- From: Dan Brickley <danbri@w3.org>
- Date: Fri, 20 Jul 2001 09:40:33 -0400 (EDT)
- To: Graham Klyne <Graham.Klyne@Baltimore.com>
- cc: Brian McBride <bwm@hplb.hpl.hp.com>, rdf core <w3c-rdfcore-wg@w3.org>
On Fri, 20 Jul 2001, Graham Klyne wrote: [...] > I'm fine with asserting the existence of the buyer service. The problem I > have is that the *description* of the buyer service asserts the existence > of something that may not actually exist. > [...] > Yet we infer that in one case the goods offered for sale definitely exist, > but in the other case no assertion is made about their existence. There > seems to be no *logical* basis for this difference in interpretation when > the only difference is a naming difference. > > Currently, it seems to me that the Existential-Conjunctive (EC) subset of > first order logic, hence RDF as I understand it, is incapable of expressing > the buyer proposition without actually asserting the existence of that > which is to be purchased. I think part of the problem here is our natural tendency to take a commonsense reading of what FOL "there exists" means, ie. reading existential quantifier as taking about some form of "existence in the world". If we try to take a strong reading of "there exists" we'll be bouncing into a whole family of (what I understand to be) fairly well known puzzles: how do we talk about pictures that depict Unicorns, future events that may not come to pass etc. Dan
Received on Friday, 20 July 2001 09:42:09 UTC