Gentoo Archives: gentoo-science

From: C Y <smustudent1@×××××.com>
To: gentoo-science@l.g.o
Subject: Re: [gentoo-science] Re: Scientific herd leadership
Date: Mon, 22 Aug 2005 16:29:21
Message-Id: 20050822162815.48625.qmail@web31713.mail.mud.yahoo.com
In Reply to: Re: [gentoo-science] Re: Scientific herd leadership by "M. Edward (Ed) Borasky"
1 --- "M. Edward (Ed) Borasky" <znmeb@×××××××.net> wrote:
2
3 > I used to be on the CMUCL and SBCL mailing lists because CMUCL is
4 > tge best overall Lisp for my purposes and SBCL is the most actively
5 > developed. In addition to Maxima, I use Lisp for algorithmic
6 > composition, and right now CMUCL is the only one that fully supports
7 > Rick Taube's Common Music.
8
9 Cool! I didn't know Common Music was in active use. I should look at
10 that more carefully.
11
12 > I guess the fact that there are four "competing" open source Lisps
13 > is a "problem", just as the fact that there are two competing open
14 > source "emacs" packages is a "problem". Still, I think the various
15 > compatibilities/standards/etc. between Maxima and the four
16 > currently-existing Lisps are best suited by the current mechanism ...
17
18 I guess it's a problem, but I like that there are so many different
19 environments experimenting with different things. The biggest fly in
20 the ointment right now is ANSI compatibility (*cough*gcl*cough*) - once
21 an ANSI platform can be depended on on all 4 then people can start
22 stretching the various features available on each, and maybe evolve
23 what will become ANSI Common Lisp 2.0 :-) Maxima being able to run on
24 all of them gives us a) a check that the underlying lisp isn't doing
25 unexpected things to results b) assurance that if one lisp dies we have
26 no trouble continuing on c) a good excuse to install lots of lisps ;-)
27 and d) an extra way to shake bugs out of our code (not that we need any
28 help right now, granted...)
29
30 > they're all available in Portage in at least one "stable" form,
31 > however ancient that might be, and available in "testing"
32 > or "unstable" for us amateur beta testers. I don't have a problem
33 > joining mailing lists for packages I use and filing bug reports
34 > upstream -- ask the Texmacs, R, or Common Music and SBCL folks
35 > about that. :)
36
37 My main complaint is that, in many cases, the newer release of the
38 lisp/CAS/scientific program in general is likely to contain many fixes,
39 and in fact be more desirable to use generally than the older version.
40 In fact, keeping the older, incorrect version around is an active
41 disservice to people using it thinking it's correct. Perhaps in the
42 case of the various lisps I can see waiting a bit (except for gcl ;-)
43 but in the case of Maxima, when we DO put out a new release there tend
44 to be fixes for mathematically incorrect behavior and other things that
45 make no sense to wait on. It is improbable that Maxima will regress
46 between versions - there's too much room for improvement. I would
47 expect the same would hold for Axiom. Even if there WERE regression,
48 it might not be caught for quite some time, since exercising the
49 capabilities of a CAS is not trivial in and of itself.
50
51 > Gentoo is about choice, right? If the choice, however, must be
52 > where to allocate finite (or in some cases zero) volunteer
53 > developer/maintainer time, I'd cast my vote for CMUCL as the Lisp
54 > of choice, at least until SBCL hits 1.0 and gets the callback thing
55 > worked out for Common Music.
56
57 I would tend to agree, except I don't know how CMUCL does on non-x86
58 platforms. Clisp at least tends to run on ANYTHING, despite it's not
59 being the performance choice. IIRC that's why Clisp was originally set
60 as the no-choice-given default for Maxima, although that may have
61 changed since.
62
63 > GCL is a tad faster on some benchmarks, and CLISP has "readline"
64 > built in.
65
66 GCL can use readline too, at this point - not sure if it's in the
67 ebuild by default.
68
69 Oh, I recommend rlwrap for cmucl, by the way - assuming one is being
70 foolish like me and not using SLIME ;-). Must learn SLIME, must learn
71 SLIME...
72
73 > I'm not sure there's a "stable" Axiom in the minds of the upstream
74 > people just yet.
75
76 Hmm, good question.
77
78 > I had Axiom on my Debian systems when I was running Debian but
79 > never had a chance to play with it. It takes several hours to
80 > build from source on a fast machine.
81
82 Yep - reminds me of acl2's build, or to a lesser extent brl-cad's :-).
83 All the good stuff takes forever to build ;-).
84
85 > I've forgotten what the core symbol crunching engine is -- IIRC it
86 > carries an older version of one of the four open-source Lisps.
87
88 GCL, although of late they have been staying relatively current.
89
90 > In any event, I'll join the chorus of "let's have as much support
91 > for Axiom as we can."
92
93 :-)
94
95 [snip - interesting about Octave vs. R - maybe the Octave interface
96 should be bolted onto the R core!]
97
98 > I don't know what brl-cad and acl2 are/do, so I can't help you there.
99
100 Acl2 is... well, here's their own description:
101
102 "ACL2 is both a programming language in which you can model computer
103 systems and a tool to help you prove properties of those models."
104
105 http://www.cs.utexas.edu/users/moore/acl2/
106
107 Here's one of its more famous uses:
108
109 "Remember the Intel FDIV bug? The first Pentium [trademark, Intel, Inc]
110 could not divide floating-point numbers correctly and it reportedly
111 cost Intel $500 million to fix. At the time that was happening, we used
112 ACL2 to verify that the floating point division microcode on AMD's
113 competing microprocessor, the AMD-K5, was correct. AMD used ACL2 to
114 verify the elementary floating-point operations for the recently
115 released Athlon."
116
117 This and HOL4 would be my first two candidates for proof systems to
118 create ebuilds for. I took the bugzilla one for 2.8 and tried it on
119 2.9 (the current version) - it seemed to build successfully, although
120 I've not done any fancy testing with it yet.
121
122 BRL-CAD (http://brlcad.org/)
123 BRL stands for Ballistic Research Laboratory - this is a powerful
124 constructive solid geometry (CSG) modeling package used by the army for
125 decades to do real world simulation. It was recently released as open
126 source software, and it has the potential to plug a major gap in the
127 lineup of open source software solutions. Unfortunately, it's history
128 is so long that it predates Linux and the library naming conventions
129 that go with it, and uses tweaked versions of other standard libraries.
130 The result is that an improper install will wreck havock on an
131 unsuspecting Gentoo box :-/. I ultimately had to reinstall. What it
132 needs is to have all relevant libraries in their own directory (say
133 /usr/lib/brlcad/) and go from there, but I don't know how to make an
134 ebuild that a) builds it with those targets and b) updates the system
135 correctly so brl-cad doesn't try to use the wrong libraries (e.g. in
136 /usr/lib) or not know about the ones in /usr/lib/brlcad.
137
138 I'm a bit like you - I love to tinker. The odds I'll do any serious
139 CAD work with BRL-CAD are remote, but I like the idea of it being
140 available in case I DO hit something that makes it necessary/worthwhile
141 to put in the time. So if you've got the itch, I recommend BRL-CAD
142 :-). (incidently, the main interaction environment is not called
143 brlcad - it's something I forget at the moment. But if typing brlcad
144 doesn't do anything it's not unexpected.)
145
146 > Where *I* would focus "gentoo-science" -- indeed, all of Gentoo --
147 > is on packages that are vibrant, alive and even chaotic upstream.
148 > Right now, that's R, Maxima, SBCL, Axiom, Ruby/Rails, Xen, TeXmacs,
149 > NS/NAM, etc.
150 >
151 > ... the stuff that's on *my* hard drive. :)
152
153 You definitely want to check out BRL-CAD - it's still actively used by
154 quite a few people, IIRC. The project lead is Sean Morrison - very
155 helpful guy. Even if you don't do CAD, worth a look as a potential
156 assist to open source in general :-). I always make sure I've got
157 Blender installed for the same reasons (or non-reasons :-P) even though
158 you can fit my artistic talent in the head of a pin :-/.
159
160 Cheers,
161 CY
162
163
164
165 ____________________________________________________
166 Start your day with Yahoo! - make it your home page
167 http://www.yahoo.com/r/hs
168
169 --
170 gentoo-science@g.o mailing list