Path: utzoo!utgpu!attcan!uunet!seismo!sundc!pitstop!sun!quintus!ok
From: ok@quintus.uucp (Richard A. O'Keefe)
Newsgroups: comp.lang.prolog
Subject: Re: Determining order of argument unification
Message-ID: <629@quintus.UUCP>
Date: 4 Nov 88 12:26:19 GMT
References: <8192@burdvax.PRC.Unisys.COM>
Sender: news@quintus.UUCP
Reply-To: ok@quintus.UUCP (Richard A. O'Keefe)
Organization: Quintus Computer Systems, Inc.
Lines: 23

In article <8192@burdvax.PRC.Unisys.COM> lang@zeta.PRC.Unisys.COM (Francois-Michel Lang) writes:
>Is it possible to determine (by writing a Prolog program)
>which of these two orders is used for given a Prolog system,
>and, if so, how?

>Credit is due to Kim Marriott and Harald Sondergaard
>(Department of Computer Science, University of Melbourne)
>for inspiring the question.

Anyone who has read Marriott & Sondergaard's paper on the Occurs Check
problem will think they know that the answer is yes, and will have some
idea of what to try.

However, the answer is *no*.  The reason is that there are lots of other
things a Prolog system could do.  For example, the compiler could pick
either order at random, or based on some property such as the spelling
of the function symbol.  Or the compiler might quite plausibly do some
of the arguments out of order if they look nice (I think I wrote a note
about this at Edinburgh).  So f(X,Y+Z,W) might have its arguments
processed in a different order from f(X,W,Y+Z).  More to the point, the
Prolog system might, like SICStus Prolog, have a unification algorithm
which can hack cyclic terms (though other parts of SICStus Prolog didn't,
the last time I tried).