Using RDFa to produce self-describing HTML

Dear TAG,

With regards to:

http://www.w3.org/2001/tag/doc/selfDescribingDocuments#UsingRDFa

"Note: at this time, drafts of the [RDFa] specification are available, but  
the media-type registration for HTML itself has not been updated to  
reflect RDFa. As described in TAG Finding [AuthoritativeMetadata],  
conventions like RDFa are normative only if provided for in the applicable  
specification for the media-type in which they are used. Thus, for RDFa to  
be fully integrated with 2 The Web's Standard Retrieval Algorithm, the  
HTML and/or XHTML media-type registrations must be be updated. Use of RDFa  
is in any case encouraged in the interim until that happens."

Both the XHTML2 WG and the RDF in XHTML Task Force (responsible for RDFa)  
disagree with this conclusion.

While we do believe that the namespace document for XHTML would be a  
reasonable place to document the use of RDFa in XHTML, we do not believe  
that in order to be able to extract anything at all from an XHTML document  
it is necessary to record this in the media type.

The media type points to the definition of XHTML at W3C, and the  
definition of XHTML says how you can process it. We do not believe that  
this situation needs to be changed.

By the way, we are only referring to the use of RDFa in XHTML here, for  
which we are chartered.

For the WG and TF,

Best wishes,

Steven Pemberton

Received on Wednesday, 28 May 2008 11:57:49 UTC