- From: Patrick Stickler <patrick.stickler@nokia.com>
- Date: Fri, 18 Jan 2002 18:34:05 +0200
- To: ext Dan Brickley <danbri@w3.org>, RDF Core <w3c-rdfcore-wg@w3.org>
On 2002-01-18 18:07, "ext Dan Brickley" <danbri@w3.org> wrote:
> ...
>
> <foaf:Person foaf:name="Dan Brickley"
> xml:base="http://rdfweb.org/people/danbri/">
>
> ...
> <h:p>An experimental <h:a
h:href="../../example.html">hyperlink</h:a></h:p>
> <!-- does the xml:base affect this hyperlink? -->
I would say no. The xml:base is relevant if the URI is to
be expanded to complete form. But this does not happen
because the link in question is not a URI, insofar as the RDF
parser is concerned. It is part of a literal. It
is simply a sequence of characters.
Thus whether an xml:base is in scope or not depends on
whether the *RDF parser* evaluates a relative URI, not
whether some string may sometime in the future be evaluated
as a relative URI.
Thus, we should support xml:base wherever it occurs
in the RDF/XML, but make it clear to users that an RDF
parser does not evaluate any of the content of a literal
and therefore xml:base attributes only apply to relative
URI refs in attributes that are relevant to the RDF
parser -- i.e. rdf:resource.
One would expect that literals with structured XML content
would retain all relative URIs -- otherwise, they loose
their modularity and portability in context of reuse.
Well, at least that's my take on it.
Cheers,
Patrick
--
Patrick Stickler Phone: +358 50 483 9453
Senior Research Scientist Fax: +358 7180 35409
Nokia Research Center Email: patrick.stickler@nokia.com
Received on Friday, 18 January 2002 11:33:15 UTC