- From: Pierre-Antoine Champin <pierre-antoine.champin@liris.cnrs.fr>
- Date: Tue, 15 Jun 2010 11:07:48 +0200
- To: "Evain, Jean-Pierre" <evain@ebu.ch>
- CC: vmalaise <vmalaise@few.vu.nl>, "Chris.Poppe@UGent.be" <Chris.Poppe@UGent.be>, "Bailer, Werner" <werner.bailer@joanneum.at>, Tobias Bürger <tobias.buerger@salzburgresearch.at>, "public-media-annotation@w3.org" <public-media-annotation@w3.org>
(catching up with the discussion) On 12/06/2010 09:03, Evain, Jean-Pierre wrote: > Yes, I have introduced some fucntional properties. At least things > like an identifier dataproperty shall be functional. I disagree. A resource may have serevarl identifiers (especially if you allow different kinds of identifiers, like URI or ISBN...). ma:identifier should be *inverse functional* though, which is how it can *identify* a resource. NB: although even that can be over-commitment. I once have been told that ISBN number can be reused in some situtation, making it non inverse-functional... But this is really a pathological case, I think... pa
Received on Tuesday, 15 June 2010 09:08:25 UTC