- From: Danny Ayers <danny.ayers@gmail.com>
- Date: Thu, 16 Mar 2006 22:31:34 +0100
- To: "Peter F. Patel-Schneider" <pfps@research.bell-labs.com>
- Cc: hans.teijgeler@quicknet.nl, bdn_01@hotmail.com, semantic-web@w3.org
On 3/16/06, Peter F. Patel-Schneider <pfps@research.bell-labs.com> wrote: > > From: "Hans Teijgeler" <hans.teijgeler@quicknet.nl> > Subject: RE: OWL Web Ontology Language > Date: Thu, 16 Mar 2006 21:29:07 +0100 > > > Or the example I raised in the past: > > How do you represent the fact that the Ford Company (an Individual) > > manufactures Mustangs (a Class)? > > Hans > > Well, again, just what do you mean here? A very good question. Although OWL has the concept of Class, maybe what you want is the class of Concept. I'm not entirely sure, but I've a feeling Ford Mustangs could be modelled pretty usefully using SKOS [1] and/or FRBR [2]. The former has Broader/NarrowerInstantive, the latter distinguishes between a Work and a Manifestation (of the work). This kind of loosening/indirection does mean that subsumption reasoning isn't directly available, but being able to make a more accurate model of real world may in many cases be worth the cost. Cheers, Danny. [1] http://www.w3.org/TR/swbp-skos-core-guide [2] http://vocab.org/frbr/core -- http://dannyayers.com
Received on Thursday, 16 March 2006 21:31:46 UTC