease]; That way the object _will_ exist for the scope of the current method (unless you're doing odd autorelease pool stuff). Even cooler, say you have a -dealloc method like: -(void)dealloc { [oneObject release]; [otherObject release]; [super dealloc]; } You can change the -release calls to -autorelease, and they will still be released during the current pool's release. On the one hand, this means that objects might live longer than otherwise (for instance, if the -dealloc is from a non-autorelease -release call), on the other hand it means that you can continue to reference the object throughout the -dealloc method with no averse consequences. Of course, if you generally use -autorelease instead of -release, the object's longer lifespan isn't really a problem, because it will always be released soon after the owner was deallocated. 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>
Date: Fri, 27 Jun 1997 13:53:04 -0600 From: idis-info@idiscorp.com Subject: Re: OS/NT with Citrix/WinFrame Newsgroups: comp.sys.next.programmer Message-ID: <867436811.27373@dejanews.com> Organization: Deja News Usenet Posting Service To: gtupar@ctp.com References: <5ols03$agj$1@concorde.ctp.com> Yup! We sell(distribution), install (for Turnkey solutions) and support Citrix WinFrame, as well as use it here in our shop. The standard answer to the "Will WinFrame support a {specific} application?" is - if the app works on NT 3.51 it will work on WinFrame. After that come the Caveats: 1) If the app is a DOS app and the client machines (Citrix is Thin Client, remote control) are Windows, you may not get a full screen window when running that app across Citrix. 2) A DOS application that uses keyboard polling may use up the entire resources of your WinFrame machine for one or two users (Citrix comes with a 15 user base liscense). 3) If the App uses graphics other than standad windows GUI format (like FoxPro for Windows) it may be excruciatingly slow across Citrix. Other than that I am unaware of any bugabears with WinFrame. If you have more questions or need any help, email us at idis-info@idiscorp.com or call us at 1-800-232-5322 In article <5ols03$agj$1@concorde.ctp.com>, gtupar@ctp.com wrote: > > Hey Folks! > > The client asked us if our system will work on top of NT product, called > WinFrame (from Citrix). It seams to provide a sort of functionality similar to > NXHosing (I was told so...). > > Any ideas, rumors, experience? > > Thanks > > G.T. > > -- > ------- > /\/\ Georg Tuparev <georg_tuparev@ctp.com> | Currently in Dublin > / /_ \ Cambridge Technology Partners 118/119 Lower Baggot Street > \ / / Apollo House, Apollolaan 15 Dublin 2, Republic of Ireland > \/\/ 1077 AB Amsterdam, The Netherlands Tel: +353(1)607-9083 > Tel: +31(20)575-0492 Fax: +31(20)575-0500 WWW: http://www.ctp.com -------------------==== Posted via Deja News ====----------------------- http://www.dejanews.com/ Search, Read, Post to Usenet
From: a;dslfj@a;slkdf.com Newsgroups: comp.sys.next.programmer Subject: Hot New Teenage Site young4.jpg Date: 27 Jun 1997 10:02:06 GMT Organization: All USENET -- http://www.Supernews.com Message-ID: <5p032u$d6k@usenet84.supernews.com> Come visit the hottest new sex site on the internet Sexy-Girls Are you ready to visit the hottest teen slut wonderland? The come visit.. http://www.sexy-girls.com All Models represented on Sexy-Girls are 18 years of age or older.
From: thomas@zippy.sonoma.edu (Thomas Poff) Newsgroups: comp.sys.next.programmer Subject: Developer UI Date: 27 Jun 1997 18:56:16 GMT Organization: Information Resources and Technology Message-ID: <5p12cg$ar2$3@nuke.csu.net> Perhaps I can't really whine too much about it but I am wondering if anyone misses the simpler, arguably more elegant version of Project Builder? It would be great if we could eventually get it back, either via Apple or a third-party development toolmaker. The old environment was simple and people did used to complain about that but the flipside was that it used very little screen real-estate and was in _very_ easy to use. Thomas -- <>+<> ////// __v__ __\/__ `\|||/ /---\ """"""" | _ - | (_____) . / ^ _ \ . (q p) | o o | <^-@-@-^> (| o O |) .( O O ), |\| (o)(o) |/| _ooO_<_>_Ooo_ooO_U_Ooo_ooO__v__Ooo_ooO_u_Ooo_ooO__(_)__Ooa__oOO_()_OOo___ [_____}_____!____.}_____{_____|_____}_____i____.}_____!_____{_____}_____] __.}____.|_____{_____!____.}_____|_____{.____}_____|_____}_____|_____!___ [_____{_____}_____|_____}_____i_____}_____|_____}_____i_____{_____}_____]
From: Matthew_Seaman@plsys.co.uk (Matthew Seaman) Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software Subject: Re: NeXT Newbie Questions Date: 27 Jun 1997 12:55:40 GMT Organization: P&L Systems Message-ID: <5p0d8c$h14$1@ironhorse.plsys.co.uk> References: <erich-2606972251150001@ppp-207-105-88-12.snrf01.pacbell.net> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 In <erich-2606972251150001@ppp-207-105-88-12.snrf01.pacbell.net> Eric Harley wrote: > Hello NeXT Developers, > I was wondering if anybody out there could answer some questions for me. > > 1) If I develop a program under OpenStep for Mach on Intel, can I run that > application on Windows NT/95 without installing OpenStep on the Windows > machine? Yes, but: i) You will have to program strictly to the OpenStep api --- avoiding having a few unix-isms creeping in can be tricky. ii) You will need to recompile your app for WinNT/95. Once Yellow Box is released, you will be able to program in Java, and so run without any need for recompilation. iii) You will need to supply the OpenStep runtime dll's with your app if it's the first OpenStep app to be installed on the WinNT/95 machine. Apple lets you do this for free. > 2) Does anybody know about an OpenStep API for the Macintosh? I dont mean > Rhapsody. Apple has announced a "90% compatible" version of the yellow box to run under MacOS 8, which should start becoming available sometime in '98. > 3) Will OpenStep 4.2 run on a Turbo Slab? Yes. > Thanks alot for any and all help! You're welcome. Matthew [Posted and Mailed] -- Certe, Toto, sentio nos in Kansate nin iam adesse. Matthew Seaman P&L Systems, 12 The Broadway, Amersham, Bucks., HP7 0HP, UK Tel: +44 1494 432422 Fax: +44 1494 432478
From: thomas@zippy.sonoma.edu (Thomas Poff) Newsgroups: comp.sys.mac.advocacy,comp.sys.next.programmer Subject: Sun Sparc TOO! (was Re: Prelude2Rhapsody on black hardware!) Followup-To: comp.sys.mac.advocacy,comp.sys.next.programmer Date: 27 Jun 1997 18:46:39 GMT Organization: Information Resources and Technology Message-ID: <5p11qf$ar2$2@nuke.csu.net> References: <rbarris-ya023280001106971149270001@news.intelenet.com> <5o3mv4$rh7$1@news.platinum.com> Prelude installed on a friend's Sparc trivially. No muss no fuss. It was a trivial installation. Thomas Gary W. Longsine (gary-nospam-@screaming.org) wrote: : In <rbarris-ya023280001106971149270001@news.intelenet.com> it appeared that : Rob Barris wrote: : > I finally got around to trying the Prelude OpenStep 4.2 CDROM on my : > 68040 NextStation Turbo. After mounting the disc, running Upgrader.app and : > then clearing off some disk space by its advice, the install worked! : > : > After having heard from any number of people "nahhh, it's an Intel only : > release" it was a pleasant surprise indeed to see that the discs from WWDC : > were actually fat builds for both Intel and "black" NeXT hardware. (So far : > I've only installed the User side FWIW) : > : > However I only have a 240MB HD in that machine, and now it's full. Looks : > like I need to swap in a 1GB drive or something before I continue with : > installing the "Developer" disc. : Hi Rob, : You may know this already, but when slipping a new disk into a NeXTstation, : remember to stay away from the newest (7200RPM) disks like the SeaGate : Baracuda. It's an excellent disk, but it's too hot for the machine. If you : keep the room temperature very cool (less than 70 or so) it would be fine, : but if the room temp gets up to about 80, you'll probably start to see kernel : panics which result from an overheated CPU. : /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. -- <>+<> ////// __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: sanguish@digifix.com (Scott Anguish) Newsgroups: comp.sys.next.programmer Subject: Re: Accessing web page data Date: 27 Jun 1997 21:02:58 GMT Organization: Digital Fix Development Message-ID: <5p19q2$k88$1@news.digifix.com> References: <beauvois-ya023580002706970653030001@news.utk.edu> In-Reply-To: <beauvois-ya023580002706970653030001@news.utk.edu> On 06/27/97, beauvois@usa.net wrote: > > I'm working on an app and would like to include a feature that would >access a web site and strip it of some of its available data for use in >updating parts of my app. Here's an example: > > I'd like my app to periodically check with a site that maintains a local >weather report, grab this weather data (text/images) and have use it as a >reference. > > I have a few ideas on how to do this, but would greatly welcome any others >and/or known existing examples of anything similar. > There is an example of how you can do this type of thing using the NSFileHandle to grab data from http servers. Its on Stepwise in the FactBase... http://www2.stepwise.com/FactBase Look under Rhapsody Development FAQs at the ToolHandle Example.. -- Scott Anguish <sanguish@digifix.com> NEXTSTEP/OpenStep Information <URL:http://www.stepwise.com>
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5p032u$d6k@usenet84.supernews.com> Date: 27 Jun 1997 19:37:54 GMT Control: cancel <5p032u$d6k@usenet84.supernews.com> Message-ID: <cancel.5p032u$d6k@usenet84.supernews.com> Sender: a;dslfj@a;slkdf.com Spam cancelled by sef@kithrup.com
From: MaRK_BeSSeY@NeXT.CoM (Mark Bessey) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 27 Jun 1997 23:29:08 GMT Organization: Apple Computer, Inc. Message-ID: <5p1ic4$rmg$1@news.apple.com> References: <5p12cg$ar2$3@nuke.csu.net> Thomas Poff writes > Perhaps I can't really whine too much about it but I am wondering > if anyone misses the simpler, arguably more elegant version of Project > Builder? If you think the 4.2 version is complicated, wait'll you see the *NEW* Project Builder... -Mark -- Mark Bessey Apple Computer, Inc. -->I DON'T SPEAK FOR APPLE<--
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 28 Jun 1997 00:06:37 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5p1kid$k5s$1@owl.slip.net> References: <5p12cg$ar2$3@nuke.csu.net> Thomas Poff <thomas@zippy.sonoma.edu> wrote: > Perhaps I can't really whine too much about it but I am wondering > if anyone misses the simpler, arguably more elegant version of Project > Builder? This is exactly how I felt until I started using frameworks. I would prefer it if they eliminated the text view beneath the priject file browser and made an open spec for taking to 3rd party class editors. -- Steve Dekorte - OpenStep consultant - San Francisco
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5p32ep$n1q$214@news.internetmci.com> Date: 28 Jun 1997 17:22:33 GMT Control: cancel <5p32ep$n1q$214@news.internetmci.com> Message-ID: <cancel.5p32ep$n1q$214@news.internetmci.com> Sender: asdfjsdlk@;okajsflsd.com Spam cancelled by sef@kithrup.com
From: droleary@alpha.temporal.org (Doc O'Leary) Newsgroups: comp.sys.next.programmer Subject: Re: Accessing web page data Date: 28 Jun 1997 04:16:27 GMT Organization: Netcom Message-ID: <5p236r$85b@dfw-ixnews1.ix.netcom.com> References: <beauvois-ya023580002706970653030001@news.utk.edu> On 27 Jun 1997 10:51:54 GMT, beauvois@usa.net <beauvois@usa.net> asked about extracting data from HTML... I have a framework started that does this. What it actually does is convert SGML into ObjC objects and then allows you to access the hierarchy of elements with SQL-like methods. Unfortunately, it is only used internally and not complete enough to license as a commercial product. If this isn't an immediate need, I should have something more usable in a few month. Anyone with an interest should let me know; I currently have other projects at a higher priority. --------- Doc -- Copyright 1997 by Doc O'Leary. Author of the wildly unsuccessful "DOS and Windows for People Who Still Have a Clue"
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: IbInspector..hwo to deactive color wells Date: 28 Jun 1997 22:54:44 GMT Organization: 21st Century Software, New York City Sender: daver@ts3-4.nj.cnct.com Message-ID: <5p44nk$444$1@darla.visi.com> References: <5p3506$oo9$1@concorde.ctp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 28 Jun 1997 17:54:44 CDT Cc: tsengel@immd9.informatik.uni-erlangen.de In <5p3506$oo9$1@concorde.ctp.com> Thomas Engel wrote: > P.S. Why isn't there any docu for that important class (its not in the > framework :-( ) and why is there no docu for a ton of other features. > PB in 4.2 (and the release in general) looks really good...but that lacking > docu is a niightmare (can you say what NSeraserect does without digging out > the 3.x docu ?.. ect..etc..pp.) I agree. I would totally kill for updated docs for the IB framework. It seems like palettizing stuff is a lot easier now, it's just a lot of guesswork since you have to judged from: - the headers - 3.3 docs - the "Other Kits" conversion guide What would be really helpful here would be for NeXT to update the TTools example to 4.x. The ProgressViewPalette example is okay, but it's too simplistic.. -- :: 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: herren@flannet.middlebury.edu (David Herren) Newsgroups: comp.sys.next.programmer Subject: Help! Openstep on trial Date: Sat, 28 Jun 1997 17:15:32 -0400 Organization: Language Schools of Middlebury College Message-ID: <msg79407.thr-2fcbe9.f4cdd@flannet.middlebury.edu> Mime-Version: 1.0 Content-Type: text/enriched; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Content-ID: <msg79407.thr-2fcbe9.f4cdd.part0@flannet.middlebury.edu> (my apologies for this cross post. I've stopped reading advocacy myself a= nd yet there are likely some individuals who read it but aren't programme= rs, and yet might be able to help us) My organization is a recent, but enthusiastic adopter of Openstep develop= ment technology. With the acquisition, and the promise of maintaining one= set of source code and delivering our academic software across many plat= forms, we feel we've made the right choice, particularly since all of our applications must be Unicode capabl= e (we develop primarily language learning tools). Recently however we have come under attack by microsoft apologists on cam= pus who, for no reason other than old boy politics--certainly not experti= se--have more influence than they warrant. At our busiest time of year, j= ust a few weeks after moving our entire operation to a newly renovated building, and days before our summe= r programs begin, we are being called upon to prepare a presentation whic= h will justify our decision to go with openstep development. Can you help= us? In preparation for the presentation, we would love to include the following: - a list of large corporations who use openstep or nextstep for mission c= ritical applications - a list of academic institutions using openstep or nextstep - the _exact_ citation from Byte magazine in which I have heard they said= "Openstep is probably the most respected software on the planet." - a quotation from the gentlemen who invented the Web at Cern using Nexts= tep as his development platform (and a citation for verification). I reca= ll seeing that he had been quoted as saying that the tools allowed him to= create the Web much faster than anything else. For stupid reasons it's just possible that our vice president will try to= tell us what compiler we can use unless we can present compelling and la= y person understandable information about Openstep. Otherwise we may be t= old we have to use Visual C++. (any information about Visual C++ capabilities/limitations with respect to Uni= code would also be welcome--it's cross platform capabilities would be int= eresting to note too) If you have done any programming, contract or otherwise, for a big corpor= ation and are not prevented from telling us so (no details necessary, jus= t the company name and the nature of the project), please respond to me d= irectly. Any other information which you may be able to provide that we can include in our presentation would = be welcomed as well. If you have chosen openstep development and would ca= re to share information about your own decision making process, we'd love= to hear about that too. = We intend to prepare the presentation in html for posting on the web and = I will gladly make the URL available to these newsgroups--perhaps someone= else may be able to use the information we prepare. Many thanks in advance. Please respond directly to: herren@flannet.middlebury.edu -- = David Herren -------------------------------------------------- Web: http://www.middlebury.edu/~herren/ General: herren@flannet.middlebury.edu NeXTMail only: herren@barcelona.cet.middlebury.edu
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: oirueou@OUREOWUR.COM Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5p49ev$3rt$3138@news.internetmci.com> Control: cancel <5p49ev$3rt$3138@news.internetmci.com> Date: 29 Jun 1997 00:17:39 GMT Message-ID: <cancel.5p49ev$3rt$3138@news.internetmci.com> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: adfkj@woeiruffs.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5p3vt2$i15$2@usenet89.supernews.com> Control: cancel <5p3vt2$i15$2@usenet89.supernews.com> Date: 29 Jun 1997 00:39:25 GMT Message-ID: <cancel.5p3vt2$i15$2@usenet89.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: andrew_abernathy@omnigroup.com Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 29 Jun 1997 02:40:03 GMT Organization: Omni Development, Inc. Message-ID: <5p4hu3$4j0$1@gaea.omnigroup.com> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> lparkyn@prosoft.com (Lyle Parkyn) wrote: > In article <5p12cg$ar2$3@nuke.csu.net> thomas@zippy.sonoma.edu (Thomas > Poff) writes: > > Perhaps I can't really whine too much about it but I am wondering > > if anyone misses the simpler, arguably more elegant version of Project > > Builder? > > > Yes! NS 3.3 was the last good one. The original visionaries are long gone > and it shows in the tools. I disagree in the case of ProjectBuilder. While I do agree that it could (and should) be more flexible for people who want to use multiple apps for the different functionality that PB covers, I personally find the new PB _far_ superior to the old PB, and it's excruciating to go back nowadays when I have to do some 3.3 development. I know a number of others who feel the same way. The new PB makes keeping the open files organized _much_ easier, it's far easier to see who's been modified, switch between files, do global finds/replaces, debug, and a host of other things. I sympathize with those who don't appreciate the new PB - no one environment fits all. But I resent the blanket statement that the new PB is inferior - that's a far from global opinion. Note that there are some tricks you can pull to make PB more to your liking if you don't want the project browser attached to the files, etc. Experiment with it, and you can probably gain back most of what you've lost based on your preferences, while taking advantage of the many improvements. -- andrew_abernathy@omnigroup.com - NeXTmail & MIME ok
From: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: IbInspector..hwo to deactive color wells Date: 28 Jun 1997 13:53:10 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5p3506$oo9$1@concorde.ctp.com> Hi, while working on an Openstep aplette I have the same "trouble" like usual... I need to to deactivete the color well when I get swapped out. Sadly the headers inside teh InterfaceBuidler Framework don't contain alot of docu strings. So which is the proper method to subclass from IbInspector to deactive the color well ? Do I get an ok: or touch: ? or revert ? I would suspect it is an ok..but maybe some knwos this from the top of his head. Thanxs. Aloha Tomi P.S. Why isn't there any docu for that important class (its not in the framework :-( ) and why is there no docu for a ton of other features. PB in 4.2 (and the release in general) looks really good...but that lacking docu is a niightmare (can you say what NSeraserect does without digging out the 3.x docu ?.. ect..etc..pp.)
From: Steve Dekorte <dekorte@slip.net> Newsgroups: comp.sys.next.advocacy,comp.sys.next.misc,comp.sys.next.programmer,comp.sys.next.software Subject: Re: NeXT Newbie Questions Date: 29 Jun 1997 03:36:55 GMT Organization: Slip.Net (http://www.slip.net) Message-ID: <5p4l8n$ine$2@owl.slip.net> References: <erich-2606972251150001@ppp-207-105-88-12.snrf01.pacbell.net> In comp.sys.next.programmer Eric Harley <erich@powerwareintl.com> wrote: > 1) If I develop a program under OpenStep for Mach on Intel, can I run that > application on Windows NT/95 without installing OpenStep on the Windows > machine? No. > 2) Does anybody know about an OpenStep API for the Macintosh? I dont mean > Rhapsody. (I'm assuming you mean MacOS) Someone may be working on GNUstep for Mac. There was also something in Macweek about an OpenStep for MacOS. > 3) Will OpenStep 4.2 run on a Turbo Slab? Yes. -- Steve Dekorte - OpenStep consultant - San Francisco
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: nobody@gateway.aoc.gov Newsgroups: 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.palmtops,comp.sys.pen Subject: cmsg cancel <33b58ee2.3402054@192.136.24.2> Control: cancel <33b58ee2.3402054@192.136.24.2> Date: 29 Jun 1997 03:05:25 GMT Message-ID: <cancel.33b58ee2.3402054@192.136.24.2> 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: 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: <10436866952022@digifix.com> Date: 29 Jun 1997 03:56:52 GMT Organization: Digital Fix Development Message-ID: <1452867556825@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: kc@ignem.omnigroup.com (Ken Case) Newsgroups: comp.sys.next.programmer Subject: Re: OpenStep: threads vs garbage collection Date: 29 Jun 1997 04:30:24 GMT Organization: Omni Development, Inc. Message-ID: <5p4od0$63j$2@gaea.omnigroup.com> References: <5oegkq$sf@news-rocq.inria.fr> <5ofmco$2jr$1@vader.wolfware.ipc.net> <5op5it$lpo$2@solaris.cc.vt.edu> <SCOTT.97Jun27124619@slave.doubleu.com> <5p23cm$fcr$1@vader.wolfware.ipc.net> Christopher Wolf (cwolf@wolfware.com) wrote: : For these reasons and others my personal preference is to always use : explicit releases unless I have specific reason to use autorelease : (i.e. it's a value which I need to return from the method.) And for those reasons and others that's our policy on using -release vs. -autorelease here at Omni. One of the other not-yet-stated reasons to avoid -autorelease is that -release is MUCH more efficient than -autorelease (particularly in subclasses of our OmniFoundation root class OFObject, where we override -retain, -release, and -retainCount to manage an inline retain count for faster access than the default -retain/-release implementation). -- Ken Case kc@omnigroup.com Omni Development, Inc. http://www.omnigroup.com
From: kc@ignem.omnigroup.com (Ken Case) Newsgroups: comp.sys.next.programmer Subject: Re: Does OS4.2 finally fix the printf %g bug? Date: 29 Jun 1997 04:15:13 GMT Organization: Omni Development, Inc. Message-ID: <5p4ngh$63j$1@gaea.omnigroup.com> References: <5npof5$jlf@tribune.usask.ca> <5o94tp$atk@news.orst.edu> <SCOTT.97Jun19093800@slave.doubleu.com> Scott Hess (scott@doubleu.com) wrote: : What if they are actively supporting the broken behaviour? I can't : imagine it would be more than a couple line change, no need to go to : BSD4.4. Certainly there are programs that rely on the current behavior, but I suppose it's reasonable to break that old code in favor of matching every other vendor's implementation. (But it is quite possible that some of the BSD 4.3 stuff relies on that behavior.) Note that the current %g behavior is exactly the behavior that was in BSD 4.3, BSD 4.2, and most other C libraries with which I was familiar (other UNIX, VMS, etc.) before the new "standard" behavior was documented. In fact, since it seems that every vendor had to fix their libraries to conform to the new %g standard, I'm suspicious that it was a documentation error in the standard in the first place: that they meant to document the existing behavior, but stated it imprecisely (using "significant figures" when that wasn't really the right term), and thus when people reading the standard pressed the issue everyone was forced to change their libraries to match the misstated standard. (Of course, this is complete speculation.) Oh, well! It's certainly time to conform, whether that phrasing in the standard was intended or not: it's certainly not likely anyone will go back and change the wording of the standard at this point (though I did have hope for a year or two). -- Ken Case kc@omnigroup.com Omni Development, Inc. http://www.omnigroup.com
From: scott@doubleu.com (Scott Hess) Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <SCOTT.97Jun27124619@slave.doubleu.com> Control: cancel <SCOTT.97Jun27124619@slave.doubleu.com> Date: 28 Jun 1997 16:00:36 GMT Organization: Is a sign of weakness Message-ID: <SCOTT.97Jun28103347@slave.doubleu.com>
From: ttakeo@camel.oasis.tcp-ip.or.jp (Tetuya Takeo) Newsgroups: comp.sys.next.programmer Subject: lookupd bug is fixed? Date: 28 Jun 1997 17:51:46 GMT Organization: Tokai Communication Platform Network Message-ID: <5p3ivi$sag@martini.tcp-net.ad.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-2022-jp Hi, There is a fatal bug on OPENSTEP4.1's lookupd (at least on OPENSTEP4.1J). The bug is that a connection from IP address which NetInfo does not know makes system panic. This bug prevents me from introducing OPENSTEP to my customers. So, I wonder whether the bug is fixed on release 4.2. Thanks in advance. --- Tetuya TAKEO ttakeo@tcp-ip.or.jp (NeXTmail welcomed!)
Newsgroups: comp.sys.next.programmer From: lparkyn@prosoft.com (Lyle Parkyn) Subject: Re: Developer UI Message-ID: <ECHvor.H1t@prosoft.com> Sender: lparkyn@prosoft.com (Lyle Parkyn) Organization: ProSoft Solutions Inc. (RDS Site) References: <5p12cg$ar2$3@nuke.csu.net> Date: Sat, 28 Jun 1997 16:41:15 GMT In article <5p12cg$ar2$3@nuke.csu.net> thomas@zippy.sonoma.edu (Thomas Poff) writes: > Perhaps I can't really whine too much about it but I am wondering > if anyone misses the simpler, arguably more elegant version of Project > Builder? > Yes! NS 3.3 was the last good one. The original visionaries are long gone and it shows in the tools. -- Lyle Parkyn ProSoft Solutions Inc. (http://www.prosoft.com) lparkyn@prosoft.com (NeXTmail, text or MIME formats welcome) Bus:(604)324-3311 Fax:(604)538-7694
From: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 29 Jun 1997 08:26:32 GMT Organization: WolfWare Message-ID: <5p567o$7m4$1@vader.wolfware.ipc.net> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> In-Reply-To: <5p4hu3$4j0$1@gaea.omnigroup.com> On 06/28/97, andrew_abernathy@omnigroup.com wrote: >lparkyn@prosoft.com (Lyle Parkyn) wrote: >> In article <5p12cg$ar2$3@nuke.csu.net> thomas@zippy.sonoma.edu (Thomas >> Poff) writes: >> > Perhaps I can't really whine too much about it but I am wondering >> > if anyone misses the simpler, arguably more elegant version of Project >> > Builder? >> > >> Yes! NS 3.3 was the last good one. The original visionaries are long gone >> and it shows in the tools. > >I disagree in the case of ProjectBuilder. While I do agree that it >could (and should) be more flexible for people who want to use multiple >apps for the different functionality that PB covers, 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, it's far easier to see who's been >modified, switch between files, do global finds/replaces, debug, and a >host of other things. I sympathize with those who don't appreciate the >new PB - no one environment fits all. But I resent the blanket >statement that the new PB is inferior - that's a far from global >opinion. It took me a while to get used to the new Project Builder and the instabilities in the 4.1 release made it frustrating to use. The 4.2 version is a pleasure to use though now that I've gotten used to it. I'm forced to use 3.3 PB at work and continually find myself frustrated by the lack of many of the 4.2 features. I'd advise the original poster to give it a honest chance -- you may be surprised at it's functionality. I still wish they'd make the auto-indentation a bit more smarter or flexible and if I double click on a particular file in the workspace I really wish PB would bring that project and file to the front but other than that I don't find PB getting in my way at all. >Note that there are some tricks you can pull to make PB more to your >liking if you don't want the project browser attached to the files, >etc. Experiment with it, and you can probably gain back most of what >you've lost based on your preferences, while taking advantage of the >many improvements. > >-- >andrew_abernathy@omnigroup.com - NeXTmail & MIME ok > -- _______________________________________________________________________ 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: cwolf@wolfware.com (Christopher Wolf) Newsgroups: comp.sys.next.programmer Subject: Re: OpenStep: threads vs garbage collection Date: 29 Jun 1997 08:39:00 GMT Organization: WolfWare Message-ID: <5p56v4$7ud$1@vader.wolfware.ipc.net> References: <5oegkq$sf@news-rocq.inria.fr> <5ofmco$2jr$1@vader.wolfware.ipc.net> <5op5it$lpo$2@solaris.cc.vt.edu> <SCOTT.97Jun27124619@slave.doubleu.com> <5p23cm$fcr$1@vader.wolfware.ipc.net> <5p4od0$63j$2@gaea.omnigroup.com> In-Reply-To: <5p4od0$63j$2@gaea.omnigroup.com> On 06/28/97, Ken Case wrote: >Christopher Wolf (cwolf@wolfware.com) wrote: >: For these reasons and others my personal preference is to always use >: explicit releases unless I have specific reason to use autorelease >: (i.e. it's a value which I need to return from the method.) > >And for those reasons and others that's our policy on using -release >vs. -autorelease here at Omni. > >One of the other not-yet-stated reasons to avoid -autorelease is that >-release is MUCH more efficient than -autorelease (particularly in >subclasses of our OmniFoundation root class OFObject, where we >override -retain, -release, and -retainCount to manage an inline >retain count for faster access than the default -retain/-release >implementation). Another tip from the trenches for anyone who's listening and maybe learning: If you're going to create an object which you want to stick around it's more efficient to use the alloc/init pair to create the object then to use one of the class factory methods which returns an autoreleased object and then issue an additional retain: i.e. I suggest using this: [[NSMutableArray alloc] initWithCapacity: 16]; rather than this: [[NSMutableArray arrayWithCapacity: 16] retain]; Both statements have the same net result but the first one avoids the overhead of ever adding the object to the autorelease pool. In this cases when you really do want an autoreleased object then use the factory class methods. Making this distinction not only is more efficient but if it's consistently applied helps indicate the author's intent for the scope of the object's life. >-- >Ken Case kc@omnigroup.com >Omni Development, Inc. http://www.omnigroup.com > -- _______________________________________________________________________ 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: tsengel@immd9.informatik.uni-erlangen.de (Thomas Engel) Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: 29 Jun 1997 11:40:24 GMT Organization: Cambridge Technology Partners, Inc. Message-ID: <5p5hj8$i8t$1@concorde.ctp.com> References: <5p12cg$ar2$3@nuke.csu.net> <ECHvor.H1t@prosoft.com> <5p4hu3$4j0$1@gaea.omnigroup.com> andrew_abernathy@omnigroup.com wrote: >lparkyn@prosoft.com (Lyle Parkyn) wrote: >> Yes! NS 3.3 was the last good one. The original visionaries are long gone >> and it shows in the tools. > >I disagree in the case of ProjectBuilder. ... I personally find >the new PB _far_ superior to the old PB, and it's excruciating to go >back nowadays when I have to do some 3.3 development. I know a number >of others who feel the same way. PB 4.2 for President! 4.2 is really stable and by far a more productive environment then 3.3 ever was. PB 4.2 is a lot more fun...looks better and does not flood my screen with windows. I still have a "ton" of suggestions that I could make to make it even better (esp. integration of PB-Ib and how docu is being handled)...but give that rhapsody will be different once more I think we'll just wait and see. Aloha Tomi P.S... I can only dream and prey that they will redesign PB to become the Talignet/Dylan dynamic browser system. Boy would that be great...with plugins and such....yess..yes..yes
From: marcel@system.de Newsgroups: comp.sys.next.programmer Subject: Re: Printing on Non-PostScript Printers Date: 29 Jun 1997 12:39:13 GMT Organization: Technical University Berlin, Germany Distribution: world Message-ID: <5p5l1h$qhj$1@brachio.zrz.TU-Berlin.DE> References: <5p3isq$sag@martini.tcp-net.ad.jp> In article <5p3isq$sag@martini.tcp-net.ad.jp> ttakeo@camel.oasis.tcp-ip.or.jp (Tetuya Takeo) writes: > Hi, > > I'm seeking for the third party products which print on OPENSTEP Enterprise's > PostScript code to non-PostScript printer, Since OPENSTEP Enterprise does not > have such an ability (Although it can rasterize PS code on a screen). Anyone > knows? Our OpenStep printer drivers are coming along nicely, and a test port to NT has also been successful. Marcel Weiher System & Project GmbH Berlin
From: beauvois@usa.net Newsgroups: comp.sys.next.programmer Subject: Round Windows Date: Mon, 30 Jun 1997 06:55:02 -0500 Organization: University of Tennessee Message-ID: <beauvois-3006970655030001@amour.la.utk.edu> Following Thomas Poff's idea, I think round windows would be a very interesting addition. But how would one implement such a thing ?
From: asdfjsdlk@;okajsflsd.com Newsgroups: comp.sys.next.programmer Subject: -Hot Anal Teen Pics * teen59.jpg Date: 28 Jun 1997 13:09:45 GMT Organization: Internet MCI Message-ID: <5p32ep$n1q$214@news.internetmci.com> Looking for XXX Pictures of Teens having Sex ? Then you've come to the right place. Teen-Porno has tons of Teens Fucking and Sucking Cock. Come Visit the hottest New Teen Site on the Internet !!! http://www.teen-porno.com All Models Represented on Teen-Porno are 18 Years of age or older. The Name Teen-Porno Represents Models that are 18 Years of age or older.
From: Cosmo Roadkill <cosmo.roadkill%bofh.int@rauug.mil.wi.us> Sender: ap8@ap8.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33b4efd3.0@news.genesisnetwork.net> Control: cancel <33b4efd3.0@news.genesisnetwork.net> Date: 30 Jun 1997 09:08:52 GMT Message-ID: <cancel.33b4efd3.0@news.genesisnetwork.net> Organization: BOFH Space Command, Usenet Division Article cancelled as EMP/ECP, exceeding a BI of 20. A report will be published shortly on news.admin.net-abuse.bulletins. Sick-O-Spam, Spam-B-Gon!
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: Re: OpenStep: threads vs garbage collection Date: Mon, 30 Jun 1997 06:29:08 -0600 Organization: Southwest Cyberport Message-ID: <33B7A571.50BC@ergotech.com> References: <5oegkq$sf@news-rocq.inria.fr> <eskimo1-2306970939470001@guy-smiley.apple.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Quinn wrote: > Pretty much. Each thread *does* have its own autorelease pool. When that > pool is released depends on the design of your thread. If your thread is > modelled on some sort of periodic action, you could have your thread > release the autorelease the pool at the end of that periodic action, much > like the AppKit does. This is most probably a good idea. Alternatively, > releasing the autorelease pool at the end of the thread is the default > behaviour. > Unless you are creating and destroying threads pretty quickly, or create almost no objects within the threads, this essentially means that you should manage the autorelease pools yourself. If you don't the memory usage of your application will tend to grow pretty fast. This worst case of this is creating a thread when you start the application that is supposed to run forever. Memory usage will increase forever until your system runs out of swap space. It used to be true, under 3.3, that timed procedures used some sort of default autorelease pool that was not freed until the application ended. I don't know whether this is true under OpenStep or not (anyone know?). If so, unless you're writing a service or some other short lived application, you need to manage the autorelease pools again. Jim
From: u8222015@cc.nctu.edu.tw (Spencer Yu) Newsgroups: comp.sys.next.programmer Subject: Prolog for NeXTStation? Date: 29 Jun 1997 15:52:13 GMT Organization: National Chiao Tung University, Hsinchu, Taiwan Message-ID: <5p60bd$30k$1@news2.nctu.edu.tw> Hi I am looking for a Prolog interpreter for my NeXTStaion...where can I find a free one? I would prefer one with GUI but anything will do thanx!
From: Richard Cave <richard_cave@claris.com> Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: Sun, 29 Jun 1997 09:40:45 -0700 Organization: Claris Message-ID: <33B6900C.3C25@claris.com> References: <5p12cg$ar2$3@nuke.csu.net> <5p1ic4$rmg$1@news.apple.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Mark Bessey wrote: > If you think the 4.2 version is complicated, wait'll you see the *NEW* > Project Builder... What does that mean? Richard Cave <speaking for myself>
From: Jim Redman <jim@ergotech.com> Newsgroups: comp.sys.next.programmer Subject: Re: Developer UI Date: Mon, 30 Jun 1997 06:37:45 -0600 Organization: Southwest Cyberport Message-ID: <33B7A899.4A36@ergotech.com> References: <5p12cg$ar2$3@nuke.csu.net> <5p1kid$k5s$1@owl.slip.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit One thing that, at least to me, is a major interface flaw is the launcher window. The buttons on the top mean that you need enough screen real estate to see the whole window. With the buttons at the bottom, you need only enough for the buttons and the last couple of lines of text. Of course, thanks to the marvels of Interface Builder, you can edit the launcher nib and make this change. I highly recommend it. Jim
From: Greg_Anderson@afs.com (Gregory H. Anderson) Newsgroups: comp.sys.next.programmer Subject: Re: Getting OpenStep/Mach program to run on Windows 95 Date: 30 Jun 1997 14:27:21 GMT Organization: Anderson Financial Systems Inc. Message-ID: <5p8fo9$scq@shelob.afs.com> References: <erich-3006970646150001@ppp-207-104-16-89.snrf01.pacbell.net> 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. (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. -- 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: erich@powerwareintl.com (Eric Harley) Newsgroups: comp.sys.next.programmer Subject: Getting OpenStep/Mach program to run on Windows 95 Date: Mon, 30 Jun 1997 06:46:15 -0800 Organization: EdgeMedia Networks Message-ID: <erich-3006970646150001@ppp-207-104-16-89.snrf01.pacbell.net> 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? -Eric
From: Stefan Ried <ried@mpip-mainz.mpg.de> Newsgroups: comp.sys.next.programmer Subject: Re: lookupd bug is fixed? Date: Mon, 30 Jun 1997 15:02:11 +0200 Organization: Max Plank Institut for Polymer Research Message-ID: <33B7AE53.41C6@mpip-mainz.mpg.de> References: <5p3ivi$sag@martini.tcp-net.ad.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Tetuya Takeo wrote: > > Hi, > > There is a fatal bug on OPENSTEP4.1's lookupd (at least on OPENSTEP4.1J). > The bug is that a connection from IP address which NetInfo does not know > makes system panic. This bug prevents me from introducing OPENSTEP to my > customers. So, I wonder whether the bug is fixed on release 4.2. > > Thanks in advance. > > --- > Tetuya TAKEO > ttakeo@tcp-ip.or.jp (NeXTmail welcomed!) 4.2 has a new lookupd. There was a bug conserning netgroups which is fixed. I can't say something something about your behavior. 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\_/ \_/| \______________________________________________________________________/
From: sef@kithrup.com Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <5p7g1r$945$2878@cadmium.aware.nl> Date: 30 Jun 1997 15:03:03 GMT Control: cancel <5p7g1r$945$2878@cadmium.aware.nl> Message-ID: <cancel.5p7g1r$945$2878@cadmium.aware.nl> Sender: <remove@freemail.nll> Spam cancelled by sef@kithrup.com
From: mtrombin@ix.netcom.com (Mark Trombino) Newsgroups: comp.sys.next.programmer Subject: [Q] OpenStep Garbage Collection Date: 30 Jun 1997 18:37:18 GMT Organization: Egghead Billy, Inc. Message-ID: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii I believe that I finally have a handle on how/when to properly retain objects in OpenStep: 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. 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? Thanks! -- |-------------------------------------------------------------------------| | Mark Trombino | J A M S o f t | | mtrombin@ix.netcom.com | Audio DSP Tools for Openstep & Rhapsody | |-------------------------------------------------------------------------|
From: ftouhi@IRO.UMontreal.CA (Majid Ftouhi) Newsgroups: comp.sys.next.programmer Subject: Compiling LEDA for NextStep Date: 30 Jun 1997 19:29:37 GMT Organization: Universite de Montreal Distribution: world Message-ID: <5p91f1$5o9@epervier.CC.UMontreal.CA> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Hi there: I'm trying to compile LEDA for NextStep on Intel. I get this errors g++ -DLEDA_INSTALL -I../../incl -O -I/usr/X11R6/include -c _x_basic.c _x_basic.c:56: X11/X.h: No such file or directory _x_basic.c:57: X11/Xlib.h: No such file or directory _x_basic.c:58: X11/Xutil.h: No such file or directory _x_basic.c:59: X11/keysym.h: No such file or directory *** Exit 1 (ignored) Any one can tell me where are the equivalente files: X.h, Xlib.h, Xutil.h and keysym.h in X11R6? THANKS IN ADVANCE
From: dwy@ace.net (David Young) Newsgroups: comp.sys.next.programmer Subject: Re: [Q] OpenStep Garbage Collection Date: 30 Jun 1997 19:10:50 GMT Organization: 21st Century Software, New York City Sender: daver@ts3-3.nj.cnct.com Message-ID: <5p90bq$r1u$1@darla.visi.com> References: <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 30 Jun 1997 14:10:50 CDT Cc: mtrombin@ix.netcom.com In <5p8ucu$khl@dfw-ixnews10.ix.netcom.com> Mark Trombino wrote: > I believe that I finally have a handle on how/when to properly retain objects > in OpenStep: Almost. :) > 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. That's not quite right. the +array and +dictionary and +stringWith (etc) methods also create objects with a retainCount of 1; however, these objects are added to the current NSAutoreleasePool, which means at the end of the pool's lifetime, they will receive a -release, decrementing their retainCount (and freeing them). In desktop applications, the current pool has a lifetime which ends at the end of the current event. In other applications, you could put a pool around, say, I/O processing from a socket, or some such. If you want to keep these objects around, you need to explicitly -retain them, so that they will have a retainCount of 2 (or more) before the NSAutoreleasePool dies and sends its -release message. IMHO, this system usually works pretty well, and I do prefer this to actual garbage collection as you can get away with simple assignments in places you need to. 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. See NeXT's documentation for NSObject and NSAutoreleasePool. It's been very helpful, at least to me. -- :: 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: Charles William Swiger <cs4w+@andrew.cmu.edu> Newsgroups: comp.sys.next.programmer Subject: Re: Compiling LEDA for NextStep Date: Mon, 30 Jun 1997 18:16:02 -0400 Organization: Fifth yr. senior, Computer Science, Carnegie Mellon, Pittsburgh, PA Distribution: world Message-ID: <Ini30Wm00iVC08=m80@andrew.cmu.edu> References: <5p91f1$5o9@epervier.CC.UMontreal.CA> In-Reply-To: <5p91f1$5o9@epervier.CC.UMontreal.CA> Excerpts from netnews.comp.sys.next.programmer: 30-Jun-97 Compiling LEDA for NextStep by Majid Ftouhi@IRO.UMontre > Any one can tell me where are the equivalente files: X.h, Xlib.h, Xutil.h > and keysym.h in X11R6? By default, they don't exist under NEXTSTEP since NEXTSTEP doesn't ship with X-Windows. You need to get some version of X that comes with the X development environment (headers and libraries)-- your choices include commercial products like co-Xist and freeware like Xnext. -Chuck Charles Swiger | cs4w@andrew.cmu.edu | standard disclaimer ----------------+---------------------+--------------------- I know you're an optimist if you think I'm a pessimist.
From: litagent345@aol.com (litagent345) Newsgroups: comp.sys.next.programmer Subject: cmsg cancel <33b47b5e.0@news1.ibm.net> Control: cancel <33b47b5e.0@news1.ibm.net> Date: 30 Jun 1997 15:57:35 -0400 Organization: AOL http://www.aol.com/ Sender: root@newsbf02.news.aol.com Message-ID: <-33b47b5e.0@news1.ibm.net> Please cancel this posting
From: frank@this.NO_SPAM.net (Frank M. Siegert) Newsgroups: comp.sys.next.programmer Subject: Re: Prolog for NeXTStation? Date: 29 Jun 1997 20:22:12 GMT Organization: Frank's Area 51 Message-ID: <5p6g5k$bi8$1@orista.ipc.uni-tuebingen.de> References: <5p60bd$30k$1@news2.nctu.edu.tw> Cc: u8222015@cc.nctu.edu.tw In <5p60bd$30k$1@news2.nctu.edu.tw> Spencer Yu wrote: > Hi I am looking for a Prolog interpreter for my NeXTStaion...where > can I find a free one? I would prefer one with GUI but anything will do > thanx! See the NeXT archives on peak and peanuts, there is an sbprolog on ftp://ftp.cs.tu-berlin.de/pub/NeXT/developer/languages I cannot comment about its quality as I am not a prolog guy... :-) -- * Frank M. Siegert [frank@this.net] - Home http://www.this.net * NeXTSTEP, Linux, BeOS & PostScript Guy
From: dbriggs@stem.com Newsgroups: comp.sys.next.programmer,comp.sys.next.software Subject: Re: CVS and .nib files Date: 27 Jun 1997 17:14:38 GMT Organization: Systemix, Inc. Message-ID: <5p0sdu$nea@iserver.stem.com> References: <5p0g4q$dlh@fido.transit.ru> 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! A SOLUTION: Bob Vadnais wrote a Palette for InterfaceBuilder, called CVSPalette. CVSPalette attaches to InterfaceBuilder and fixes the problem; it copies the CVS stuff when saving a *.nib document. Christopher Lane recently updated it for InterfaceBuilder 4.2. Those who use NEXTSTEP 3.x and CVS can get CVSPalette from Bob's archive submission: ftp://next-ftp.peak.org/pub/next/sourcelibrary/palettes/ CVS.NIHS.bs.tar.gz CVS.ReadMe Those who use OPENSTEP 4.2 are invited to e-mail me; I'll send you this work by Bob and Christopher. (I've not tested it in 4.0 or 4.1, and this does not apply to OPENSTEP/NT -- CVS is UNIX.) If Bob Vadnais reads this, -- Hey, thanks, Bob! -- he might want to submit this update. Unfortunately, I was unable to contact him at PDH and NeXT. Don Briggs dbriggs@stem.com <standard disclaimer> Of course, fixing InterfaceBuilder itself would be preferable, but the NeXT folks are probably all pretty busy, right now ...
From: "Mark" <mp@digiplace.com> Newsgroups: comp.sys.next.programmer,comp.unix.programmer,comp.unix.questions Subject: Are these entry-level-job handicaps show-stoppers?" Date: 30 Jun 1997 22:07:59 GMT Organization: zarfism Message-ID: <01bc85a2$01ec70c0$bc65a8c6@metnews.hip.cam.org> References: <332DA1F1.41A4@hom.net> <5h7eed$ntq$1@news.platinum.com> A few months back Gary Longsine posted a thoughtful, detailed response (quoted at the bottom of this post) in reply to somebody wondering about getting an entry-level programming job. I'm wondering if he or anyone else would like to comment on my situation - it's similar but with a difference: I don't have the CS degree. I lack no confidence that with some diligent work I can become competent in Unix programming, Objective-C, the OpenStep API, and fundamental OOP principles. However I'm not confident that I can land an entry-level programming job of some kind because * I'm in my mid-thirties * never finished my CS university degree * and I have no workplace programming experience. I fear that even if follow Longsine's prescription to the letter, I'll be shut out of the marketplace. (It could be though that I have an unrealistic and unnecessarily intimidating image of that marketplace.) So - are the handicaps I've listed show-stoppers? Or I am blowing these problems out of proportion? Does it amount to doing something unheard of -- or merely unconventional? I'm kind of in a Catch-22 where I can't afford to go back to finish the degree without getting a good job first and I can't get a decent job without getting that degree. Or is this a self-imposed Catch-22? What's your _hard-headed_ take on this? -- Mark PS Naturally, I do understand that even if the entry-level job was feasible without the degree, that longer-term career considerations are hindered by the lack of it. = - = - = - = - = - = - = - = - = - = - = - = - = - = - = - = - = - = - = Gary W. Longsine <gary-nospam-@screaming.org> wrote in article <5h7eed$ntq$1@news.platinum.com>... > In <332DA1F1.41A4@hom.net> it appeared that jeff parsons wrote: > > Hi, anyone got any suggestions on how to write business application > > to help someone with a B.A. in CS to get hired as a programmer. I > > have some programming skills but not the full stuff everyone wants? > > > > Thanks, > > > > Jeff > > Hi Jeff, > > There are lots of variables that go into picking your path. > > How desperately do you need a job? > How much time are you willing to put into it? > How much can you spend on hardware/software before you get a job? > Do you want to be a god, and learn cool stuff, or just be a Visual Basic > programmer for the next 20 years? > > Here's my advice: > > If you want to be *really* good, then learn to program in C, under UNIX. If > you start on UNIX, and start in C, you will likely master the skills required > to quickly learn any new development environment, OS, or programming language > that you will ever encounter. (of course many of them you'll not like at > that point, but it won't bother you much.) After a couple of years, you'll > get to the point where you can pick up a new language in about a week, if > you're motivated to do so. > > If you are still a student, get a copy of the OPENSTEP for MachOS 4.1 > Academic Kit > OpenStep runs on intel PCs, but not just *any* pc -- check out the NeXT web > site to make sure hardware is supported BEFORE you buy hardware. (same goes > for any UNIX on intel). Alternatively, you could buy a used NeXTstation Mono > fairly cheap (get 32MB of RAM at least and a nice new 2gig disk put in it). > check out: > > Spherical Solutions http://www.orb.com/ > DeepSpace Technologies http://www.deepspacetech.com/ > > Both of these companies have good reputations for dealing in used NeXT > hardware. > > ($300 through your campus bookstore, if they don't have a clue, call NeXT and > they'll help you help them figure it out). > > http://www.next.com > (800) TRY-NEXT > http://www.stepwise.com > http://www.misckit.com > ftp://next-ftp.peak.org > > this will get you started. buy the following books: > > Teach Yourself C in 21 days > Using C on the UNIX System (O'Reilly & Associates http://www.ora.com) > > Now, TYCin21Days used to be a really decent self-education book, there may be > better ones now, i don't know. I'm sure it's still OK. UsingC is great, > once you get the C basics. > > Once you get up to speed in C, start right away with Objective-C under > OpenStep's way cool IDE (integrated development environment). NeXT is a > niche player right now, but they were bought out by Apple in December, and > are providing the foundation of Rhapsody - a new OS from Apple which will run > on PowerMac and Intel based hardware. It will be fun, cool, sexy, and you > will enjoy programming in that environment (if you like programming.) > > When you get to this point, (after you're up to speed in C, not before) call > up Springer-Verlag (publisher) and get "NeXTSTEP Programming" by Garfinkek & > Mahoney -- it's a bit out of date with respect to the NeXT programming > environment, but it's still an excellent introduction in most respects (it's > very well written.) > > If you're not a student, and don't know any, get a Linux box and start in C, > then move to Java when you're up to speed in C. > > Expect to work 2-3 hours a night, 5-6 nights a week (assuming you're FT > employed) and more if you can, for about 3 months before you're feeling your > oats. Don't be frustrated if nobody wants to hire you at that point. You'll > have a better understanding of what things you'll need to teach yourself to > become valuable. At that point you can start learning more about source code > control systems, networking, other stuff. Take some public domain apps with > source code and modify them to make them do something you want. > > At some point you will have built up enough skills and confidence that you > get hired as an entry level programmer and put through the grind. > > Best of Luck. > > /gary > -- > Gary W. Longsine, Systems Engineer | ____/| OpenStep, MachOS, > PLATINUM Technologies, Inc. | \ o.O| Objective-C: > l_o_n_gsine@platinum.com (NeXTmail | =(_)= > (Can i have his spam?) & MIME) |. U Elegance is Relevant. > >
From: Stefan Ried <ried@mpip-mainz.mpg.de> Newsgroups: comp.sys.next.programmer Subject: mmap via vm_* ??? Date: Mon, 30 Jun 1997 15:06:49 +0200 Organization: Max Plank Institut for Polymer Research Message-ID: <33B7AF69.167E@mpip-mainz.mpg.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit 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 ? Thanks. 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\_/ \_/| \______________________________________________________________________/
These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Marcel Waldvogel and Netfuture.ch.