Newsgroups: comp.lang.lisp
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!europa.eng.gtefsd.com!howland.reston.ans.net!EU.net!uknet!festival!edcogsci!jeff
From: jeff@aiai.ed.ac.uk (Jeff Dalton)
Subject: Re: Why do people like C? (Was: Comparison: Beta - Lisp)
Message-ID: <Cwst88.CFD@cogsci.ed.ac.uk>
Sender: usenet@cogsci.ed.ac.uk (C News Software)
Nntp-Posting-Host: bute.aiai.ed.ac.uk
Organization: AIAI, University of Edinburgh, Scotland
References: <1994Sep19.221325.3567@cabell.vcu.edu> <uPkXsc1w165w@sytex.com>
Date: Tue, 27 Sep 1994 17:14:31 GMT
Lines: 48

In article <uPkXsc1w165w@sytex.com> smcl@sytex.com (Scott McLoughlin) writes:
>csc1alf@cabell.vcu.edu (Adrian L. Flanagan) writes:
>> 
>> I must strenously disagree with the original poster.  "Blazing
>> speed,space,etc." are that critical.  Particularly in the PC DOS
>> world with its 640K restriction, program size and efficiency of
>> compiled code made a tremendous market difference in acceptance of
>> early commercial programs.  Programmers writing in C had a large
>> advantage over programmers using the early Lisp systems [...]

If speed is so crucial, why are so many people willing to use Macs
with the cache off (3,4, 10 times slower)?

There's a lot of evidence that people are often willing to
sacrifice speed for other things.  (Not always, of course.)

In any case, what early Lisp systems were competing with C?

>> The (relative) failure of Lisp has everything to do with Lisp
>> vendors' failure to understand (even now) the needs of their
>> marketplace.  Call it Ivory Tower Syndrome.

>       1. Right. DOS+640K definitely gives advantage to C/Pascal/Asm.
>Esp with TSR's and other weirdnesses. My question refered to the folks,
>though, who are using VB and VC and BCW4.0 _now_ - big environments and
>big images. I like the general historical picture, though.

I can't really say, because it's lacking all detail.

>        2. When I say "no blazing speed", I mean -- "Not hard realtime".
>I'm assuming only a 2X to 3X speed avantage of native C over compiled
>Lisp. Any body ever benchmark Allegro vs. Borland C++ on say TAK ?

KCL is as fast as C for TAK (on Sun3s, which reflects the last time I
tried it).  But TAK is a very limited benchmark.

>        3. "Ivory Tower Syndrom" -- YES! Now I think we're getting to
>the heart of the matter ;-)

How so?

I could understand someone saying Lisp vendors went after market A
(AI researchers, perhaps?) when they should have gone after market B
(commercial DOS users?).  But the two of you seem to be saying they
misunderstood the market they went after.  Perhaps you think they
went for B and blew it when what they really did was go for A.

-- jd
