Newsgroups: comp.lang.prolog
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!uhog.mit.edu!bloom-beacon.mit.edu!gatech!howland.reston.ans.net!ix.netcom.com!netcom.com!dcf
From: dcf@netcom.com (Don Ferguson)
Subject: Re: Why hasn't Prolog Taken over the World?
Message-ID: <dcfCznotK.15n@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
 
  Richard Pitre (pitre@n5160d.nrl.navy.mil) wrote: 
X-Newsreader: TIN [version 1.2 PL1]
References: <dcfCzI6wB.9pH@netcom.com> <3aqbss$r6k@ra.nrl.navy.mil>
Date: Tue, 22 Nov 1994 06:32:08 GMT
Lines: 38

: What will save Prolog-like programming is the cost of C++-like programming?
: By cost I mean development and maintenance $$$ and I mean the cost in bug
: related deaths. Once businesses are given a believable cost comparison and
: government mandates to stop killing people with expensive junk then the 
: game will be over. 

The game will be over for C++, but that does not mean that Prolog, or
a Prolog-like language will take over.  Smalltalk stands a better chance.

One thing Prolog has going for it is that the hardware environments have
finally caught up.  With Pentiums and PowerPCs, the desktop hardware finally
has enough speed for Prolog to be practical for commercial applications.
And with GUI's chewing up gobs of memory, Prolog's footprint is not even
a major issue.

Quintus has demonstrated that Prolog can be used to develop and deliver
sophisticated client/server systems on standard hardware, and compete
head-to-head with systems written in C++.  Quintus WorkPro, which serves
the Customer Support/Help Desk markets, is a successful product in a highly
competitive market.  And Prolog enabled Quintus to develop this product
on a shoestring budget, compared to the well-funded competitors.  Yet
Quintus is considering re-writing it all in C++.  Why?  In part because
Quintus' competitors use the fact that it is written in Prolog to imply
that the technology is wierd, proprietary, and maybe flaky.

So I ask, if success is not enough to justify Prolog, what is?  There are
tidal forces that underlie technological shifts, and those forces are
more powerful than logic and reason.  

Perhaps I'm jaded.  I spent nearly 10 years at Quintus fighting the good
fight, and was the original author of Quintus WorkPro.  To see that
technology transform Quintus such that they would reject Prolog for a 
'mainstream' technology like C++ must give one pause.  If Prolog is to have
a savior, it won't be Quintus.


	Don Ferguson

