home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.arch:10954 comp.lang.misc:3808
- Path: sparky!uunet!olivea!charnel!sifon!thunder.mcrcim.mcgill.edu!snorkelwacker.mit.edu!ai-lab!life.ai.mit.edu!tmb
- From: tmb@arolla.idiap.ch (Thomas M. Breuel)
- Newsgroups: comp.arch,comp.lang.misc
- Subject: Re: Hardware Support for Numeric Algorithms
- Message-ID: <TMB.92Nov20131310@arolla.idiap.ch>
- Date: 20 Nov 92 21:13:10 GMT
- References: <1992Nov18.005857.2241@oracle.us.oracle.com> <1efg1gINNm6c@network.ucsd.edu>
- Reply-To: tmb@idiap.ch
- Organization: IDIAP (Institut Dalle Molle d'Intelligence Artificielle
- Perceptive)
- Lines: 39
- NNTP-Posting-Host: arolla.idiap.ch
- In-reply-to: mbk@lyapunov.ucsd.edu's message of 19 Nov 1992 07:34:08 GMT
-
- In article <1efg1gINNm6c@network.ucsd.edu> mbk@lyapunov.ucsd.edu (Matt Kennel) writes:
-
- : Feel free to design a computer language for mathematics. Don't pretend
- : that other computer languages (least of all C) exist to serve you.
-
- Well, when he expresses some notion of what he wants to do, without specifying
- it the way computer scientists want to hear, he gets the third degree. The
- point is that the computer scientists have the abilities and facilities to
- do what he wants (that he probably doesn't), but too many seem profoundly
- unsympathetic to even his most basic motivations and they go tell him to jerk
- off.
-
- Software companies will produce programming languages that the market
- wants. Computer scientists will produce programming languages
- incorporating features that they think the market will want 10 or 20
- years from now. Rubin's ideal language isn't being produced partly
- because he is asking for conflicting constraints to be satisfied, and
- partly because people like him have evidently failed to convince
- others that the features they want are useful and cost-effective.
-
- As an example, consider a Connection Machine---it's hardware is so fundamentally
- different, and so to even have a chance of writing intelligent programs, it's
- essential to be able to express these fundamentally new methods in software.
- (And yes, they do it.) Herman's point is that more computers than you think
- may need this, especially as supercomputer-research-type stuff starts coming
- down to earth. We don't want 87 completely different dialects of C, or tricky
- obscure code that compiles into something simple. We accept the necessity
- of hardware dependence, but we want it tamed, rather than excommunicated.
-
- That's the goal of everybody. Now, if you have a solution to this
- problem, by all means, let us know about it. So far, all the proposals
- I have seen on this newsgroup came down to "I want to be able to mess
- up the lexer/parser in arbitrary ways by defining my own syntax"
- (which has nothing to do with taking advantage of hardware), and
- "wouldn't it be nice if I could express the following machine-specific
- feature so that it runs fast on every machine" (which is obviously a
- contradiction).
-
- Thomas.
-