W3C home > Mailing lists > Public > www-tag@w3.org > September 2008

Treatment of RDFa in TAG Finding on Self-describing Web

From: <noah_mendelsohn@us.ibm.com>
Date: Mon, 8 Sep 2008 20:53:56 -0400
To: www-tag@w3.org
Cc: "Steven Pemberton" <steven.pemberton@cwi.nl>, "Mark Baker" <distobj@acm.org>, Ben Adida <ben@mit.edu>
Message-ID: <OF7E559286.08EEB1BE-ON852574BF.00021059-852574BF.0004BD88@lotus.com>

As I have mentioned previously on this list, I am trying to wrap up the 
TAG finding on the Self-Describing Web [1], and the major unresolved 
question is whether there is a chain of normative specifications, 
presumably starting with the media type registration for 
application/xhtml+xml [2], that justifies the inference of RDF triples in 
an XHTML document that uses RDFa.

I have posed this question several times [3,4], and although I've received 
some notes with helpful insights [5,6], I don't believe anyone has stepped 
up to either say conclusively:  "yes, here are the pertinent clauses of 
the pertinent normative specifications" or "no, RDFa is not currently 
supported as self-describing content using HTTP GET;  a user agent that 
infers triples from an application/xhtml+xml representation retrieved 
using HTTP is, at least in principle, running some risk that the 
representation was not intended to be interpreted in that manner."  I am 
very close to the point where it is not worth holding up the finding on 
this question.  So, I would first of all be very grateful if one or more 
of you could help me resolve the question on the facts.

Also, I want to signal that lacking such clarity, I will within a couple 
of days propose to move ahead with a story along the lines of:  "The 
community has been unable to reach consensus at this time as to whether 
the pertinent normative specifications provide for an HTTP client to do 
RDFa processing on a representation labeled with media-type 
application/xhtml+xml.  While the risk of unintended consequences of such 
processing is likely low in practice (I.e. because the markup used for 
RDFa is very unlikely to be used for other purposes), RDFa cannot at this 
time be considered a good example of self-describing Web content.", or 
words to that effect.   If you think the story is better than this, and 
can cite chapter and verse to prove it, I'd be delighted to hear the 

Please note that I must publish a draft within the next 2-3 days if TAG 
members are to have appropriate time for review ahead of our F2F meeting 
in 2 weeks;  I am hoping to get final clearance for issuance of the 
finding at that meeting.  Thank you.


[1] http://www.w3.org/2001/tag/doc/selfDescribingDocuments.html
[2] http://www.ietf.org/rfc/rfc3236.txt
[3] http://lists.w3.org/Archives/Public/www-tag/2008Aug/0116.html
[4] http://lists.w3.org/Archives/Public/www-tag/2008Aug/0120.html
[5] http://lists.w3.org/Archives/Public/www-tag/2008Aug/0117.html
[6] http://lists.w3.org/Archives/Public/www-tag/2008Aug/0118.html

Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
Received on Tuesday, 9 September 2008 00:53:26 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:58 UTC