Lists Home |
Date Index |
-----BEGIN PGP SIGNED MESSAGE-----
> -----Original Message-----
> From: Roger L. Costello [mailto:email@example.com]
> Consider this RDF Syntax document:
> <rdf:RDF xmlns:rdf=“http://www.w3.org/1999/02/22-rdf-syntax-ns#”
> <book:Creator>Roger L. Costello</book:Creator>
> Read this as: "This is a Description about the
> BookCatalogue.xml resource. It has a Creator whose value is
> Roger L. Costello."
> It is my understanding that all properties (e.g., Creator)
> must be namespace-qualified. Further, the RDF Syntax spec
> requires that the properties namespace must not only be a
> label, but, in fact, must be a valid URL to an RDF Schema
> document (which defines what the property means, thus
> enabling dynamic understanding of a property). That's my
> understanding. Is it an incorrect understanding? /Roger
It's an incomplete understanding. Your example is a serialization
of RDF in XML. Once again, RDF is not an XML grammar.
If you read the RDF MT draft, which more than any W3C document
defines what RDF is, you'll find that various XML artefacts, such
as XML namespaces, qnames, xml:lang are not a part of an RDF graph.
Wrt to RDF schema documents and URL linking, your understanding is
incorrect. Elements of an RDF schema are referenced by URI refs,
not by hypertext links into a document. Again, the XML schema
document is syntactic artefact. As per RDF, an RDF Schema is
fundamentally a graph, not an XML document, though you may decide
to serialize that graph as an XML document.
Claims made about RDF in XML as if it were RDF (it is not) and as
if the XML were primary (it is not) are apt to lead you astray. It
pays to keep the two distinct; making that distinction has been a
guiding principle of the current RDF working group.
Bill de hÓra
-----BEGIN PGP SIGNATURE-----
Version: PGP 7.0.4
-----END PGP SIGNATURE-----