- From: Dave Reynolds <der@hplb.hpl.hp.com>
- Date: Tue, 18 Nov 2003 15:55:58 +0000
- To: "Miles, AJ (Alistair) " <A.J.Miles@rl.ac.uk>
- Cc: "'public-esw-thes@w3.org'" <public-esw-thes@w3.org>
Hi Alistair, > I agree, I think having a property set like ... > > <soks:semanticRelation> > ^ > | > <thes:broader> > ^ ^ ^ > | | | > <thes:broaderGeneric> <thes:broaderInstantive> <thes:broaderPartOf> > > ... would be good for thesaurus-specific applications. However, here is > where we start treading on the toes of RDF RDFS and OWL. The property > <thes:broaderGeneric> would be semantically equivalent to <rdfs:subClassOf>, > and the property <thes:broaderInstantive> would be semantically equivalent > to <rdf:type>. How do we handle this kind of overlap? I'm not sure there is a problem here. If BG and BI are truly equivalent to rdfs:subClassOf and rdf:type then just define them as equivalent using either owl:equivalentProperty or a pair of rdfs:subPropertyOf relations. That's the beauty of RDF - open world, multiple inheritance, cycles allowed. Then a thesaurus processor could take an RDFS file and realise, for example, that an rdfs:subClassOf relation implies thes:broader. Dave
Received on Tuesday, 18 November 2003 10:56:22 UTC