- From: Jakob Voss <jakob.voss@gbv.de>
- Date: Tue, 24 Oct 2006 10:51:32 +0200
- To: public-esw-thes@w3.org
Benjamin Nowack wrote: > Quentin Reul wrote: > >> I was wondering if you could explain why skos:broader and skos:narrower >> properties have been added as part of the SKOS vocabulary. My query is >> based on the fact that generalization/specification can already be >> represented by properties such as rdfs:subClassOf. > > SKOS concepts differ from rdfs/owl classes, one of the purposes of SKOS > (the main one?) is to be less precise than RDF schemas/ontologies in > order to avoid unwanted restrictions and inference effects caused by > using RDFS or OWL constructs. See http://jodi.tamu.edu/Articles/v01/i08/Doerr/#_Toc497798911 for a short description of the problem. >> Furthermore, the Semantic Web tend to claim that reuse is very >> important. However, properties such as skos:subject or >> skos:primarySubject are already present in other languages such as >> foaf:topic and foaf:primaryTopic respectively in FOAF >> <http://xmlns.com/foaf/0.1/>. I wonder if Semantic Web tends to claim that reuse is very important why it so little reuses the knowledge of library and information science. > The domain of foaf:topic is foaf:Document, the domain of skos:subject, > however, is not restricted (looks like you can tag real-life objects > like conferences etc. with SKOS). I also think SKOS adds some (informally > described) transitivity features to skos:subject (in combination with > broader/narrower). See the idea of "Limited Cost Expansion" in Alistair's NKOS presentation, slide 42-45: http://www.ukoln.ac.uk/nkos/nkos2006/presentations/miles.pdf By the way relevance of a relationship cannot directly be modeled in SKOS. >> The W3C have proposed a draft for Wordnet in RDFS and OWL >> <http://www.w3.org/2001/sw/BestPractices/WNET/wordnet-sw-20040713.html> >> and I was wondering if there was any plans to represent WordNet in SKOS. If Wordnet is used for indexing than there will probably be a SKOS representation of it. Greetings, Jakob
Received on Tuesday, 24 October 2006 08:50:49 UTC