This file contains the mails sent to the GAP forum in July-September 1993. Name Email address Mails Lines Frank Celler fceller@math.rwth-aachen.de 9 335 Martin Schoenert martin@math.rwth-aachen.de 9 210 Joachim Neubueser neubuese@math.rwth-aachen.de 7 221 Dima Pasechnik dima@maths.uwa.edu.au 4 68 Leonard Soicher L.H.Soicher@qmw.ac.uk 3 60 Jaroslav Gurican gurican@alpha.dcs.mff.uniba.cs 2 94 Alexander Hulpke hulpke@abacus.concordia.ca 2 55 Eamonn O'Brien obrien@pell.anu.edu.au 2 48 Dana-Picard Noah dana@bimacs.cs.biu.ac.il 2 44 Toni Greil greil@guug.de 2 32 Dave Donaghy Dave.Donaghy@bristol.ac.uk 2 17 Philip Osterlund osterlu@s5.math.umn.edu 1 58 Harald Boegeholz hwblist@machnix.mathematik.uni-stuttgart.de 1 49 Luis Valero valero@s5.math.umn.edu 1 41 Thomas Breuer sam@math.rwth-aachen.de 1 31 Hiroshi Matsuyama f50014@sakura.kudpc.kyoto-u.ac.jp 1 26 Johannes Mueller Johannes_Mueller@su.maus.de 1 18 Norai Rocco norai@ime.unicamp.br 1 15 Ruth Schwingel RUTH@brunb 1 14 Michael Smith smith@pell.anu.edu.au 1 12 David Sibley sibley@math.psu.edu 1 12 Prof. Knapp knapp@donau.mathematik.uni-tuebingen.de 1 10 N. S. Mendelsohn mendel@ccu.umanitoba.ca 1 9 Daniel Ruberman RUBERMAN@binah.cc.brandeis.edu 1 9 Yasuhiro HONDA honda@smlab.tutkie.tut.ac.jp 1 7 Bill Bogley bogley@math.orst.edu 1 4 Steve Linton sl25@cus.cam.ac.uk 1 3 Jackie Ford ffor@gauss.math.rochester.edu 1 2 This file is in Berkeley mail drop format, which means you can read this file with 'mail -f ' or 'mailx -f '. It is also possible however to read this file with any text editor. From obrien@pell.anu.edu.au Sun Jul 4 08:16:24 1993 From: obrien@pell.anu.edu.au "Eamonn O'Brien" Date: Sun, 4 Jul 93 08:16:24 +0200 Subject: ANU p-Quotient Program Version 1.1 Version 1.1 of the ANU p-Quotient Program is now available by anonymous FTP. This version can be run as part of the share library of GAP 3.2. The files "anupq.tar.Z" and "anupq.zoo" are available by anonymous ftp from "pub/gap" on "samson.math.rwth-aachen.de" They are also available in "pub/gap-3.2" from "pell.anu.edu.au" This program provides access to implementations of the following algorithms: -- p-quotient computations; -- p-group generation; -- a "standard presentation" algorithm which permits isomorphism testing for p-groups; -- an algorithm to compute a description of the automorphism group of a p-group. New features of the first two implementations are now accessible via GAP; the latter two are publicly available for the first time. Feedback and bug reports are welcomed. ++++++++++++++++++++++++++++++++++++++++++++++ Eamonn O'Brien School of Mathematical Sciences Australian National University Canberra, 0200 FAX +61-(0)6-249 5549 Telephone +61-(0)6-249 2963 (office) e-mail obrien@pell.anu.edu.au ++++++++++++++++++++++++++++++++++++++++++++++ From dana@bimacs.cs.biu.ac.il Wed Jul 14 13:34:34 1993 From: dana@bimacs.cs.biu.ac.il "Dana-Picard Noah" Date: Wed, 14 Jul 93 13:34:34 +0200 Subject: DimensionLoewyFactors Dear Forum, I entered some small p-groups as finitely presented groups and used the function DimensionsLoewyFactors. For the first ones everything seemed OK, but once I got the following: gap> g:=Group(a,b,c); Group( a, b, c ) gap> g.name:="g"; "g" gap> g.relators:=[a^3,b^3,c^3,a*b*a^2*b^2,a*c*b*a^2*c^2,b*c*b^2*c^2]; [ a^3, b^3, c^3, a*b*a^2*b^2, a*c*b*a^2*c^2, b*c*b^2*c^2 ] gap> Size(g); 27 gap> DimensionsLoewyFactors(g); Error, This function will not work for finitely presented groups. Please use 'OperationCosetsFpGroup' in order to construct an isomorphic permutation group in G.operations.ConjugacyClasses( G ) called from ConjugacyClasses( G ) called from G.operations.Agemo( G, p ) called from AgemoAbove( L[i], C, p ) called from G.operations.JenningsSeries( G ) called from .. brk> What happens here? Furthermore, I got the same > Please use 'OperationCosetsFpGroup' in order to construct > an isomorphic permutation group when trying to compute Centre or ConjugacyClasses of a finitely presented group. Could somebody send more information and/or examples? Thanks a lot, Thierry Dana-Picard. From neubuese@math.rwth-aachen.de Wed Jul 14 15:33:02 1993 From: neubuese@math.rwth-aachen.de "Joachim Neubueser" Date: Wed, 14 Jul 93 15:33:02 +0200 Subject: Re: DimensionLoewyFactors Thierry Dana-Picard asked: > I entered some small p-groups as finitely presented groups and used > the function DimensionsLoewyFactors. For the first ones everything > seemed OK, but once I got the following: > > gap> g:=Group(a,b,c); > Group( a, b, c ) > gap> g.name:="g"; > "g" > gap> g.relators:=[a^3,b^3,c^3,a*b*a^2*b^2,a*c*b*a^2*c^2,b*c*b^2*c^2]; > [ a^3, b^3, c^3, a*b*a^2*b^2, a*c*b*a^2*c^2, b*c*b^2*c^2 ] > gap> Size(g); > 27 > gap> DimensionsLoewyFactors(g); > Error, This function will not work for finitely presented groups. > Please use 'OperationCosetsFpGroup' in order to construct > an isomorphic permutation group in > G.operations.ConjugacyClasses( G ) called from > ConjugacyClasses( G ) called from > G.operations.Agemo( G, p ) called from > AgemoAbove( L[i], C, p ) called from > G.operations.JenningsSeries( G ) called from > .. > brk> > > What happens here? > > Furthermore, I got the same > > Please use 'OperationCosetsFpGroup' in order to construct > > an isomorphic permutation group > when trying to compute Centre or ConjugacyClasses of a finitely > presented group. > Could somebody send more information and/or examples? The answer is fairly simple: There are no general algorithms to find for a finitely presented group its conjugacy classes, centre or Jennings series (from which in turn the dimensions of the factors of the Loewy series are obtained). This is a consequence of the unsolvability of the word problem. However, if the finitely presented group is indeed finite, a faithful permutation representation can be found and used. GAP does not try to do this automatically but asks the user to use 'OperationCosetsFpGroup' to do just that. Just in very special cases, e.g. if the presentation exhibits that the group is in fact finite abelian, the function 'DimensionsLoewyFactors' will recognise this and directly give the answer. This is perhaps a slightly misleading act of courtesy since it is then in fact puzzling why a function that did answer for some very small examples suddenly does not work. By the way, if you want to compute the dimensions of the Loewy factors of a p-group it is advisable to describe the group by an AG (polycyclic) presentation from which GAP can work directly. I hope that this clarifies the situation. Joachim Neubueser From mendel@ccu.umanitoba.ca Mon Jul 19 22:25:17 1993 From: mendel@ccu.umanitoba.ca "N. S. Mendelsohn" Date: Mon, 19 Jul 93 22:25:17 +0200 Subject: GAP 3.2 on the Mac For those not contacted by H. Lakser, Gap 32 on the Mac may be obtained from ftp anonymous at the address ccu.umanitoba.ca. It is archived as /pub/mac/macgap32. The directory of macgap32 contains the following four files for downloading: MacGap32.sea.hqx; doc32.sea.hqx; lib32.sea.hqx; and macgap32.info. These files are all self extracting. N. S. Mendelsohn From dana@bimacs.cs.biu.ac.il Wed Jul 21 11:02:25 1993 From: dana@bimacs.cs.biu.ac.il "Dana-Picard Noah" Date: Wed, 21 Jul 93 11:02:25 +0200 Subject: Conjugacy classes Dear Forum, Let t be the character table of a group g, given by GAP. Is there a way to get the elements of the various conjugacy classes of g and the correspondence between these classes and the columns of t? Thanks a lot, Thierry Dana-Picard. From sam@math.rwth-aachen.de Wed Jul 21 11:40:23 1993 From: sam@math.rwth-aachen.de "Thomas Breuer" Date: Wed, 21 Jul 93 11:40:23 +0200 Subject: Re: Conjugacy classes Dear Mrs. and Mr. Forum, Thierry Dana-Picard asks the following question. > Let be the character table of a group , given by GAP. Is there a > way to get the elements of the various conjugacy classes of and the > correspondence between these classes and the columns of ? If has been computed from using 'CharTable' then the conjugacy classes of are stored in the component '.conjugacyClasses', and their ordering is the same as that of the columns in . If is a library table then there is no particular representation of the underlying group attached to , so it is not clear what it means to ``get the elements'' of a conjugacy class corresponding to a particular column of . So the problem might be to identify the columns of a library table with the conjugacy classes of a group for which is the character table. Probably many of the classes of will be easily identified by their element and centralizer orders, and maybe other invariants. If this is not sufficient, one can compute the character table of , and then find permutations of columns and rows that transform this table into (e.g. using 'TransformingPermutationsCharTables'). In some cases there is a generic description of conjugacy classes, e.g., the classes of symmetric groups are parametrized by partitions, and the classes of general linear groups are parametrized by normal forms of matrices. This may make it easier to identify the columns and the classes. Kind regards Thomas Breuer (sam@ernie.math.rwth-aachen.de) From fceller@math.rwth-aachen.de Wed Jul 21 12:52:35 1993 From: fceller@math.rwth-aachen.de "Frank Celler" Date: Wed, 21 Jul 93 12:52:35 +0200 Subject: New Grape (Version 2.1) Dear Forum, Leonard Soicher's new version of GRAPE (Version 2.1), a GAP share library package for graph algorithms, is available on our ftp server (samson.math.rwth-aachen.de) as zoo and tar archive. -r--r--r-- 1 ftp 60365 Feb 19 17:28 grp3r2.tar.Z -r--r--r-- 1 ftp 35194 Feb 19 17:28 grp3r2.zoo The following is an extract from the README. ----------------------------------------------------------------------------- GRAPE (GRaph Algorithms using PErmutation groups), Version 2.1, L.H.Soicher@qmw.ac.uk, 12 July 1993. Changes from previous versions of GRAPE other than new functions and new function parameters: (1) Nauty Version 1.7 is now used. Note that canonical labellings could be different from those produced by nauty 1.4, which was used in previous versions of GRAPE. (2) EnumColadj now returns an intersection matrix for the trivial orbital graph, in addition to the intersection matrices for the other orbital graphs. (3) MakeIntransitiveGroupGenerators now called IntransitiveGroupGenerators. (4) VertexName now returns a copy (of the vertex name). GRAPE is provided "as is", with no warranty whatsoever. Please read the copyright notice in the file COPYING. Please send comments on GRAPE, bug reports, etc. to L.H.Soicher@qmw.ac.uk. ----------------------------------------------------------------------------- best wishes Frank From fceller@math.rwth-aachen.de Wed Jul 21 14:32:49 1993 From: fceller@math.rwth-aachen.de "Frank Celler" Date: Wed, 21 Jul 93 14:32:49 +0200 Subject: Re: New Grape (Version 2.1) Dear Forum, >> -r--r--r-- 1 ftp 60365 Feb 19 17:28 grp3r2.tar.Z >> -r--r--r-- 1 ftp 35194 Feb 19 17:28 grp3r2.zoo this should be -r--r--r-- 1 ftp 456034 Jul 21 12:22 grape.tar.Z -r--r--r-- 1 ftp 366431 Jul 21 12:22 grape.zoo best wishes Frank From L.H.Soicher@qmw.ac.uk Wed Jul 21 18:12:58 1993 From: L.H.Soicher@qmw.ac.uk "Leonard Soicher" Date: Wed, 21 Jul 93 18:12:58 +0200 Subject: Important information on the new GRAPE 2.1 The new GRAPE 2.1 gap share library package for computing with graphs ------------------------------------------- *Important Information* This GAP 3.2 share library package is the latest version of GRAPE. Old GRAPE users should note that it includes online documentation, new functions, some new (optional) function parameters, and the nauty version 1.7 system (and they should alter their .gaprc so that an old version of GRAPE is not read, and insert RequirePackage("grape"); instead!). As well as being available from Aachen, GRAPE 2.1 is available via anonymous ftp from galois.maths.qmw.ac.uk (138.37.80.15) in the file pub/grape/2.1/grape.tar.Z. Also available is the file pub/grape/2.1/grapedoc.dvi.Z, the uncompressed version of which includes a description of GRAPE, the installation instructions, and documentation for GRAPE functions (when GRAPE is installed, all this documentation is available online in the usual GAP way). The file grapedoc.dvi is also part of the grape.tar archive, and you should read it! Please email L.H.Soicher@qmw.ac.uk if you install GRAPE 2.1. If you use GRAPE to solve a problem then also tell me about it, and reference L.H.Soicher, GRAPE: a system for computing with graphs and groups, in Proc. 1991 DIMACS Workshop on Groups and Computation, DIMACS Series in Discrete Mathematics and Theoretical Computer Science, to appear. If you use the automorphism group and graph isomorphism testing functions of GRAPE 2.1 then you are also using Brendan McKay's nauty 1.7 package, and should also reference B.D.McKay, nauty users guide (version 1.5), Technical Report TR-CS-90-02, Computer Science Department, Australian National University, 1990. I hope you find GRAPE useful. Please send comments on GRAPE, bug reports, etc. to L.H.Soicher@qmw.ac.uk. From honda@smlab.tutkie.tut.ac.jp Thu Jul 22 07:32:02 1993 From: honda@smlab.tutkie.tut.ac.jp "Yasuhiro HONDA" Date: Thu, 22 Jul 93 07:32:02 +0200 Subject: Gr\"{o}bner bases software Does anyone have software for Gr\"{o}bner bases algorithm on GAP? Thanks. -- Yasuhiro HONDA @ Toyohashi Univ. of Tech. From f50014@sakura.kudpc.kyoto-u.ac.jp Fri Jul 23 02:06:20 1993 From: f50014@sakura.kudpc.kyoto-u.ac.jp "Hiroshi Matsuyama" Date: Fri, 23 Jul 93 02:06:20 +0200 Subject: Problem in FactorsInt Dear Gap Forum, I use Gap on SVR4/386 (Japanese version) with intel 80486/60MHz. I tried factoring the 17-th repunit as follows. But Gap does not factorize it, and exits to shell. --------------------------------------- gap> FactorsInt((10^17-1)/9); Gasman: last bag of type str and size 4 has overwritten the free bag. $ --------------------------------------- Of course, Gap on Sparc station and Gap386.exe can factor (10^17-1)/9 corrrectly. This implies that the problem is in the machine (NEC) or in the Japanes version of UNIX SVR4/386. What the message means ? Please give me any advice. Hiroshi Matsuyama F50014@sakura.kudpc.kyoto-u.ac.jp Dept. of Math. Hyougo Univ. of Teacher Edu. Japan From gurican@alpha.dcs.mff.uniba.cs Fri Jul 23 09:48:22 1993 From: gurican@alpha.dcs.mff.uniba.cs "Jaroslav Gurican" Date: Fri, 23 Jul 93 09:48:22 +0200 Subject: PQ on VAX/VMS Dear Professor, I would like to get sources of PQ and/or NQ for VAX/VMS if you provide them. I tried to get something from your ftp server pell.anu.edu.au, but acces bits are bad, I got message : permission denied for files in PQ, NQ and gap-3.2 directories (but I could list these directories). Sincerely, Jaroslav Gurican, Dept. of Algebra, Comenius University at Bratislava, Slovakia (former Czechoslovakia) From martin@math.rwth-aachen.de Mon Jul 26 14:53:10 1993 From: martin@math.rwth-aachen.de "Martin Schoenert" Date: Mon, 26 Jul 93 14:53:10 +0200 Subject: Upgrade for GAP 3.2 patchlevel 2 (V3R2P2) to patchlevel 3 (V3R2P3) This is to announce the availibility of the third upgrade for GAP 3.2. This upgrade brings version 3 release 2 patchlevel 2 (V3R2P2) to version 3 release 2 patchlevel 3 (V3R2P3). The priority of this upgrade is medium. This upgrade adds adds a group of size 96 that was missing from the library of solvable groups of size <= 100. It also adds symmetric groups on 0 and 1 points. This upgrade fixes the reported problem with subgroups in a composition series of a permutation group having a trivial elements list. Also the functions for permutation groups are a little bit more carefull to avoid creation of unnecessary lists. Coset enumerations in finitely presented groups will now enter a break loop if more than a certain number of cosets have been defined, to give the user the choice to abandon the enumeration. A problem with the computation of double cosets in very small groups is fixed. A problem in the implementation of Dixon's method to compute the character table of a group is fixed. Two functions that assumed that the trivial character is always first in the character table were fixed. 'DisplayCharTable' now handles character tables that have no powermap. A function 'InterpolatedPolynomial' was added. The function that divides polynomials now checks if the divisor and not the dividend is nonzero. An incorrect test in the function that composes mappings is fixed. A function 'LogInputTo' was added to the kernel. The kernel didn't know about the new ranges with three operands, so the third operand was sometimes deleted by the garbage collection, leading to address errors and core dumps. The upgrade is available as file 'upg3r2p3.dif.Z' on the 'ftp' server 'samson.math.rwth-aachen.de'. It should be available on the other 'ftp' servers soon. This time I do not send out this upgrade as e-mail, it is simply too large (about 100 KBytes 'compress'-ed). First unpack the upgrade with 'uncompress upg3r1p3.dif.Z'. Then read the beginning of the unpacked file 'upg3r2p3.dif', which contains instructions how to apply this upgrade. Martin. -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany From hulpke@abacus.concordia.ca Mon Jul 26 16:49:52 1993 From: hulpke@abacus.concordia.ca "Alexander Hulpke" Date: Mon, 26 Jul 93 16:49:52 +0200 Subject: Re: Conjugacy classes Dear Mrs. and Mr. Forum, Thomas Breuer already answered the question of Thierry Dana-Picard. however, there are two minor comments, I'd like to add: When the ``easy'' invariants (orders, centralizers, permutation character etc.) are not sufficient, and one has to compute the character table of , it may be sufficient, to compute this table only partially (especially, if the group is large, the computation of the whole table is not feasible at all). A lot of this identification process could be automized. The routine would be given a group, a character table and (if possible) further informations, as the type of the permutation character (when using permutation representations of small degree, this character is sometimes determined unique by the degree, even without knowing the values on the conjugacy classes). The routine then would try to find a unique mapping and compute further information (conjugacy classes, powermap, partial character table) etc. itself until the correspondence is determined completely (As the character table is known already, some of these computations would be much more easy, then in the general case, as sometimes the character table will give information, that can save tedious computations, e.g. conjugacy tests). Some time ago, I had considered writing a routine of this type, since I needed something similar, but for the particular case, working ``by hand'' had been faster. If there is sufficiently demand for that kind of routine, I would reconsider my decision. (I can not promise anything about availiability of a routine of this type yet.) In case, you would be interested in such a routine please send me a short eMail message (to the adress in the footer, not to the gap-forum itself, as most readers won't be interested in particular answers). Alexander Hulpke (hulpke@bert.math.rwth-aachen.de) From martin@math.rwth-aachen.de Tue Jul 27 07:19:48 1993 From: martin@math.rwth-aachen.de "Martin Schoenert" Date: Tue, 27 Jul 93 07:19:48 +0200 Subject: Re: Gr\"{o}bner bases software Yasuhiro Honda writes in his e-mail message of 1993/07/22 Does anyone have software for Gr\"{o}bner bases algorithm on GAP? Unfortunately no. Note that the current implementation of multivarite polynomials in GAP is not very efficient. Martin. -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany From martin@math.rwth-aachen.de Tue Jul 27 07:33:24 1993 From: martin@math.rwth-aachen.de "Martin Schoenert" Date: Tue, 27 Jul 93 07:33:24 +0200 Subject: Re: Problem in FactorsInt Hiroshi Matsuyama writes in his e-mail message of 1993/07/23 I use Gap on SVR4/386 (Japanese version) with intel 80486/60MHz. I tried factoring the 17-th repunit as follows. But Gap does not factorize it, and exits to shell. --------------------------------------- gap> FactorsInt((10^17-1)/9); Gasman: last bag of type str and size 4 has overwritten the free bag. $ --------------------------------------- Of course, Gap on Sparc station and Gap386.exe can factor (10^17-1)/9 corrrectly. This implies that the problem is in the machine (NEC) or in the Japanes version of UNIX SVR4/386. What the message means ? Please give me any advice. This is a message from Gasman -- GAP's storage manager. Each time an object is allocated Gasman checks that the word behind the previously allocated object has not been overwritten. This check failed in the example above. I.e. the last bag, which was a string with 3 characters (plus a terminating '\0' character) has overwritten the following word. Does the executable work for other problems? Or does it fail for any nontrivial computation? Which C compiler did you use? Have you compiled GAP without the optimizer (i.e., without the '-O' option)? Please send your answer to 'gap-trouble@samson.math.rwth-aachen.de'. If anything interesting comes out of this I will summarize in the GAP forum. Martin. -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany From martin@math.rwth-aachen.de Tue Jul 27 07:39:53 1993 From: martin@math.rwth-aachen.de "Martin Schoenert" Date: Tue, 27 Jul 93 07:39:53 +0200 Subject: Re: PQ on VAX/VMS Jaroslav Gurican writes in his e-mail message of 1993/07/23 I would like to get sources of PQ and/or NQ for VAX/VMS if you provide them. I tried to get something from your ftp server pell.anu.edu.au, but acces bits are bad, I got message : permission denied for files in PQ, NQ and gap-3.2 directories (but I could list these directories). Very strange. I could get the files from 'pell.anu.edu.au' in the directories 'pub/PQ', 'pub/NQ', and 'pub/gap' without any problems. Be that as it may. You can also take the files 'anupq.tar.Z' and 'nq.tar.Z' from 'samson.math.rwth-aachen.de'. Martin. -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany From osterlu@s5.math.umn.edu Tue Jul 27 07:51:57 1993 From: osterlu@s5.math.umn.edu "Philip Osterlund" Date: Tue, 27 Jul 93 07:51:57 +0200 Subject: Redefining "+, -, *, =," etc. Dear Forum: I would like to be able to create a domain in which I redefine addition and multiplication. I was able to do so in version 3.1, but I'm having difficulty in version 3.2. These are defined as functios, as shown below. gap> G := SymmetricGroup(5);; gap> a := AbstractGenerator("a");;b :=AbstractGenerator("b");; gap> a5:= Group(a,b);;a5.relators := [a^2,b^3,(a*b)^5];; gap> D := DirectProduct(G,a5); Group( DirectProductElement( (1,5), IdWord ), DirectProductElement( (2,5), IdWord ), DirectProductElement( (3,5), IdWord ), DirectProductElement( (4,5 ), IdWord ), DirectProductElement( (), a ), DirectProductElement( (), b ) ) gap> ops := D.identity.operations; rec( = := function ( a, b ) ... end, < := function ( a, b ) ... end, * := function ( a, b ) ... end, / := function ( a, b ) ... end, mod := function ( a, b ) ... end, ^ := function ( a, b ) ... end, Comm := function ( a, b ) ... end, Print := function ( a ) ... end ) But I can't access these functions. The obvious guess is: gap> ops.=; Syntax error: record component name expected ops.=; ^ What used to work is: gap> ops.'='; Syntax error: record component name expected ops.'='; ^ Another possible guess is: gap> ops."="; Syntax error: record component name expected ops."="; ^ Just to check that I'm looking at the right object: gap> ops.Print; function ( a ) ... end Thanks for any suggestions. -Philip Osterlund From martin@math.rwth-aachen.de Tue Jul 27 08:01:39 1993 From: martin@math.rwth-aachen.de "Martin Schoenert" Date: Tue, 27 Jul 93 08:01:39 +0200 Subject: Re: Redefining "+, -, *, =," etc. Philip Osterlund writes in his e-mail message of 1993/07/27 I would like to be able to create a domain in which I redefine addition and multiplication. I was able to do so in version 3.1, but I'm having difficulty in version 3.2. These are defined as functios, as shown below. ...example deleted... But I can't access these functions. The obvious guess is: gap> ops.=; Syntax error: record component name expected What used to work is: gap> ops.'='; Syntax error: record component name expected Another possible guess is: gap> ops."="; Syntax error: record component name expected The single quotes are now used to specify character constants, i.e., '=' is the equal character. There are two possibilities. gap> ops.\=; function ( left, right ) ... end gap> ops.("="); function ( left, right ) ... end The first is preferrable, because in the second one the string value is looked up at runtime. Martin. -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany From ffor@gauss.math.rochester.edu Tue Jul 27 14:18:28 1993 From: ffor@gauss.math.rochester.edu "Jackie Ford" Date: Tue, 27 Jul 93 14:18:28 +0200 Subject: unsubscribe From RUBERMAN@binah.cc.brandeis.edu Tue Aug 10 16:28:35 1993 From: RUBERMAN@binah.cc.brandeis.edu "Daniel Ruberman" Date: Tue, 10 Aug 93 16:28:35 +0200 Subject: SL_2 Dear Gap Forum--I was hoping to do some investigations concerning SL_2 and PSL_2 of the ring of integers in Q(sqrt(p)). These groups are finitely presented (I believe) but I haven't been able to find an explicit presentation in the literature. Specific examples (e.g. p=5) would be of interest as well as the general case. Thank you, Daniel Ruberman From greil@guug.de Tue Aug 10 16:55:13 1993 From: greil@guug.de "Toni Greil" Date: Tue, 10 Aug 93 16:55:13 +0200 Subject: Re: SL_2 > > Dear Gap Forum--I was hoping to do some investigations concerning > SL_2 and PSL_2 of the ring of integers in Q(sqrt(p)). These groups are > finitely presented (I believe) but I haven't been able to find an explicit > presentation in the literature. Specific examples (e.g. p=5) would be of > interest as well as the general case. > > Thank you, > Daniel Ruberman > One reference is: Benjamin Fine, "Algebraic Theory of the Bianchi Groups" - Monographs and textbooks in pure and applied mathematics; 129 - Marcel Dekker, Inc. New York and Basel (1989) Toni Greil, Muenchen (Germany) greil@guug.de From obrien@pell.anu.edu.au Tue Aug 10 19:38:33 1993 From: obrien@pell.anu.edu.au "Eamonn O'Brien" Date: Tue, 10 Aug 93 19:38:33 +0200 Subject: break command in GAP language As a follow-up to recent discussions with Gap developers I propose that a "break;" command be provided in the GAP language. It would perform an identical function to the comparable command in C -- namely to exit from a loop. Are there other GAP users who wish to have such provided? Martin has given public assurance that he will be guided by the opinion of the forum on this topic. Eamonn O'Brien From sibley@math.psu.edu Tue Aug 10 21:37:56 1993 From: sibley@math.psu.edu "David Sibley" Date: Tue, 10 Aug 93 21:37:56 +0200 Subject: Re: break command in GAP language > As a follow-up to recent discussions with Gap developers > I propose that a "break;" command be provided in the GAP > language. It would perform an identical function to the > comparable command in C -- namely to exit from a loop. > > Are there other GAP users who wish to have such provided? > Martin has given public assurance that he will be guided > by the opinion of the forum on this topic. Yes. I would certainly use such a command. From bogley@math.orst.edu Tue Aug 10 23:23:29 1993 From: bogley@math.orst.edu "Bill Bogley" Date: Tue, 10 Aug 93 23:23:29 +0200 Subject: Re: break command in GAP language A built-in command to abort runaway processes would be very helpful. Bill Bogley From dima@maths.uwa.edu.au Wed Aug 11 04:57:05 1993 From: dima@maths.uwa.edu.au "Dima Pasechnik" Date: Wed, 11 Aug 93 04:57:05 +0200 Subject: Re: break command in GAP language > > As a follow-up to recent discussions with Gap developers > I propose that a "break;" command be provided in the GAP > language. It would perform an identical function to the > comparable command in C -- namely to exit from a loop. > > Are there other GAP users who wish to have such provided? > Martin has given public assurance that he will be guided > by the opinion of the forum on this topic. > > Eamonn O'Brien Yes, such a command would be quite useful. (Adepts of "structural" programming would disagree.) Dmitri Pasechnik From dima@maths.uwa.edu.au Wed Aug 11 05:04:08 1993 From: dima@maths.uwa.edu.au "Dima Pasechnik" Date: Wed, 11 Aug 93 05:04:08 +0200 Subject: stack overflow Dear Forum, How one can recognise the sutuation when a recursive GAP function causes the stack overflow. How one can find the size of space avaliable for such a stack. Thanks in advance, Dima Pasechnik From martin@math.rwth-aachen.de Tue Aug 24 01:18:28 1993 From: martin@math.rwth-aachen.de "Martin Schoenert" Date: Tue, 24 Aug 93 01:18:28 +0200 Subject: Re: stack overflow Dima Pasechnik writes in his e-mail message of 1993/08/11 How one can recognise the sutuation when a recursive GAP function causes the stack overflow. How one can find the size of space avaliable for such a stack. The answer to the first question is unfortunatly ``not at all''. Apart from getting a core dump, that is. The answer to the second question depends on the operating system. Under most variants of UNIX there is something called 'limit' or 'ulimit', which can be used to get and set the process parameters, such as maximal data space or stack space. Martin. -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany From martin@math.rwth-aachen.de Tue Aug 24 01:27:10 1993 From: martin@math.rwth-aachen.de "Martin Schoenert" Date: Tue, 24 Aug 93 01:27:10 +0200 Subject: Re: break command in GAP language Eamonn O'Brien writes his e-mail message of 1993/08/10 As a follow-up to recent discussions with Gap developers I propose that a "break;" command be provided in the GAP language. It would perform an identical function to the comparable command in C -- namely to exit from a loop. Ok, I have 6 votes for a 'break' statement, and no vote against it. I will put this on my to-do list. Martin. -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany From dima@maths.uwa.edu.au Tue Aug 24 03:15:43 1993 From: dima@maths.uwa.edu.au "Dima Pasechnik" Date: Tue, 24 Aug 93 03:15:43 +0200 Subject: Re: stack overflow > How one can recognise the sutuation when a recursive GAP function > causes the stack overflow. How one can find the size of space avaliable > for such a stack. > > The answer to the first question is unfortunatly ``not at all''. Apart > from getting a core dump, that is. The answer to the second question > depends on the operating system. Under most variants of UNIX there is > something called 'limit' or 'ulimit', which can be used to get and set > the process parameters, such as maximal data space or stack space. Let me suggest the memory control to be implemented in future versions, to avoid those problems. I'd guess it wouldn't affect the speed. Regards, Dima Pasechnik From martin@math.rwth-aachen.de Tue Aug 24 07:58:04 1993 From: martin@math.rwth-aachen.de "Martin Schoenert" Date: Tue, 24 Aug 93 07:58:04 +0200 Subject: Unsubscriptions This is a reminder that to unsubscribe one should send an e-mail message to 'listserv@samson.math.rwth-aachen.de' (not to 'gap-forum') containing the line 'unsubscribe gap-forum'. Thank you, Martin. -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany From Dave.Donaghy@bristol.ac.uk Tue Aug 24 10:34:47 1993 From: Dave.Donaghy@bristol.ac.uk "Dave Donaghy" Date: Tue, 24 Aug 93 10:34:47 +0200 Subject: Calculating Invariants I know this isn't really a GAP question, but does anyone out there know if there's a package available that will calculate the invariants of the action of a (small) finite group on a purely transcendental field extension over some (arbitrary but infinite) base field? The action can be given explicitly as the action on the transcendental generators, but ISN'T a permutation action. (Even though the group is often the symmetric group.) Hopeful thanks, Dave Donaghy From knapp@donau.mathematik.uni-tuebingen.de Tue Aug 24 11:54:35 1993 From: knapp@donau.mathematik.uni-tuebingen.de "Prof. Knapp" Date: Tue, 24 Aug 93 11:54:35 +0200 Subject: GAP3.3 While attending the group theory conference in Galway I was told by the GAP people who ran the GAP workshop that the new release GAP3.3 were available since August 1st. I tried to get the new release from the samson server in Aachen, but apparently the new release is not yet available on pub/gap as all file names indicate the release 3.2 . What has happened? A readme file on the directory pub/gap would also be desirable. Wolfgang Knapp , University of Tuebingen, Germany KnappWD@mailserv.zdv.uni-tuebingen.de From neubuese@math.rwth-aachen.de Tue Aug 24 12:02:38 1993 From: neubuese@math.rwth-aachen.de "Joachim Neubueser" Date: Tue, 24 Aug 93 12:02:38 +0200 Subject: Re: Calculating Invariants I do not have an answer to Dave Donaghy's question: > I know this isn't really a GAP question, but does anyone out there know > if there's a package available that will calculate the invariants of > the action of a (small) finite group on a purely transcendental field > extension over some (arbitrary but infinite) base field? The action can be > given explicitly as the action on the transcendental generators, but ISN'T > a permutation action. (Even though the group is often the symmetric group.) but the recommendation to contact Larry Grove (Tucson) (grove@math.arizona.edu) who has used GAP and Maple in the investigation of invariants during a visit to Aachen and might have some advice. Since this is an area that we have been asked about more often, I would welcome further comments in the GAP-forum, if somebody has some hints. Joachim Neubueser From neubuese@math.rwth-aachen.de Tue Aug 24 12:28:47 1993 From: neubuese@math.rwth-aachen.de "Joachim Neubueser" Date: Tue, 24 Aug 93 12:28:47 +0200 Subject: Re: GAP3.3 Wolfgang Knapp writes: > While attending the group theory conference in Galway I was told by the GAP > people who ran the GAP workshop that the new release GAP3.3 were available > since August 1st. I tried to get the new release from the samson server in > Aachen, but apparently the new release is not yet available on pub/gap as > all file names indicate the release 3.2 . What has happened? A readme file > on the directory pub/gap would also be desirable. Sorry, this must be a misunderstanding. We did use a prerelease copy of GAP 3.3 in Galway, but the release of this version should still take a few weeks, e.g. we want to fix a few things that we learned through the use in Galway. At present the latest version is still 3.2, patchlevel 3, i.e. three patches for version 3.2 have been issued. As a general remark, let me add that version 3.3 will not differ dramatically from version 3.2, it is more a consolidation of the latter, fixing the bugs that have been noticed, of course, and replacing some functions by more efficient ones, having some but not very many new functions. Note that recently some new share libraries have become available . We hope to bring in more new stuff with the next version after 3.3, which however should definitely not be expected before the end of the year. The readme file with the GAP announcement is contained in the directory pub/gap as pub/gap/README. Joachim Neubueser From martin@math.rwth-aachen.de Tue Aug 24 12:50:04 1993 From: martin@math.rwth-aachen.de "Martin Schoenert" Date: Tue, 24 Aug 93 12:50:04 +0200 Subject: Re: Re: stack overflow Dima Pasechnik writes in his e-mail message of 1993/08/24 Let me suggest the memory control to be implemented in future versions, to avoid those problems. I'd guess it wouldn't affect the speed. I never tried it, but I think that it would indeed affect the performance. It would mean (at least) one more increment/decrement and test of a global variable in 'EvFuncCall', and this function is already a series bottleneck for many computations. Martin. -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany From dima@maths.uwa.edu.au Tue Aug 24 13:29:58 1993 From: dima@maths.uwa.edu.au "Dima Pasechnik" Date: Tue, 24 Aug 93 13:29:58 +0200 Subject: Re: Re: stack overflow > > Dima Pasechnik writes in his e-mail message of 1993/08/24 > > Let me suggest the memory control to be implemented in future versions, > to avoid those problems. I'd guess it wouldn't affect the speed. > > I never tried it, but I think that it would indeed affect the > performance. It would mean (at least) one more increment/decrement and > test of a global variable in 'EvFuncCall', and this function is already a > series bottleneck for many computations. I'd give it less than 0.1% of the speed anyway. Indeed, each memory allocation in the interpreted languages (say, Basic, or GAP) takes hundreds of commands. > > Martin. > > -- .- .-. - .. -. .-.. --- ...- . ... .- -. -. .. -.- .- > Martin Sch"onert, Martin.Schoenert@Math.RWTH-Aachen.DE, +49 241 804551 > Lehrstuhl D f"ur Mathematik, Templergraben 64, RWTH, 52056 Aachen, Germany -- Dima Pasechnik, Dept. of Mathematics, Univ. of Western Australia, Nedlands WA 6009, Australia e-mail:dima@maths.uwa.edu.au From sl25@cus.cam.ac.uk Tue Aug 24 13:39:43 1993 From: sl25@cus.cam.ac.uk "Steve Linton" Date: Tue, 24 Aug 93 13:39:43 +0200 Subject: Re: Calculating Invariants I believe Leonard Soicher had a program to do domething of this kind. Steve From valero@s5.math.umn.edu Tue Aug 24 23:16:47 1993 From: valero@s5.math.umn.edu "Luis Valero" Date: Tue, 24 Aug 93 23:16:47 +0200 Subject: GroupHomomorphismByImages I have been having trouble using GroupHomomorphismByImages, which seems to take a very long time in a fairly straightforward situation. Here is an example: g:=Group((1,2,3,4,5,6),(1,2)); R:=[]; for k in [1..2] do M:=NullMat(6,6,GF(3)); for i in [1..6] do j:=i^g.generators[k]; M[i][j]:=Z(3)^0; od; R[k]:=M; od; GroupHomomorphismByImages(g,GeneralLinearGroup(6,3),g.generators,R); This piece of code sets up a pair of matrices R{1] and R[2] which represent the permutation action of the two generators of the symmetric group of degree 6 on the natural permutation representation. Everything except the last line takes hardly any time. The final line then takes some minutes to run. Evidently GroupHomomorphismByImages is doing a lengthy computation, but by looking at the manual I can't tell what it is. From looking at interrupt messages it appears to be computing a stablizer chain. I can think of several things it might be doing that would take rather a long time, but am not sure which is actually taking place. Could you please shed some light on this situation? My main aim is to set up a record which stores some of the information in GroupHomomorphismByImages, but I don't necessarily need all this information. Is there any way to bypass the computation that GroupHomomorphismByImages is doing? I am running version 3.2 of GAP without any of the subsequent patches. With thanks for your help, Luis Valero valero@math.umn.edu School of Mathematics, University of Minnesota, Minneapolis MN 55455 From fceller@math.rwth-aachen.de Wed Aug 25 14:26:26 1993 From: fceller@math.rwth-aachen.de "Frank Celler" Date: Wed, 25 Aug 93 14:26:26 +0200 Subject: Re: GroupHomomorphismByImages Dear Luis Valero, >> I have been having trouble using GroupHomomorphismByImages, which seems >> to take a very long time in a fairly straightforward situation. Here is >> an example: unfortunately 'GroupHomomorphismByImages' has a lot of hidden pitfalls, we are still trying to clean this up. The following is going wrong: the function tries to decide if the set is a subset if the general linear group, although this should be straight forward, this is done at the moment using a permutation representation for GL. You can avoid this check by calling h := g.operations.GroupHom..(g,GL(6,3),g.generators,R) directly instead of >> GroupHomomorphismByImages(g,GeneralLinearGroup(6,3),g.generators,R); >> Could you please shed some light on this situation? >> in GroupHomomorphismByImages, but I don't necessarily need all this >> information. Is there any way to bypass the computation that >> GroupHomomorphismByImages is doing? However, this is still not good enough, because 'GroupHom..' does not trust you. You must set h.isMapping := true; h.isHomomorphism := true; h.isGroupHomomorphism := true; in order to avoid unnecessary calculations. 'GL' and 'GroupHom...' will be changed to be more user friendly, but I am afraid that this will happen *after* the release of 3.3. best wishes Frank Celler From L.H.Soicher@qmw.ac.uk Wed Aug 25 16:23:08 1993 From: L.H.Soicher@qmw.ac.uk "Leonard Soicher" Date: Wed, 25 Aug 93 16:23:08 +0200 Subject: Re: Calculating Invariants My (old, PASCAL) programs for invariants are for linear groups over finite fields, but the algorithms should work for linear groups over an arbitrary field. To be more precise, given matrices generating a subgroup G of GL(n,q), and a natural mnumber k, my programs calculate a basis for the vector space of the n-variable degree k polynomials invariant under G. Regards, Leonard Soicher. From Dave.Donaghy@bristol.ac.uk Thu Aug 26 13:41:54 1993 From: Dave.Donaghy@bristol.ac.uk "Dave Donaghy" Date: Thu, 26 Aug 93 13:41:54 +0200 Subject: Re: Calculating Invariants Dear Leonard, Many thanks for your reply. Unfortunately, my group action isn't linear. Still, never mind, eh? Dave Donaghy From smith@pell.anu.edu.au Fri Aug 27 05:47:41 1993 From: smith@pell.anu.edu.au "Michael Smith" Date: Fri, 27 Aug 93 05:47:41 +0200 Subject: Compiling GAP under MiNT on 68040 Mac Forum readers, I am trying to get GAP running under a port of MiNT to the Mac. When compiled with "make st-tos-gcc2", the resulting executable exits immediately with memory allocation problems (I cannot remember the precise message, something about corrupted workspace). Anyway, my question is: has anyone compiled and run GAP under MiNT, and if so, were any patches required to achieve this? Thanks, Michael. From norai@ime.unicamp.br Mon Aug 30 21:20:17 1993 From: norai@ime.unicamp.br "Norai Rocco - Prof. Visitante" Date: Mon, 30 Aug 93 21:20:17 +0200 Subject: call a GAP session? Dear Gap-forum, A student here has been working with a (fortran) program to find out a kind of "first vector" in communication theory. One of its variables is to run certain iterations for a min-max problem, under a (matrix) group action, say X^T*g*X, g \in G, norm(X)=1. The first idea was to call GAP during this phase, not by simply storing a list "Elements(G)" and then using the list (this should take a lot of storage space), but to take advantage of GAP facilities to run things like an orbit algorithm. Could anuyone give me some hint concerning this "dynamical" use of GAP ? Thanks in advance, Norai Rocco. From fceller@math.rwth-aachen.de Wed Sep 1 12:29:43 1993 From: fceller@math.rwth-aachen.de "Frank Celler" Date: Wed, 1 Sep 93 12:29:43 +0200 Subject: Re: Calculating Invariants Dear Forum, some letter to the forum where rejected by our list-server, because they were sent from different accounts than usual. sorry for any delay Frank ----------------------------------------------------------------------------- Date: Tue, 24 Aug 93 13:55:31 +0200 Comment: GAP Forum Originator: gap-forum@samson.math.rwth-aachen.de Errors-To: martin@samson Reply-To: Sender: gap-forum@samson Version: 5.31 -- Copyright (c) 1991, Anastasios Kotsikonas From: gap-forum@samson To: martin@samson Subject: Re: Calculating Invariants I might have two (!) answers for Mr Donaghy's request; both are programs written in Maple. They calculate the ring of invariants of a finite linear group defined over K=Q . In other words, the group acts on the indeterminates of a purely transcendental extension of K by linear transformations. "Calculate" means that they calculate a Cohen-Macaulay basis for the ring of invariants. The first program has been written by Janet McShane together with Larry Grove, whom Herr Neubueser already mentioned in his reply (email grove@math.arizona.edu). The second one is by me. It has been published in the Maple Share Library and can be obtained by anonymous ftp on neptune.inf.ethz.ch (129.132.101.33), but also directly from me by email. In my program you can also take an algebraic number field for K, and it can calculate syzygies between the generating invariants. Hope I won't get kicked out of the forum for advertising Maple programs! Gregor Kemper From fceller@math.rwth-aachen.de Wed Sep 1 12:41:48 1993 From: fceller@math.rwth-aachen.de "Frank Celler" Date: Wed, 1 Sep 93 12:41:48 +0200 Subject: Invariants Rejected message sent to gap-forum@samson.math.rwth-aachen.de ------------------------------------------------------------------------------- Date: Tue, 24 Aug 93 18:58:43 +0200 Comment: GAP Forum Originator: gap-forum@samson.math.rwth-aachen.de Errors-To: martin@samson Reply-To: Sender: gap-forum@samson Version: 5.31 -- Copyright (c) 1991, Anastasios Kotsikonas From: gap-forum@samson To: martin@samson Subject: Invariants Having published a paper last year in Communications in Algebra in which I computed the ring of invarinats of the simple group of order 660 in a 5 dimensional complex representation, I can point to some experience with such computations. I am not aware of any package for doing them and if there was, I would certainly want to use it. Based on my experience with the case I published and some others, I have some ideas on how to attack such problems and what kinds of software tools would be useful. Unfortunately, I am not a very good programmer and I would need to work with someone who is. Actually, I did work with someone on this a couple of years ago, but since I can't pay anyone to do this work and since programmers have a tendency to be busy for long periods of time with things that they do get paid to do, we didn't get very far. In addition, since the programmer in question has asserted some of his rights over the code, I do not feel free to distribute it, although I do tinker with it a little. I would actually prefer to bury the package as it exists at the moment and start from scratch to produce something that I can legally distribute freely to all who want to use it or who want to help develop it further. In addition, the programmer in question would have to know more mathematics than most programmers that I have seen know. One of the reasons for the painfully slow pace of development is that I could not take any knowledge for granted on the part of the programmer and had to explain even fairly routine computations in minute detail with examples that I had to work out by hand. As the computations got more complicated, this was more and more of a burden on me. It was also a burden that the programmer had no idea whether the code was doing what it was supposed to do at any level beyond conformity with the examples I had worked out by hand. I should add that I may not be as feeble a programmer as I thought. The evidence is that after my collaborator left the project, I went through the code he had written minutely and wrote detailed comentary on exactly what was happening at each moment. In this way, I found and corrected some of the outstanding bugs that caused performance to deviate from the description of the manual. In the process, I learned a certain amount about REDUCE symbolic mode programming conventions and the resources lying around in the REDUCE source code. But I still have a long way to go. GAP has attractive features that lend itself to such computations, namely the facility for working with groups, character tables, representations, etc. But it is weak in more traditional computer algebra such as polynomial manipulation. What is needed is an interface between GAP and other packages. If I am to be involved in a project like this, I would have to have access to any computer algebra packages that would be used in it. At the moment, the only computer algebra package I have access to that is close to being adequate for my needs is REDUCE. However, I am often frustrated by what I perceive as a "proprietary mentality" when I try to learn about its inner workings. I think it is something like the following at work: whereas I think of the programmer as someone who is helping me develop a package and educating me an all that I want to know in order to understand the final product (in particular, that I am in charge of it and need to know minutely what I am in charge of), the corporate mentality instead views me as the individual whose job it is to provide the programmer with algorithms in order to produce proprietary software which it ultimately doesn't matter whether I understand. In short, it is the point of view that I am part of a work group in which my "need to know" is sharply curtailed by my perceived function. Perhaps that is just what life is like when one works with commercial packages. For that reason, I would much prefer to work with a public domain algebra package. But a sufficiently powerful such package doesn't seem to exist yet. In fairness to the REDUCE people, I feel that they do a much better job of documenting the inner workings of their product than the MACSYMA people do. In both packages, there is a computer algebra package running over a LISP substrate. In MACSYMA that substrate and the interface to it are not documented while in REDUCE they are somewhat documented. Also, there are now books available that discuss symbolic mode programming in REDUCE. In addition, REDUCE comes with source code for REDUCE over the LISP substrate and one can pay an extra $500 to get the source code for the underlying LISP too. At any rate, I am more familiar with REDUCE and although I am willing to learn other pacakges, my learning curve would have to be taken into account, including the millions of naive and apparently frivolous questions that I would have about it the software and its inner workings, not to mention my insistence on having access to complete source code. If one is willing to put up with that and with my insistence of distributing the resulting code freely, read on. Actually, on second thought, I am probably making impossible demands by insisting that in addition to having the program I also have a certain working relationship to the programmers. So I will do the following: (1) When I have a moment, I will make a revised copy of the manual available by anonymous ftp. It will be typeset in TeX (it already is). Anyone is then free to read the descriptions of the functions contained therein and write their own functions that duplicate the described behavior. If you do, please let me know. If it is in a package I don't have access to, I would appreciate it if you find a way to provide me access to the necessary packages. (2) The manual only describes the work as far as we got with it. There are lot more functions that need to be included which are not described in the manual. I will be happy to collaborate with people in the development of those functions. I would like to add that since I have been unemployed for 7 years and have no income, I would not object if someone found a way to compensate me for my efforts. However, that is not a condition for the cooperation I have offered above. Allan Adler ara@altdorf.ai.mit.edu From fceller@math.rwth-aachen.de Wed Sep 1 12:53:26 1993 From: fceller@math.rwth-aachen.de "Frank Celler" Date: Wed, 1 Sep 93 12:53:26 +0200 Subject: GAP for the Mac Rejected message sent to gap-forum@samson.math.rwth-aachen.de follows. ------------------------------------------------------------------------------- Is a port of GAP 3.2 (or 3.1) to the Macintosh available at any FTP site? Thanks in advance. Bob Rowley ------------------------------------------------------------------- | Dept. of Computer Science | email: rowley@research.cs.orst.edu | | Oregon State University | phone: (503) 737-3371 | | Corvallis, OR 97331 USA | | ------------------------------------------------------------------- From neubuese@math.rwth-aachen.de Wed Sep 1 13:54:53 1993 From: neubuese@math.rwth-aachen.de "Joachim Neubueser" Date: Wed, 1 Sep 93 13:54:53 +0200 Subject: Re: Invariants Alan Adler's long letter to the GAP-forum of August 24 was unfortunately rejected by the mail server because the sender's address was not that of a member of the GAP-forum. It was then sent to the forum by Frank Celler. In order to avoid such delays please make sure that you write to the GAP-forum from the address with which you originally registered with the forum. When you change your address, please 'unsubscribe' and subscribe again with your new address. As to Alan Adler's letter: Yes, GAP does not provide all that a CA system like Reduce or Maple has. (For that reason we do ourselves use in particular Maple in our work and - to comment on Gregor Kemper's last remark - Maple or Reduce or some of the others are no no-no-words in the GAP-forum at all!) In particular while GAP got a (hopefully reasonable) polynomial arithmetic for univariate polynomials with version 3.2 it does not have (yet?!) an efficient multivariate polynomial arithmetic. (Since polynomials can be formed over arbitrary coefficient rings you can mimic a polynomial in two variables by a polynomial over the ring of univariate polynomials and so on, but that is not suitable for serious efficient computation). However in answer to some of the problems, Alan Adler addresses: There are systems for handling multivariate puolynomials very efficiently that are free of charge (although maybe formally not public domain in the same way that GAP is not public domain, namely in that a copyright is formulated that forbids commercialisation by third parties). I particularly mention MACAULAY and CoCoA. By the way, if you want rather comprehensive information on presently available CA systems (including CGT), you may want to consult the report "Computeralgebra in Deutschland" published by "Fachgruppe Computeralgebra of GI, DMV, and GAMM" which in spite of its understating title gives a fairly broad description on the present worldwide state of activities in algebraic and symbolic computation as well as good characterisations of 39 systems and packages for such computations together with information where and under which conditions to obtain them. Joachim Neubueser From neubuese@math.rwth-aachen.de Wed Sep 1 14:05:04 1993 From: neubuese@math.rwth-aachen.de "Joachim Neubueser" Date: Wed, 1 Sep 93 14:05:04 +0200 Subject: GAP 3.2 on the Mac (fwd) In answer to Bob Rowley's question I reprint the message of Prof. Mendelsohn of July 19. Joachim Neubueser (Moral: Read the forum regularly, it has good news even at vacation time!) ================================================================== Forwarded message: > From gap-forum@samson.math.rwth-aachen.de Mon Jul 19 22:26:28 1993 > Date: Mon, 19 Jul 93 22:25:17 +0200 > Message-Id: <9307192022.AA07404@ccu.umanitoba.ca> > Comment: GAP Forum > Originator: gap-forum@samson.math.rwth-aachen.de > Errors-To: martin@samson.math.rwth-aachen.de > Reply-To: > Sender: gap-forum@samson.math.rwth-aachen.de > Version: 5.31 -- Copyright (c) 1991, Anastasios Kotsikonas > From: mendel@ccu.umanitoba.ca (N. S. Mendelsohn) > To: Multiple recipients of list > Subject: GAP 3.2 on the Mac > > For those not contacted by H. Lakser, Gap 3.2 on the Mac may be obtained > from ftp anonymous at the address ccu.umanitoba.ca. It is archived as > /pub/mac/macgap32. > The directory of macgap32 contains the following four files for > downloading: MacGap32.sea.hqx; doc32.sea.hqx; lib32.sea.hqx; and > macgap32.info. These files are all self extracting. > > N. S. Mendelsohn > > > From neubuese@math.rwth-aachen.de Wed Sep 1 15:16:10 1993 From: neubuese@math.rwth-aachen.de "Joachim Neubueser" Date: Wed, 1 Sep 93 15:16:10 +0200 Subject: Re: call a GAP session? Norai Rocco asks for advice how best to call GAP functions from a FORTRAN program. Nobody has answered to that question in the forum yet. Since I try to make sure that each question to the forum is at least acknowledged in it, I have asked people in our department, but to my regret must report that nobody here seems to have experience with such a situation. I hope that still somebody else has some hint. I know that Leonard Soicher in GRAPE has been using FORTRAN programs together with GAP, but as I understand it in the other direction. Leonard is on holidays at present, perhaps he can comment when he is back. Joachim Neubueser From hulpke@abacus.concordia.ca Wed Sep 1 17:14:26 1993 From: hulpke@abacus.concordia.ca "Alexander Hulpke" Date: Wed, 1 Sep 93 17:14:26 +0200 Subject: Re: Polynomials in GAP Dear GAP-forum, In his message of 24th of August, Allan Adler writes: > GAP has attractive features that lend itself to such computations, > namely the facility for working with groups, character tables, > representations, etc. But it is weak in more traditional computer > algebra such as polynomial manipulation. Invariant computations (which were the initiator for this thread) surely will need polynomial arithmetic in several variables, which is - at the moment - not very feasible in GAP, as Joachim Neub"user already pointed out. However, if you need polynomial factorisations (over Q or an algebraic extension Q(alpha)) and basic polynomial routines like resultant etc.; there will be routines in a future version, that provide these functions. (They are basically finished and just have to be incorporated.) Probably this is of some use. Alexander Hulpke From greil@guug.de Fri Sep 3 11:50:09 1993 From: greil@guug.de "Toni Greil" Date: Fri, 3 Sep 93 11:50:09 +0200 Subject: GAP users in Muenchen ? Hello GAP users, are there any of you living/working in Muenchen (Germany) ? I'd like to get an impression, what you undertake with GAP. Toni Greil, Muenchen Tel. 089-268562 greil@guug.de From fceller@math.rwth-aachen.de Mon Sep 6 18:59:18 1993 From: fceller@math.rwth-aachen.de "Frank Celler" Date: Mon, 6 Sep 93 18:59:18 +0200 Subject: Missing Mail Dear Forum, due to a hard disk fault some mail to users on "math.rwth-aachen.de" or "tiffy.math.rwth-aachen.de" might have been lost. If you have sent any mail to the above machines and didn't get an answer, please resent the mail. The gap-forum itself is not affected. best wishes Frank Celler From fceller@math.rwth-aachen.de Mon Sep 6 18:50:19 1993 From: fceller@math.rwth-aachen.de "Frank Celler" Date: Mon, 6 Sep 93 18:50:19 +0200 Subject: Re: GAP users in Muenchen ? Dear Toni Greil, >> are there any of you living/working in Muenchen (Germany) ? >> I'd like to get an impression, what you undertake with GAP. sending a message containing a single line "review gap-forum" to "listserv@samson.math.rwth-aachen.de" will return a list of all subscribers of the gap-forum. Maybe you find someone from Muenchen in this list. best wishes Frank Celler From RUTH@brunb Fri Sep 10 15:32:22 1993 From: RUTH@brunb "Ruth Schwingel" Date: Fri, 10 Sep 93 15:32:22 +0200 Subject: solvable factor group Dear GAP-forum, I'm studying Alexander Wegner's dissertation entitled "Implementation eines Algorithmus zur Berechnung endlicher aufl"osbarer Faktorgruppen einer endlich pr"asentierten Gruppe im gruppentheoretischen Programmsystem GAP". But apparently, this algorithm is not implemented in GAP-3.2. I would like to know if there is such algorithm that computes finite solvable factor groups of a finitely presented group implemented in any other system. Ruth Schwingel From Johannes_Mueller@su.maus.de Mon Sep 13 18:48:34 1993 From: Johannes_Mueller@su.maus.de "Johannes Mueller" Date: Mon, 13 Sep 93 18:48:34 +0200 Subject: Running GAP under OS/2 -P54785@K Dear Forum, I have a problem with GAP running under OS/2: When I start a DOS-Session from the desktop and call GAP386.EXE, I get the following message: CPU must be in REAL mode (not V86 mode) to run this program without VCPI. (If you are using an EMS emulator, make sure that EMS isn't disabled) Now I don't know how to react, the Online-Help and the very small OS/2-Manual told me nothing, that led to a positive change. Who can tell me in details, how to configure OS/2 to run GAP there? Hope to get help, Johannes M|ller From fceller@math.rwth-aachen.de Tue Sep 14 12:42:18 1993 From: fceller@math.rwth-aachen.de "Frank Celler" Date: Tue, 14 Sep 93 12:42:18 +0200 Subject: Re: Running GAP under OS/2 Dear Johannes Mueller >> CPU must be in REAL mode (not V86 mode) to run this program >> without VCPI. (If you are using an EMS emulator, make sure that >> EMS isn't disabled) >> >> Now I don't know how to react, the Online-Help and the very small >> OS/2-Manual told me nothing, that led to a positive change. >> >> Who can tell me in details, how to configure OS/2 to run GAP there? Harald Boegeholz has done a port to OS/2. best wishes Frank Celler ----------------------------------------------------------------------------- Date: Thu, 4 Mar 93 17:23:31 +0100 Comment: GAP Forum Originator: gap-forum@samson.math.rwth-aachen.de Errors-To: martin@samson Reply-To: Sender: gap-forum@samson Version: 5.31 -- Copyright (c) 1991, Anastasios Kotsikonas From: hwblist@machnix.mathematik.uni-stuttgart.de (Harald Boegeholz) To: Multiple recipients of list Subject: A port of GAP to OS/2 2.0 and DOS Hello! I have started porting GAP 3.2 to emx, a free software development system for OS/2 2.0 and DOS. I mainly did this because I wanted an OS/2 version of GAP, but the resulting executable should run under DOS, too. This is my first try at porting GAP, and I have spent only very little time on it. But if you need an OS/2 version or if you are curious about another DOS version, give it a try! I put the executable on ftp.uni-stuttgart.de:/soft/os2/ProgLang/Languages/gapemx.zip Please read the enclosed documentation. Harald -- Harald Boegeholz |Home: hwb@texnix.stgt.sub.org (read daily) |University: hwb@machnix.mathematik.uni-stuttgart.de |please don't send large (>100k) mail to my home address. From gurican@alpha.dcs.mff.uniba.cs Thu Sep 16 13:43:19 1993 From: gurican@alpha.dcs.mff.uniba.cs "Jaroslav Gurican" Date: Thu, 16 Sep 93 13:43:19 +0200 Subject: GAP for VAX/VMS and Alpha AXP/OpenVMS Dear GAP Forum, I would like to notify you about the porting of gap to VAX/VMS and ALPHA AXP/OpenVMS enviroments. This port has been done by me and my colleague/student Pavol Mederly. As far as the ALPHA AXP/OpenVMS is concerned, we had only 1 day to make the port. The first thing we tried was translate VAX/VMS image (gap.exe) using VEST (translator from VAX/VMS images to ALPHA AXP/OpenVMS images). This is not good job, of course, but it could serve as as the last possibility (this is not what is ALPHA AXP meant for). The program was running well. (Combinat.tst gave approximately 11000 GAPStones - ALPHA AXP/OpenVMS running on 150 MHz, 64MBytes of memory, O.K.) Then we tried to compile it from the sources. The compiler and linker were succesful, there was one linker warning. I run combinat.tst, no problems, it gave at about 71000 GAPStones at the first run, 73500 at the second run. Even if it was not new world record, ALPHA seemd to be extermely fast gay.) After it I made all examples from About Group Chapter of GAP manual without problems. There were some (many) garbage collections during the session. Finally I tried some big arithmetic and I haven't find any mistake. So, I hope we produced fully functional version of GAP also for ALPHA AXP/OpenVMS environment (we used patch level 3). This version would be distributed on DEMO CD ROM by DEC (together with Announcement). You can get gap.exe (image file for ALPHA AXP/OpenVMS) from ftp.eunet.sk via anounymous ftp, it resides in the directory itr/techreports/incoming together with sysvax.c file. (This is the only file necessary to compile GAP on both VAX/VMS (we did it on 5.5) and Alpha AXP/OpenVMS and which is not included in usual distribution.) If you want to run GAP on VAX/Alpha, you have to unpack *.tar.Z or *.tar.zoo files in UNIX environement and send it to your VAX/Alpha (e.g. via ftp), BUT AS ASCII files. If you would use binary transfer, you will have problems with line-feets - you will get approximately 5 compilation errors. These errors can be fixed easily - simply remove LF sign from the ends of apropriate lines. (The gap.exe should be send as binary file.) I would like to mention one point which could explain differences (problems) with ALPHA AXP/OSF-1 environment: The cc on ALPHA AXP/OpenVMS knows only about 16 and 32 bits integers, there are no 64 bits integers. Jaroslav Gurican Dept. Of Algebra & Number Theory Mlynska dolina Bratislava SK- 842 15 P.S.: I would like to express my congratulations (and thanks, as well) for such clean design of GAP. P.S.1: GAPStones on some VMS machines --------------------------------------------------------------------------- First column = first run of GAPu Second column = another run of GAPu first raw = first run of the test second raw = second run of the test Alpha AXP 3000-500S (118 SPECmark89 (DECchip 21064/150 MHz)) 71000 (less than 10 seconds) 71000 73500 72992 MicroVAX 3100/90 (24 VUP - Vax Unit Performance) 25510 (28s) 26099 (27s) 26505 (27s) 27100 (26s) VAXstation 4000/60 (12 VUP) 12660 (58s) 12697 (57s) 13069 (55s) 13118 (55s) VAX 6000/420 (2 proccesors 7 VUP each) 9350 (80s) 9422 (77s) 9727 (76s) 9652 (76s) VAX 4000/200 (4.8 VUP) 5006 (140s) 4976 (143s) 5128 (140s) 5173 (137s) From neubuese@math.rwth-aachen.de Thu Sep 16 19:44:57 1993 From: neubuese@math.rwth-aachen.de "Joachim Neubueser" Date: Thu, 16 Sep 93 19:44:57 +0200 Subject: Re: solvable factor group In a letter of last Friday Ruth Schwingel asked: > I'm studying Alexander Wegner's dissertation entitled > "Implementation eines Algorithmus zur Berechnung endlicher > aufl"osbarer Faktorgruppen einer endlich pr"asentierten > Gruppe im gruppentheoretischen Programmsystem GAP". > > But apparently, this algorithm is not implemented in GAP-3.2. > > I would like to know if there is such algorithm that computes > finite solvable factor groups of a finitely presented group > implemented in any other system. Alexander Wegner's Soluble Quotient program, following W. Plesken's proposal, was implemented for GAP 2.4, it did not get included into this and the later versions so far, however right now work is going on to modify it slightly and to make it available with the then latest version of GAP. Prior to full integration and soon we hope to provide a separate file with this program for extensive testing. Another Soluble Quotient using completely different methods and involving Steve Linton's Module Enumerator has been written by Alice Niemeyer in Canberra. She plans to make it available both as a standalone and as a share library with GAP also in the near future. No dates promised for both, but near future is hoped to mean what it says in both cases!! Availability of both will be announced in the GAP-forum. Charles Sims has written an experimental version of a Metabelian Quotient in connection with his paper on implementing the Baumslag- Cannonito-Miller algorithm, however this is not publicly available. I do not know of a Soluble Quotient in any other system, sorry. Joachim Neubueser From hwblist@machnix.mathematik.uni-stuttgart.de Thu Sep 23 16:34:53 1993 From: hwblist@machnix.mathematik.uni-stuttgart.de "Harald Boegeholz" Date: Thu, 23 Sep 93 16:34:53 +0200 Subject: new port of GAP for OS/2 2.x and DOS Hello! I have completed a new version of my port of GAP 3.2 to OS/2 2.x and DOS. The executable is now at patchlevel 3, and I made some more minor emx-specific improvements. You can download it from ftp.uni-stuttgart.de:/soft/os2/ProgLang/Languages/gapemx.zip This file contains an executable of GAP compiled for OS/2 2.x and DOS as well as all those files of the GAP library that were updated by the three patches to GAP. That way you can conveniently upgrade to patchlevel 3 without using PATCH and the diffs. For running GAP under OS/2 2.x, you also need the emx runtime. It is available from ftp.uni-stuttgart.de:/soft/os2/emx-0.8g/emxrt.zip. Please refer to the file gap/emx/README in gapemx.zip for more information. New features of this port: * works correctly under GNU Emacs under OS/2 (identifier completion etc.) * signal handling (Ctrl-Break) now works under DOS, too There is also a hypertext version of the GAP documentation for OS/2. It can be found at ftp.uni-stuttgart.de:/soft/os2/ProgLang/Languages/gapinf.zip Since I have had practically no feedback from people using GAP under OS/2, I would like to hear from anybody who is using this port. I would also like to hear from people who have tested this version of GAP under DOS. I don't use DOS myself, but I have briefly tested this port under DOS and it seems to work pretty well. It even seems to be a little faster than gap386.exe. Enjoy! Harald Boegeholz -- Harald Boegeholz | hwb@mathematik.uni-stuttgart.de | please note that I will discontinue using my home | address hwb@texnix... in the near future. From L.H.Soicher@qmw.ac.uk Thu Sep 30 12:26:46 1993 From: L.H.Soicher@qmw.ac.uk "Leonard Soicher" Date: Thu, 30 Sep 93 12:26:46 +0100 Subject: GRAPE reference The reference to quote if you use GRAPE has now appeared in print. It is: L.H. Soicher, GRAPE: a system for computing with graphs and groups, in "Groups and Computation" (L. Finkelstein and W.M. Kantor, eds.), DIMACS Series in Discrete Mathematics and Theoretical Computer Science 11, American Mathematical Society, 1993, pp. 287-291.