On 2012-08-29, at 13:26, Dan Brickley wrote: > > On Wednesday, 29 August 2012, Yves Raimond wrote: > Hello! > > A colleague of mine was writing a specification generation software for web ontologies recently, and as part of his testing on multiple RDFS vocabularies, noticed that most of them use the OWL namespace just for its owl:Ontology class, to describe the vocabulary itself. > > Would it make sense to create a new rdfs:Schema class, to make it simpler for people to create basic vocabularies, without having to involve OWL at all? > > > Thanks for the suggestion. I would advise against this: > > 1. There are lots of bits of OWL that it is good to use, even if you don't buy the whole DL vision. For example, knowing about class disjointness, property inverses, FPs and IFPs. Having the OWL 'Ontology' class can serve as a gateway drug to these useful extras. > > 2. You push work onto consumers; everyone who previously queried for owl: Ontology would now have to check for rdfs: Schema too. > > 3. Are there any members of one of these classes that are not also in the other? If yes, I fear this will confuse. If no, then this is purely cosmetics. +1 > 4. Both owl: and rdfs: are pre-declared prefixes in the rdfa 1.1 'initial context', and this addresses most of the cosmetic concern. New schemas ought to be written in rdfa imho. That made me double-take, but I've not seen a schema written in RDFa to see how legible that can be. - Steve -- Steve Harris, CTO Garlik, a part of Experian +44 7854 417 874 http://www.garlik.com/ Registered in England and Wales 653331 VAT # 887 1335 93 Registered office: Landmark House, Experian Way, Nottingham, Notts, NG80 1ZZ
Received on Wednesday, 29 August 2012 12:48:49 UTC