- From: David Wood <dwood@tucanatech.com>
- Date: Sat, 4 Sep 2004 13:10:11 -0400
- To: Ben Adida <ben@mit.edu>
- Cc: mark.birbeck@x-port.net, public-rdf-in-xhtml task force' <public-rdf-in-xhtml-tf@w3.org>
Hi Ben and Mark, I have been reading the XHTML 2.0 WD carefully and comparing it to Mark's excellent 14 Feb paper, "XHTML and RDF". It seems to me that Mark's suggestion on making the 'rel' attribute able to contain QNAMES is a very good one, even critical. Mark goes on to suggest that all values in 'rel' would need to be namespace-prefixed for consistency. The HTML WG seems to have done something a little bit different: The XHTML 2.0 WD seems to say that 'rel' can contain either a list of pre-defined values (given in http://www.w3.org/TR/xhtml2/mod- metaAttributes.html#s_metaAttributesmodule) or user-defined values with defined namespaces. They go on to add another variant that relates to RDF: The use of rel='profile' to point to an external RDF document. I *think* this covers the necessary cases in terms of XHTML syntax, at the expense of a lot of syntactic noise. Am I reading this correctly? I'm not sure there isn't a presumption in the use of rel='profile' that RDF is accessible solely by pointing to RDF/XML documents. I *think* that to really make a Semantic Web, one has to presume the ability to navigate cleanly from "Web Space" (often in HTML) to "RDF Space" (graph traversal) and back again. I don't think that simply pointing to RDF/XML documents necessarily does that. Pardon the rant, I am still formulating some thoughts on the matter and would appreciate comments. Regards, Dave -- David Wood CTO, Tucana Technologies, Inc. http://www.tucanatech.com --------------------> Information Fusion. Tucana. <--------------------
Received on Saturday, 4 September 2004 17:10:14 UTC