- From: Luc Moreau <L.Moreau@ecs.soton.ac.uk>
- Date: Tue, 14 Feb 2012 03:27:37 +0000
- To: Timothy Lebo <lebot@rpi.edu>
- CC: James Cheney <jcheney@inf.ed.ac.uk>, "public-prov-wg@w3.org" <public-prov-wg@w3.org>
Hi Tim You're right. I am not trying to defend attributes in hasAnnotation at all cost. I think they are present for uniformity reason. Can we provide a rationale why there should not be attributes here? It would be the only relation without them. Professor Luc Moreau Electronics and Computer Science University of Southampton Southampton SO17 1BJ United Kingdom On 14 Feb 2012, at 01:03, "Timothy Lebo" <lebot@rpi.edu> wrote: > > On Feb 13, 2012, at 5:31 AM, Luc Moreau wrote: > >> Hi James, >> >>> >>> note(n2,[ex:style="dotted"]) >>> hasAnnotation(u1,n2) >>> >>> and >>> hasAnnotation(u1,n2,[ex:style="dotted") >>> >> >> To me they are *not* equivalent. >> >>> There are no examples in the DM document showing hasAnnotation with a non-empty list of attributes. >>> >> >> I think we could subtype the relation hasAnnotation: hasTrustAnnotation, hasReputationAnnotation, ... >> > > > It seems to me that we could achieve this by subtyping Note and avoiding the need to qualify hadAnnotation. > > :myNote prov:hadAnnotation :myMetaNote . > > :myMetaNote a prov:Note, my:TrustNote; > rdfs:comment "THAT NOTE OVER THERE IS THE MOST TRUSTWORTHY NOTE EVER. :: signed :: Tim." . > > etc. > > -Tim >
Received on Tuesday, 14 February 2012 03:28:37 UTC