comment/question on the latest RDFa editor's draft: role attribute

Shane, Mark

the reference you give is to

http://www.w3.org/TR/2007/WD-xhtml-role-20071004

whereas the latest version seems to be:

http://www.w3.org/TR/2008/WD-xhtml-role-20080407

I have a more substantial question/comment, however. The RDFa spec 
refers to 'role' as one of the possible values for @rel/@rev. It then says:

[[[
Indicates the purpose of the resource. For some possible values, see 
[XHTMLROLE] module.
]]]

I tried to understand what this means for RDFa. If I use 'role' as a 
@rev/@rel value, this means I would have, as triples

<> xhv:role <SOMEURI> .

The [XHTMLROLE] document describes possible values of the @role when no 
namespace is used, and those are listed in [1]. However, the current 
RDFa spec does not allow the usage of

<span rel="role" resource="[banner]"/>

(the prefix-less thing is only expanded for @rel/@rev), so the only 
legal usage is

<span rel="role" resource="[xhv:banner]"/>

(or, alternatively any other URI in another namespace...). I am afraid 
this may be a bit misleading for the reader because the text in the Role 
document might mislead him/her...

If this is all true, could we maybe shortcut the text in the RDFa 
document, without referring to the XHTMLROLE document, with, eg:

[[[
Indicates the purpose of the resource. For some possible values in the 
xhv namespace, see [XHTMLVOCAB] module.
]]]

which has an added bonus: we do not have a dependency on the advancement 
of the Role module as a recommendation (I am not sure what the timetable 
is there, with a bit of bad luck we might be suspended to go to Rec!)

I also have a conformance question: is an RDFa implementation supposed 
to check the resources associated to @rel="role"?

Ivan

[1] http://www.w3.org/1999/xhtml/vocab/

-- 

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf