home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math.symbolic
- Path: sparky!uunet!usc!zaphod.mps.ohio-state.edu!uwm.edu!ux1.cso.uiuc.edu!news.cso.uiuc.edu!mm-mac17.mse.uiuc.edu!gaylord
- From: Richard J. Gaylord <gaylord@ux1.cso.uiuc.edu>
- Subject: Re: Serious Programming
- References: <1992Nov17.232239.9748@access.usask.ca>
- Message-ID: <BxwrMF.Ctw@news.cso.uiuc.edu>
- X-Xxdate: Wed, 18 Nov 92 04:55:54 GMT
- Sender: usenet@news.cso.uiuc.edu (Net Noise owner)
- X-Useragent: Nuntius v1.1.1d12
- Organization: University of Illinois
- Date: Wed, 18 Nov 1992 10:54:13 GMT
- X-Xxmessage-Id: <A72F7D5AD4029B11@mm-mac17.mse.uiuc.edu>
- Lines: 49
-
- first of all, my apologies for the intemperate use of the phrase "
- serious programming".
- that was just a response to the initial ridiculous (in my opinion)
- comment that Maple is somewhat more powerful [than Mathematica] in just
- about every way. Mathematica's only advantages are in pattern matching. "
-
- obviously Maple is a serious cas [ and release 2 looks like its going to
- be really nice] and since most of it is written in the Maple language
- the Maple language must be serious, indeed.
- ==================
- let's bring some focus to this thread.
-
- i am comparing cas languages ONLY in terms of their ability to do
- functional programming . I am excluding graphics, algebraic manipulation,
- numerics, etc. i understand that there are many other issues involved in
- comparing cas languages but this is the issue i am competent to discuss.
-
- i maintain that mathematica does the best job of bringing the functional
- style of programming into scientific computing. (no one is using ML or
- Scheme or Haskell or FP to do scientific computing ).
-
- here are the good 'functional' things i find in the Mathematica
- programming language
-
- (1) mathematica's use of fully spelled out english names for its
- built-in functions and its use ofa mathematical equational style is
- superior to the syntax of other languages (with the exception of APL
- whose runic symbols and one way evaluation scheme are especially
- appealing ).
-
- (2) mathematica definitely has one of the most sophisticated pattern
- matchers around. it should not be dismissed as unimportant. it is an
- incredibly useful programming tool.
-
- (3) the use of a rewrite rule system, while certainly problemmatic, is
- very useful.
-
- (4) while the built-in list manipulation and higher-order functions are
- not a part of functional programming per se, they are extremely useful to
- have save me considerable time in writing programs
-
- all of these 4 specific features are part of the 'functional' style of
- programming [ i know that what is a functional feature is often argued
- but these features above are common to most functional languages eg.,
- ML)] and i don't think that any other language (cas or otherwise) comes
- close to mathematica in providing these specific functional programming
- tools for scientific computing.
-
- am i wrong in saying this?
-