- From: Jonathan Borden <jonathan@openhealth.org>
- Date: Tue, 24 Sep 2002 23:01:44 -0400
- To: "pat hayes" <phayes@ai.uwf.edu>
- Cc: <www-webont-wg@w3.org>
Pat, > > Well, yes, as I said, this is a good point. But I don't think that > one can rationally agree with this implication but not with the J > entailment (which I understood you did not like, Jonathan, correct?) > ie that any superclass of a range is also a range. Since if you > accept the intersection rule, then you have to concede both that > ranges can sensibly be inside one another (which on my 'intensional' > interpretation is kind of silly, although not exactly incoherent) and > that the minimal range might not be the one that you first thought > of, as it were; and then there seems to be no objection to allowing > larger sets to also be ranges (why not?) all the way up to the > universe. At the time I was under the mistaken impression that disjunctive semantics applied to multiple rdfs:range's (which is why I mistakenly thought the entailment was wrong). I guess the real problem with having multiple rdfs:ranges is that it poses exactly the problem you note -- any larger set is technically an rdfs:range because the _real_ rdfs:range is the intersection of all the rdfs:ranges. I see your aesthetic issue with foo rdfs:range rdfs:Resource . as it ultimately always is... but, no, if we assume conjunctive semantics then I don't see a _logical_ problem. Shrug, I don't have a better solution. Jonathan
Received on Tuesday, 24 September 2002 23:19:34 UTC