- From: Peter F. Patel-Schneider <pfps@research.bell-labs.com>
- Date: Tue, 31 Dec 2002 09:37:02 -0500 (EST)
- To: jjc@hpl.hp.com
- Cc: www-webont-wg@w3.org
From: Jeremy Carroll <jjc@hpl.hp.com> Subject: RE: keeping Jeremy happy Date: Tue, 31 Dec 2002 13:16:27 +0100 > Thanks Peter, again, for these improvements ... > > I hope I am not the only who cares about making OWL DL useable with RDF. > > My main concern at the level of the documents is that the restrictions > implicit in this mapping are not expressed clearly enough elsewhere. In fact, > I would go as far as to say that the other document authors appear unaware of > the restrictions. I would agree with this. > I still feel that the extent of this restrictions is excessive and should be > revised further downward further. > > I have some clarification questions, mainly so that I can make correct > comments about the other documents. > > 1: Is a document of a single triple e.g. > > _:x <eg:a> <eg:b> . > > or > > <eg:x> <eg:d> "foo" . > > a legal OWL Lite document? No. The documents are ambiguous, in that they could either be a statement about an OWL individual or a ``non-logical'' annotation. > 2: Can an OWL/DL document import one that is not OWL/DL. > Can an OWL/Lite doc import one that is not OWL/Lite? Sure, but the result might not be OWL/DL or OWL/Lite. I have not put adequate thought into this issue. I may be able to work on it today. > 3: Can an OWL DL document import two OWL DL documents that use the same URI > refs differently (e.g. <eg:a> as an Individual ID in one, and as an > ObjectPropertyID in the other). No. Again this issue has not been adequately addressed. (The idea is addressed in the RDFS semantics section, by the way.) > 4: Given the following file at <a> > > <a#p> rdf:type owl:DatatypeProperty . > <a#c> rdf:type owl:Class . > > is the following file in OWL DL? > > <> owl:imports <a> . > <x> rdf:type owl:Class . > <x> owl:subClassOf _:r . > _:r rdf:type owl:Restiction . > _:r owl:onProperty <a#p> . > _:r owl:minCardinality "1"^^xsd:integer . > > or the following file: > > <> owl:imports <a> . > <y> rdf:type <a#c> . Not right now. They probably should be, and I will see if I can arrange it so. I *think* that a simple change - to using what is in effect the OWL imports closure of an RDF graph instead of the graph itself - is adequate. > Jeremy peter
Received on Tuesday, 31 December 2002 09:37:11 UTC