- From: Thomas B. Passin <tpassin@comcast.net>
- Date: Thu, 20 May 2004 09:34:51 -0400
- To: www-rdf-interest@w3.org
Richard Lennox wrote: > Yes it would be plausible to set up refinements of dc:relation (I think > thats what you mean) > ie. x:isTranslationOf rdfs:subPropertyOf dc:relation > > however there are those relations that I will not be able to define (as I > could never get every concievable relation) so the default would be: > C1 dc:relationC2 > > Thus an additional comment on the relation woul be nice to describe it in > some textual way. It follows the same problem of > C1 my:relation C2 > C3 my:relation C4 > my:relation x:comment "blah blah" if rthe comment is only for that > particular relation. > The idea would be that each statement uses a different instance of the property. You might have to simulate that by using unique subproperties - C1 my:relation_1 C2 C3 my:relation_2 C4 my:relation_1 subPropertyOf dc:relation my:relation_2 subpropertyOf dc:relation You don't have to further constrain these properties, so they could all "mean" the same thing. Or maybe you could just type them as instances - someone who knows the semantics better would be able to say more about this - i.e. my:relation_1 rdfs:type dc:relation my:relation_2 rdfs:type dc:relation Either way, you would only use a given one of these predicates once in a given graph. Cheers, Tom P
Received on Thursday, 20 May 2004 09:34:17 UTC