🚀 go-pugleaf

if your database has nextto(a,b), nextto(c,d) etc. then use a different
predicate to define the symmetric version of the relation. I.e.

next-to(A,B) :- nextto(A,B).
next-to(A,B) :- nextto(B,A).

Matt.

-------Dr. Matthew Crocker-----------------------------------------
 Dept. of Artificial Intelligence  | Phone: DAI: +44 -31 650 2702 
   Centre for Cognitive Science    |        CCS: +44 -31 650 4432
     University of Edinburgh       |        FAX: +44 -31 226 2730
         80 South Bridge           | E-mail: M.Crocker@ed.ac.uk or,
   Edinburgh, Scotland, EH1 1HN    |         mwc@aisb.ed.ac.uk

In article <C1C63u.J6n@watserv1.uwaterloo.ca> jssloka@monet.uwaterloo.ca (Scott Sloka) writes:
>
>   Howdy.  My name is Scott Sloka and I am an AI enthusiast at the University of
>   Waterloo.  I have a list of facts such as: nextto(a,b) and I would like
>   nextto(a,b) to be equivalent to nextto(b,a).  This does not work:
>
>   nextto(X,Y):-nextto(Y,X).
>
>   because an endless loop is set up.  I am just beginning Prolog programming
>   because I am doing a project on abstraction and I would like to include this
>   parameter switch equalization in my routines for some of my predicates.  Any
>   pointers would be greatly appreciated.  Thanks in advance.
>
>		   -Scott

Just define

nextto_sym(X,Y) :- nextto(X,Y).
nextto_sym(X,Y) :- nextto(Y,X).

Ciao Jens

% From: jssloka@monet.uwaterloo.ca (Scott Sloka)
% Subject: Help: making a binary predicate reversibly equal
% 
% Howdy.  My name is Scott Sloka and I am an AI enthusiast at the University of
% Waterloo.

Howdy.

%            I have a list of facts such as: nextto(a,b) and I would like
% nextto(a,b) to be equivalent to nextto(b,a).  This does not work:
% 
% nextto(X,Y):-nextto(Y,X).
% 
% because an endless loop is set up.  I am just beginning Prolog programming
% because I am doing a project on abstraction and I would like to include this
% parameter switch equalization in my routines for some of my predicates.  Any
% pointers would be greatly appreciated.  Thanks in advance.

This is a real common problem for beginners.  You need _two_ predicates, one
to handle the symmetry of the relation, and the other to provide the basic
facts.  Try something like:

	nextto( X, Y ) :- nextto_fact( X, Y ).
	nextto( X, Y ) :- nextto_fact( Y, X ).

	nextto_fact( a, b ).
	% and whatever other 'facts' are involved.

%                 -Scott
	-- Scot
+------------------------+-----------------------------------------------+
|Scot Dyer_Rivenburgh    | "Land of song," said the warrior bard,	 |
|dyer@eagle.sharebase.com| "Though all the world betrays thee,		 |
+------------------------+  One sword, at least, thy rights shall guard, |
|(He's right, you know.) |  One faithful harp shall praise thee."	 |
+------------------------+-----------------------------------------------+

In article <C1C63u.J6n@watserv1.uwaterloo.ca> jssloka@monet.uwaterloo.ca
(Scott Sloka) writes:

#  I have a list of facts such as: nextto(a,b) and I would like nextto(a,b)
#  to be equivalent to nextto(b,a). 


This is a common problem. The usual solution is

next_to(A,B) :-
	next_to_fact(A,B).

next_to(A,B) :-
	next_to_fact(B,A).


next_to_fact(e,r).
next_to_fact(i,o).
next_to_fact(o,p).
next_to_fact(q,w).
next_to_fact(r,t).
next_to_fact(t,y).
next_to_fact(u,i).
next_to_fact(w,e).
next_to_fact(y,u)


Example run

| ?- next_to(q,w).

yes
| ?- next_to(w,q).

yes
| ?- next_to(a,b).
no


-- 
Son, all the pretty, intelligent, healthy     #    I have discovered of this
young women are taken. It's a basic rule of   #    basic rule of the universe
the universe, and if you don't like it, go    #    a most remarkable proof,
somewhere else.        -- my dad  1906-1992   #    which this .signature is too