Path: utzoo!attcan!uunet!lll-winken!ames!ucsd!rutgers!apple!bloom-beacon!bu-cs!mirror!rayssd!raybed2!linus!mbunix!bwk
From: bwk@mbunix.mitre.org (Barry W. Kort)
Newsgroups: comp.lang.prolog
Subject: Re: Prolog and Smalltalk
Summary: Try Digitalk Smalltalk V/286
Keywords: Computer Animation, Simulation Modeling, Expert Systems, LANs
Message-ID: <43450@linus.UUCP>
Date: 7 Jan 89 16:13:51 GMT
References: <15242@mimsy.UUCP>
Sender: news@linus.UUCP
Reply-To: bwk@mbunix (Barry Kort)
Organization: The MITRE Corporation, Bedford, Mass.
Lines: 26

In article <15242@mimsy.UUCP> chomicki@brillig.umd.edu writes:

 > I'd appreciate very much any information about the existence and
 > availability of Prolog systems written in Smalltalk or integrated
 > in some other way with a Smalltalk programming environment.

Jan, we have been using Digitalk's Smalltalk V/286 for the AT class
PC.  It has an integral Prolog which closely resembles Clocksin
and Mellish Prolog.  The code seems to work OK, but the documentation
is a bit thin.  One problem is that the Smalltalk debugger doesn't
know how to handle Prolog methods.  The notation for Prolog lists
conflicts with the notation for Smalltalk Block Contexts (both use
square brackets), and this causes some confusion.  Prolog code must
be compiled in a special browser (but can be viewed in the regular
Smalltalk browser).

Despite these annoyances (which we expect will vanish over time),
we are cautiously optimistic that we can successfuly integrate
a Prolog diagnostic expert system into our Smalltalk model of
a local area network.  Most of the work is being done by a senior
at UConn, who is bravely wading into uncharted territory here.
Her rate of progress is encouraging, and the work is both
intellectually stimulating and attention grabbing.

--Barry Kort
MITRE Network Center