Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!munnari!moncskermit!basser!elecvax!cheops!claude
From: claude@cheops.OZ (Claude Sammut)
Newsgroups: net.lang.prolog
Subject: Re: looking for Prolog
Message-ID: <496@cheops.OZ>
Date: Wed, 11-Jun-86 00:41:15 EDT
Article-I.D.: cheops.496
Posted: Wed Jun 11 00:41:15 1986
Date-Received: Sat, 14-Jun-86 04:49:24 EDT
References: <546@glasgow.glasgow.UUCP> <70900001@uicsl> <177@bubba.UUCP> <98@philtis.UUCP>
Organization: Computer Science, Uni of NSW, Aust.
Lines: 21

After Frank de Bruyn's article it looks like I'll have to step in with
the authentic story on UNSW Prolog. As far as I can tell UNSW Prolog
predates C-Prolog. The first version was completed in 1979.  (Please
let me know if I'm wrong. I know that C-Prolog was based on an earlier
interpreter for an ICL machine).  The entire system was written without
knowledge of how any other Prolog was implemented. In fact I wrote it
to teach myself how Prolog works because there was so little
information available here in Australia at that time. This independent
development also explains why there are differences in the system's
behaviour when compared with "Edinburgh" Prologs. I had to make design
decisions without the benefit of anyone's advice. As it turned out
almost all the decisions comformed with the Edinburgh implementations,
however, there remain some differences. These are mostly only minor
datails or matters of taste. The most important differences are that
cut in UNSW Prolog, when used inside an "or" or "->", does not cut the
entire clause. Also, my bagof is really only findall.

In the interests of historical accuracy,

Claude Sammut
University of New South Wales