🚀 go-pugleaf

liggio@cunixf.cc.columbia.edu (Vincent J. Liggio) writes:

>Is there a source out there where the original 'evolving' uncompiled prolog
What is 'evolving' uncompiled prolog ?
>can be gotten from, or is Turbo (which I have) and other compiled versions
>the only ones available?
Since when does there exist such a thing as Turbo "prolog" other than in
the minds of those who try to sell it.

I am not sure but I think that any real prolog will fit your
requirements.

>Vince

In article <1992Oct7.172542.21894@news.columbia.edu>, liggio@cunixf.cc.columbia.edu (Vincent J. Liggio) writes:
> Is there a source out there where the original 'evolving' uncompiled prolog
> can be gotten from, or is Turbo (which I have) and other compiled versions
> the only ones available?

I do not know what "evolving" or "uncompiled" means.

The original Prolog is Marseilles Prolog; I believe that existed in several
languages but the one I've seen documentation for was coded in reasonably
portable Fortran.  That was an interpreter.  Someone may still have a copy
of the sources on a backup tape somewhere.  There were two other Prolog
implementations that have some claim to "originality" and still have some
influence:  Waterloo Prolog started out as someone's Masters project; it
was developed from published descriptions of the _ideas_ of Prolog, but I
heard that the author had no access to the sources of any Prolog.  Aspects
of that design live on in IBM Prolog.  Sources for the original Waterloo
Prolog may still be on a backup tape somewhere.  UNSW Prolog (described in
the Australian Computer Journal) was again developed from published _ideas_
without access to code.  That too was an interpreter.

ALS Prolog is an example of a Prolog system where clauses can be asserted,
listed, retracted, and so on, and yet they are always compiled.  Prolog-X
(Byrd & Clocksin) was (is?) another.

PDC (Turbo) Prolog is a logic programming language offering reasonable
performance on a PC (though there are conventional Prologs which are said
to outperform it) and a development environment which some people like
very much.  However, it has a rather inexpressive type system (compared,
say, with the DEC-10 Prolog type checker) and there are so many differences
between it and the family of programs advertised as "Edinburgh-compatible"
that it is better for practical purposes to think of it as a different
language.

If you have an 80386, SWI Prolog may be of use to you (free).

In article <15039@goanna.cs.rmit.oz.au>, ok@goanna.cs.rmit.oz.au (Richard A. O'Keefe) writes:

>                                             There were two other Prolog
> implementations that have some claim to "originality" and still have some
> influence:  Waterloo Prolog started out as someone's Masters project; it
> was developed from published descriptions of the _ideas_ of Prolog, but I
> heard that the author had no access to the sources of any Prolog.  Aspects
> of that design live on in IBM Prolog.  

That "someone" was Grant Roberts; probably his Masters thesis is still
available from the University of Waterloo (I read the thesis some years ago;
it appeared to give a fast implementation using very different techniques
from those used at Edinburgh).  It's a pity that Grant Robert's work has
languished in obscurity because he picked the wrong target hardware.

- peter ludemann		ludemann@quintus.com

Let's not forget IC-Prolog, the original Prolog with co-routining developed
by Frank McCabe at Imperial College in the late seventies. Its model of
co-routining led, on the one hand, to the Parlog-Concurrent Prolog class
of parallel logic languages, and on the other hand to the Mu-Prolog type
co-routining Prologs.

Kave Eshghi