This is related to my recent query about the use of dc:identifier in RDF
rather than specifically to Phil's transforms but they do highlight an
issue that I still don't feel I've got to the bottom of.
In Phil's examples, the XHTML construct
<meta name="dc.identifier" scheme="dcterms.URI"
content="http://www.codestyle.org/test/rdf-DCOriginal.shtml" />
generates the triple (T1):
<http://www.codestyle.org/test/rdf-DCOriginal.shtml>
<http://purl.org/dc/elements/1.1/identifier>
<http://www.codestyle.org/test/rdf-DCOriginal.shtml> .
i.e. the resource X has-as-identifier the resource X.
I'm happy enough with the case (T2)
<http://www.codestyle.org/test/rdf-DCOriginal.shtml>
<http://purl.org/dc/elements/1.1/identifier> "UniqueIDRef:12345" .
i.e. the resource X has-as-identifier the literal "UniqueIDRef:12345"
(as in the DC-RDF doc [1]).
And I'm happy with the case (T3)
<http://www.codestyle.org/test/rdf-DCOriginal.shtml>
<http://purl.org/dc/elements/1.1/identifier>
<http://example.org/something> .
i.e. the resource X has-as-identifier the resource Y (a person is
identified by their passport - the example used by Mikael and Roland a
few weeks back). And the subcase where the subject is a blank node makes
sense (as in the DCQ-RDF doc [2]).
Although I don't think I've seen it used, I guess I could also make
sense of something like (T4)
<http://www.codestyle.org/test/rdf-DCOriginal.shtml>
<http://purl.org/dc/elements/1.1/identifier>
"http://www.codestyle.org/test/rdf-DCOriginal.shtml"^^<http://www.w3.org
/2001/XMLSchema#anyURI> .
i.e. the resource X has-as-identifier the datatyped literal
"http://www.codestyle.org/test/rdf-DCOriginal.shtml", type=xsd:anyURI.
But (T1) seems rather odd to me.
Can the subject and object of a statement using the dc:identifier
property be the _same_ resource?
Pete
[1] DC-RDF http://dublincore.org/documents/dcmes-xml/
[2] DCQ-RDF http://dublincore.org/documents/dcq-rdf-xml/
|