- From: Aaron Swartz <aswartz@upclink.com>
- Date: Thu, 17 May 2001 16:37:08 -0500
- To: Dave Beckett <dave.beckett@bristol.ac.uk>, <w3c-rdfcore-wg@w3.org>
Dave Beckett <dave.beckett@bristol.ac.uk> wrote:
> The confusion is the different interpretation of rdf:ID in
> propertyElt when the element is empty / non-empty.
It was my understanding that the issue was the fact that statements using
resources as objects could not be reified. For example:
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns:e="http://rdf.example.org/#">
<rdf:Description rdf:about="#bar">
<e:someProperty rdf:ID="foo" rdf:resource="#baz" />
</rdf:Description>
</rdf:RDF>
would be expected to create the triples:
<#bar> e:someProperty <#baz> .
<#foo> rdf:type rdf:Statement ;
rdf:subject <#bar> ;
rdf:property e:someProperty ;
rdf:object <#baz> .
But the grammar does not allow this because of the (somewhat unexpected and
little-known) usage of ID to name a new property.
Along with Jan Grant's proposal for removing the creation of new resources
on empty propElts, I think that the use of ID to name them should also be
removed. Thus your option 2 would always be the case.
--
[ Aaron Swartz | me@aaronsw.com | http://www.aaronsw.com ]
Received on Thursday, 17 May 2001 17:37:19 UTC