- From: Dan Connolly <connolly@w3.org>
- Date: 29 May 2002 14:16:04 -0500
- To: Jos De_Roo <jos.deroo.jd@belgium.agfa.com>
- Cc: w3c-rdfcore-wg <w3c-rdfcore-wg@w3.org>
On Fri, 2002-05-24 at 11:17, Jos De_Roo wrote: > Dan, > > In the RDFCore telecon d.d. 2002-05-24 agenda item 11 > > [[[ > 11: daml:collection > ACTION:2002-05-03#7 danc summarise to the list the options for > collections, with the objections that've been raised > (or likelihood that "they'll fly") > > > See: > http://www.w3.org/2000/03/rdf-tracking/#rdfms-seq-representation > http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Apr/0334.html > ]]] > > I took the action to create a testcase for this issue. This pretty much works for me (I have a different preferences w.r.t. one of your points of attention below...). i.e. I'd like to use this to dispatch ACTION 2002-05-24#2 (danc) to write up a proposal for incorporating daml:collection into RDF as well. Umm... syntax editors, does this tell you what you would need in order to make the edits? > The proposed testcase is that parsing > > ---- test001.rdf > <rdf:RDF > xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" > xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#" > xmlns:eg="http://example.org/eg#"> > > <rdf:Description rdf:about="http://example.org/eg#eric"> > <rdf:type rdf:parseType="Resource"> > <eg:intersectionOf rdf:parseType="collection"> > <rdf:Description rdf:about="http://example.org/eg#Person"/> > <rdf:Description rdf:about="http://example.org/eg#Male"/> > </eg:intersectionOf> > </rdf:type> > </rdf:Description> > </rdf:RDF> > ---- > > would generate following triples > > ---- test001.nt > <http://example.org/eg#eric> > <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> _:a0 . > _:a0 <http://example.org/eg#intersectionOf> _:a1 . > _:a1 <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> > <http://www.w3.org/2000/01/rdf-schema#List> . > _:a1 <http://www.w3.org/2000/01/rdf-schema#first> > <http://example.org/eg#Person> . > _:a1 <http://www.w3.org/2000/01/rdf-schema#rest> _:a2 . > _:a2 <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> > <http://www.w3.org/2000/01/rdf-schema#List> . > _:a2 <http://www.w3.org/2000/01/rdf-schema#first> > <http://example.org/eg#Male> . > _:a2 <http://www.w3.org/2000/01/rdf-schema#rest> > <http://www.w3.org/2000/01/rdf-schema#nil> . > ---- > > Some points of attention are: > 1/ using namespace xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#" That seems to introduce a dependency from RDF syntax to RDF schema. I'd rather not do that. A new namespace would be the cleanest option, but I think on balance, if we're going to change the syntax, we might as well use this namespace for first/rest/nil/List. http://www.w3.org/1999/02/22-rdf-syntax-ns# I'm also uneasy about leaving the semantics of first/rest 'incomplete' at the RDF level; i.e. allowing layered specs to constrain their extents further. I think I can live with it, but I'd feel better if I had a more clear picture of the rdfms-assertion and MIME type issues. > 2/ using attribute rdf:parseType="collection" works for me. > 3/ just use rdf:Description for collected items of course, you can use any typednode, right? > > That's all. > > -- > Jos > -- Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Wednesday, 29 May 2002 15:16:24 UTC