- From: Dan Connolly <connolly@w3.org>
- Date: Tue, 05 Jun 2001 11:47:24 -0500
- To: Brian McBride <bwm@hplb.hpl.hp.com>
- CC: www-rdf-comments@w3.org
Brian McBride wrote: > > Dan, > > Nice one on the sub-property problem. > > I've created a new issue: > > http://www.w3.org/2000/03/rdf-tracking/#rdfms-abouteach > > Regarding the use/mention bug. I'm inclinded to ask Ralph to update the > errata. Would you agree? I'm happy for him to say, on the errata page, that "... value is a collection ..." is an error. But I'm not sure what to suggest as a correction yet. [Ralph isn't critical path for that, though, is he? If the WG decides it's an error, you can do it, no?] But I'm much more interested in the WG's disposition of the test case I submitted than the editorial details. Is it an RDF document? If so, what's the n-triples equivalent? The simplest answer that meets my needs is: "no, it's not an RDF document; there is no rdf:aboutEach attribute in RDF; we're sorry for creating any impression that there was one.": ;-) My answer continues... "Stay tuned for a rules language specification which will allow you to say express this sort of thing and much more. If you agree that there never was an rdf:aboutEach attribute in RDF, we're likely to get to a specification of rules sooner. By the way, here's an XSLT transformation that will take any XML document with rdf:aboutEach in it and produce an RDF document that, when combined with some rules, says what you probably meant to say: http://www.w3.org/2001/04rs22/aboutEachSugar.xsl http://www.w3.org/2001/04rs22/#ae " (hmm... I should probably change that aboutEachSugar.xsl and aboutEachRules.n3 to use a different namespace for the 1st-class-object version of rdf:aboutEach.) I include the test case here to reiterate it; note that it'll work fine with the rules-based interpretation of aboutEach: > > For example, consider using a subproperty of rdf:_2 > > to specify the second member of a collection: > > > > <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:ex="http://example/vocab#"> > > <r:Description r:about="#books" > > xmlns:r="http://www.w3.org/1999/02/22-rdf-syntax-ns#"> > > <r:type > > r:resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#Bag" /> > > > > <r:_1 r:resource="#book1" /> > > > > <ex:member2 r:resource="#book2" /> > > > > <r:_3 r:resource="#book3" /> > > </r:Description> > > > > <rdf:Description rdf:aboutEach="#books"> > > <dc:rights xmlns:dc="http://purl.org/dc/elements/1.1/">all > > mine!</dc:rights> > > </rdf:Description> > > > > <rdf:Description rdf:about="http://example/vocab#member2"> > > <rdfs:subPropertyOf > > rdf:resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#_2"/> > > </rdf:Description> > > </rdf:RDF> > > > > What are the members of #books? is #book2 one of them? > > I can deduce, from the specification of rdfs:subProperty, > > that it is. But knowledge of rdfs:subProperty is not > > required for parsing rdf:aboutEach syntax, is it? > > > > I tried it with SiRPAC and got: > > > > /////// > > Errors during parsing > > > > Fatal error: Unresolved internal reference to books > > (Line number = 25, Column number = 4) > > \\\\\\\ -- Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Tuesday, 5 June 2001 12:48:35 UTC