Path: utzoo!utgpu!attcan!uunet!husc6!mailrus!purdue!decwrl!labrea!csli!johnson
From: johnson@csli.STANFORD.EDU (Mark Johnson)
Newsgroups: comp.lang.prolog
Subject: Re: MAC Prolog
Message-ID: <6383@csli.STANFORD.EDU>
Date: 10 Nov 88 14:57:24 GMT
References: <1561@bingvaxu.cc.binghamton.edu> <200@internal.Apple.COM> <13330@oberon.USC.EDU>
Reply-To: johnson@csli.UUCP (Mark Johnson)
Distribution: na
Organization: Center for the Study of Language and Information, Stanford U.
Lines: 26

I don't want to be too negative, but in my opinion AAIS prolog
incorporates a number of serious design flaws, the most important
of which are the "Lisp-style CONS cell lists" (p. 50 of User's Manual)
and the ensuing nonstandard behaviour of predicates like "clause".

ALS Prolog for the Mac (it appears to be a port of the PC version) is
a little better in this regard, and their implementation is
substantially faster than AAIS as well.  However their user interface
leaves something to be desired (eg. to load a predicate definition
from a window you must save the file and load the file).

Moreover, both implementations are somewhat buggy (ALS in particular),
which can be quite annoying - you never know if the problem is in your
program or the Prolog implementation.

What we really need is a good, solid Prolog implementation of the
standard of Allegro CommonLisp for the Mac.  I would strongly advise
anyone implementing Prolog for the Mac to look long and hard at
Allegro CL, especially their interface to the Mac OS. Allegro CL shows
that you can implement a solid product for the Mac and sell it at a
reasonable price and still make a profit  (are you listening, QP?)

Mark Johnson


Both of these implem