Re: Datalog With Only Binary Relations Is Too Weak

>   [Pat Hayes:]
>
>This is not quite right.

Whoops, you are correct. I shouldn't stay up so late answering emails.

The skolem form I gave would have been appropriate for

(forall (?a ?c) (exists (?b)( <=> (grandparent ?a ?c)
	                                (and (parent ?a ?b)(parent ?b ?c))
    )) )

which isnt quite so plausible, though is in fact still correct. In 
this case, the skolem term  (f ?a ?c) is an intermediate parent/child 
if ?a is ?c's grandparent, and otherwise it is just some random thing 
which makes one of the (parent ...) assertions false. Either way, the 
biconditional comes out true.

Pat

---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes

Received on Monday, 19 March 2001 18:08:04 UTC