Path: utzoo!mnetor!uunet!husc6!psuvax1!burdvax!antares!finin
From: finin@antares.PRC.Unisys.COM (Tim Finin)
Newsgroups: comp.lang.prolog
Subject: Re: A Suggested Additional Predicate for Prolog
Message-ID: <5781@burdvax.PRC.Unisys.COM>
Date: 13 Mar 88 22:33:52 GMT
References: <10303@shemp.CS.UCLA.EDU>
Sender: news@PRC.Unisys.COM
Organization: Unisys Corporation, Paoli Research Center; Paoli, PA
Lines: 24
Summary: Micro-Planner had them!


   From: gast@CS.UCLA.EDU
   Subject: A Suggested Additional Predicate for Prolog
   Date: 12 Mar 88 07:23:48 GMT
 
   Logic programming seems to me to be an ideal paradigm for dealing with
   combinations and permutations.  If there were some way to temporarily
   retract a clause, but leave it's position in the database unchanged,
   programming permutations would be much easier and faster.
 
   In other words, the retract should be undone on backtracking.  For
   consistency sake, an assert that is undone on backtracking should also
   be included, but the rest of this comment does not concern itself with
   a "backtrackable assert."...

I recall that Micro-Planner had such predicates for assert and
retract.  I think that the normal assert (THASSERT) and retract
predicates (THREMOVE?) predicates had the desired behavior and that
there were alternates which were not undone on backup.  Tim

------------------------------------------------------------------------------
Tim Finin			finin@prc.unisys.com
Paoli Research Center		..!{psuvax1,sdcrdcf,cbmvax,bpa}!burdvax!finin
Unisys Corporation		215-648-7446 (o)