home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math.symbolic
- Path: sparky!uunet!utcsri!newsflash.concordia.ca!mizar.cc.umanitoba.ca!access.usask.ca!news
- From: jake@skatter.usask.ca
- Subject: Re: Serious Programming (was: Re: MAPLE resources reccomendation)
- Message-ID: <1992Nov17.232239.9748@access.usask.ca>
- Sender: news@access.usask.ca (USENET News System)
- Nntp-Posting-Host: skatter.usask.ca
- Organization: University of Saskatchewan
- References: <1eb8j4INNld2@agate.berkeley.edu>
- Date: Tue, 17 Nov 1992 23:22:39 GMT
- Lines: 77
-
- From article <1eb8j4INNld2@agate.berkeley.edu>, by fateman@peoplesparc.Berkeley.EDU (Richard Fateman):
- > In article <BxuxC6.Dxz@news.cso.uiuc.edu> Richard J. Gaylord <gaylord@ux1.cso.uiuc.edu> writes:
- >
- >>
- >>the basic question iswhat do you want out of your cas? if you are going
- >>to do serious programming, mathematica is totally in front of the
- >>others.
- >
- > I think that if you are doing serious programming, the mandatory use of
- > Mathematica's built-in interpreted user language is a serious handicap
- > for many purposes. It's ok for fiddling around, but if it were really
- > good, then there wouldn't be 400,000 (or more) lines of C in the
- > system. Most add-ons written in the Mathematica language have
- > serious efficiency problems, and some are outright crocks that break
- > when stressed.
- >
- > While I am not a fan of the Maple syntax, it is certainly
- > possible to write serious programs in it, as demonstrated by the Maple
- > group's having written most of Maple in it. It has been shown on this
- > newsgroup that Maple has similar functional-programming style
- > features, and is fast.
- >
-
- Having recently discovered (?) and read with some interest Mr. Fateman's
- review of the Mathematica system, and now this particular comment, I begin
- to wonder if any system, anywhere in whatever stage of development, can
- be said to embody a syntax of which Mr. Fateman would be a "fan".
- This is not intended to abuse Mr. Fateman, I rather am seriously interested
- in these issues and would like to see him state exactly what, in his opinion,
- the ideal design trade-offs in a CAS are.
-
- It is quite possible that he has done so in a forum of which I am ignorant,
- and if so, I would appreciate knowing. How one goes about constructing such
- systems is a topic of my interest every since I first came into contact
- with these systems.
-
- >
- > A system built on top of Common Lisp (like Macsyma, Reduce, Weyl,
- > Mock-MMa) can in principle provide all those (very serious) features.
- > Some of them, like the Common Lisp Object System, are not available
- > in Mathematica, and, some might argue, represent a serious deficiency.
- > The lack of any appropriate structure for maintaining
- > assumptions is another major deficiency in Mathematica's design.
- >
-
- This is a good point. (IMHO) In an article in the Mathematica Journal, it was
- (reportedly) said by Mr. Wolfram that the problem is in organizing and
- propagating through all the various bits of information. However, it
- seems to this poster that all of the other systems seem to have been
- able to deal effectively, with perhaps varying degrees of success, with this
- problem.
-
- > Most implementations of Common Lisp allow for links to other languages
- > such as calls to Fortran or C, if that is "serious" also.
- >
- > So for serious programming, I'd put Mathematica near the end of the list.
- > For non-serious programming, it is much more appealing, especially
- > if you can exercise some restraint in how much of it you use, and you
- > are able to avoid questions of scope, evaluation, and limited data
- > structure choices.
- >
- >
- >
- >
- >
- > --
- > Richard J. Fateman
- > fateman@cs.berkeley.edu 510 642-1879
-
-
- jake
-
-
- --
- Jason C. Breckenridge jake@skatter.usask.ca
-
- Sometimes the world is a mighty fine place to be dead to.
-