home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.prolog
- Path: sparky!uunet!mcsun!sunic!sics.se!torkel
- From: torkel@sics.se (Torkel Franzen)
- Subject: Re: Occurs check
- In-Reply-To: smolka@dfki.uni-sb.de's message of 22 Dec 1992 16:20:21 +0100
- Message-ID: <TORKEL.92Dec22170711@lludd.sics.se>
- Sender: news@sics.se
- Organization: Swedish Institute of Computer Science, Kista
- References: <TORKEL.92Dec18231028@lludd.sics.se> <24454@alice.att.com>
- <TORKEL.92Dec20104928@bast.sics.se>
- <1992Dec21.153147.26177@cs.brown.edu>
- <1h7bnlINN5es@eagle.dfki.uni-sb.de>
- Date: Tue, 22 Dec 1992 16:07:11 GMT
- Lines: 10
-
- In article <1h7bnlINN5es@eagle.dfki.uni-sb.de> smolka@dfki.uni-sb.de
- (Gert Smolka) writes:
-
- >This is the perfect answer with respect to
- >rational trees (there is exactly one X such that X=f(X)).
-
- Yes, but the point made by Mark Johnson was that Prolog does not return
- as an answer the conjunction C(X)&C'(Y) for which C(X)&C'(Y)->p(X) follows
- logically from program, but instead returns C(X), which is a simplified form
- of C(X)&C'(Y), equivalent to EY(C(X)&C'(Y)) only in the constraint theory.
-