[RDFa] ISSUE-34: @href everywhere and non-information resources

Hey all,

I want to bring up the issue of non-information resources again for the
purpose of the @href-everywhere discussion.

Mark, you pointed out that @resource would help point to non-information
resources, because, to quote you:

> But what about when XHTML 2 makes all @hrefs
> navigable? A precise interpretation of the mark-up (from an RDF
> standpoint) would be this:
> 
>   <> foaf:knows <http://www.ivan-herman.net/foaf.rdf> .
>   <http://www.ivan-herman.net/foaf.rdf> rdf:type foaf:Document .
http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2007Jun/0156

I think the above is incorrect, because the assertion that a URI is of
type document can be made only by ivan-herman.net. It specifically makes
that assertion by choosing what return code to issue when dereferencing
that URI, as per the TAG resolution on this. A 200 return code means
it's an information resource, while a 303 return code that redirects to
a document about that URI means it can be a non-information resource.

In other words, I can say:

=====
   I know <a rel="foaf:knows"
             href="http://www.ivan-herman.net/foaf#">
      Ivan
   </a>.
=====

and that's okay, because when I click on that link, Ivan's server will
redirect me to foaf.xhtml with a 303, so navigation works just fine, but
I'm actually linking to the non-information resource for semantic purposes.

Is it complicated for publishers? Sure, but that's not our doing, that's
the general issue with RDF and non-information resources.

In other words, I don't see this as an argument for adding @resource.

We can still discuss @resource vs. @href, but I want to make sure we get
the above issue out of the way first.

-Ben

Received on Monday, 9 July 2007 18:15:08 UTC