Noah et. al.,

The RDFa Task Force has not completed a formal reply to this document.  However, I wanted to get you some personal but more-or-less authoritative feedback prior to your meeting this week.  Sorry it is so late.

In section 5.1 this document discusses using RDFa to produce self-describing HTML.  First, thanks for including this important emerging technology in your finding.  I have only a few comments:
  1. The title and the section refer to HTML.  To date we have not defined RDFa in terms of HTML, only in terms of XHTML.  Please be sure that you are only referring to XHTML in this section.
  2. The last paragraph reads in part "For this example document to be self-describing, the pertinent media type and the specifications on which it depends must provide for the use of RDFa in XHTML; at the time of this writing, they do not."  The XHTML 2 Working Group disagrees with this statement.  The Media Type definition [1] indicates that XHTML Family document types can use application/xhtml+xml.  The specification [2] defines XHTML+RDFa, a markup language.  Further, the specification declares that the document type is in fact part of that family.  The specification also (obviously) defines how RDF is embedded in XHTML. So the specification, per force, defines the connection and enables the use use of RDFa in XHTML+RDFa. 
  3. You also mention in an Editorial note the XHTML namespace document.  The XHTML 2 Working Group and the RDFa Task Force are investigating changing the namespace document so that it explicitly refers to a GRDDL transform that would extract RDF from documents using that namespace.  This is dependent upon an XSLT implementation of RDFa - one is currently in development.  However, once complete we hope it will satisfy your requirements in that a "follow-your-nose" approach will render a document "self-describing".  However, there has been some discussion that having this in the XHTML Namespace Document (as defined in [3]) will obviate the need for other announcement mechanisms to be used in XHTML+RDFa documents (e.g., a DOCTYPE, SchemaLocation, @version attribute on the root element, or whatever).  I for one would welcome input from the TAG on whether you believe this namespace document alteration is sufficient.
  4. Finally, you have a reference to the Media Type declaration in the last paragraph.  However, it says "XHMTLMediaType" instead of "XHTMLMediaType" ;-)

There may be other formal comments from the group in the future, but these are all I have captured to date. wrote:
I'm pleased to announce the availablity of a new draft of a W3C TAG 
Finding on "The Self-Describing Web".  [1,2]  There are a number of 
significant revisions, most of which were motivated by suggestions made by 
the TAG at the face to face meeting in February, 2008 [3].  I have also 
attempted to address the several suggestions I've received in email since 
the previous draft [4] was published.

I expect that this new draft will be discussed at the upcoming face to 
face meeting of the TAG in Bristol, and it's my hope that we'll find it 
more or less ready for final publication.  I know of at least two areas in 
which a little bit of further cleanup will be needed:

* Tim Berners-Lee has asked me to include a diagram that he's prepared 
that shows in some detail the "algorithm" used to explore information on 
the Web.  I have included it [5], but the only copy available to me is a 
.png that is not suitable for editing or cleanup.  There are at least a 
few typos in it, and the size is a problem when printing. I'm hoping that 
Tim can get me the source, or else can help me get a cleaner copy for 
inclusion in the final finding.

* I have updated the discussion on RDFa [6], reflecting in part the latest 
RDFa drafts [7,8] and also some guidance received from members of the 
group(s) working on RDFa.  I encourage members of the TAG to satisfy 
themselves that the revised text conveys the right messages, and I of 
course encourage those involved with RDFa to make sure I haven't 
misunderstood their intentions.  If that all goes OK, then we'll have to 
decide whether to point to the RDFa working drafts, or else to hold up the 
self-description finding until RDFa progresses.  (FWIW: I'd really like to 
ship the self-desc. finding).

Comments should be sent to the mailing list.  Thank you.



Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142


Shane P. McCarron                          Phone: +1 763 786-8160 x120
Managing Director                            Fax: +1 763 786-8180
ApTest Minnesota                            Inet: