- From: Graham Klyne <Graham.Klyne@MIMEsweeper.com>
- Date: Fri, 15 Feb 2002 10:33:38 +0000
- To: Pat Hayes <phayes@ai.uwf.edu>
- Cc: Frank Manola <fmanola@mitre.org>, w3c-rdfcore-wg@w3.org
At 06:52 PM 2/14/02 -0500, Pat Hayes wrote: >It occurs to me that the obvious way to do that would be to have another >special property <rdf:source> of a reification, so the 'normal' case would >be 5 triples (a quintriplet?) rather than 4, where the extra one links the >subject to its source, using an appropriate URI, eg (check it out) > > _:r <rdf:type> <rdf:Statement> . > _:r <rdf:subject> _:x . > _:r <rdf:predicate> <http://example.org/dt#USdate> . > _:r <rdf:object> "05-08-67" . > _:r <rdf:source> <http://www.agfa.com/w3c/n3/p7.nt> > >Then we really could say some MT stuff about what it has to mean, eg this >is *false* if that graph doesn't contain the described triple. I think this is a great example of what *could* be done with reification. But I don't think that means *we* (i.e. this working group) should feel compelled to do it. But if different folks do different things, the agreed-upon 'quad' vocabulary may be the thing that helps the various ideas to be cross-related. (Pat, I think I'm mostly agreeing with sentiments you've expressed here and elsewhere.) #g ------------------------------------------------------------ Graham Klyne MIMEsweeper Group Strategic Research <http://www.mimesweeper.com> <Graham.Klyne@MIMEsweeper.com>
Received on Friday, 15 February 2002 06:40:05 UTC