Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!im4u!ut-sally!husc6!sri-unix!sri-spam!ames!elroy!cit-vax!ucla-cs!reeves
From: reeves@ucla-cs.UUCP
Newsgroups: comp.lang.prolog
Subject: forall/2 and negation as failure
Message-ID: <8781@shemp.UCLA.EDU>
Date: Thu, 22-Oct-87 04:43:54 EDT
Article-I.D.: shemp.8781
Posted: Thu Oct 22 04:43:54 1987
Date-Received: Sat, 24-Oct-87 21:16:34 EDT
Sender: root@CS.UCLA.EDU
Reply-To: reeves@CS.UCLA.EDU (John F. Reeves)
Organization: UCLA Computer Science Department
Lines: 34

I'm having a problem reading forall/2 and wondered if anyone  could  help.
In    Sterling    and    Shapiro   p.   270-271,   the   claim   is   that
forall(Goal,Condition) is true for all solutions  of  Goal,  Condition  is
true.  Under this reading not(Goal, not(Condition)) works fine as far as I
can tell.

They claim that

    forall(father(X,C),male(C))

is checking _for_ fathers that have all male children.  My reading is that
it  is  checking  _that_  all  fathers  have  all  male  children.   Their
implementation returns two solutions for the query: the fathers that  have
only  male  children.  In  the database, however, there are fathers who do
not have only male children, so (I would think) the query should fail.

Is there a reading for forall/2 for the following implementation (from S&S)?

	forall(G,C) :- setof(C,G,Cases), check(Cases).

	check([Case|Cases]) :- Case, check(Cases).
	check([]).

Alternatively, is there a case where:

	forall(G,C) :- not(G, not C).

behaves counterintuitively, due to the use of not as negation-as-failure?

Fanx,

  John Reeves
  UCLA CS Department, AI Lab      UUCP: {ihnp4,ucbvax}!ucla-cs!reeves
  3531 Boelter Hall, LA CA 90024  ARPA: reeves@CS.UCLA.EDU