- From: Steven Pemberton <steven.pemberton@cwi.nl>
- Date: Fri, 06 Jun 2008 12:38:12 +0200
- To: TAG <www-tag@w3.org>
- Cc: "XHTML WG" <public-xhtml2@w3.org>, RDFa <public-rdf-in-xhtml-tf@w3.org>
Dear TAG members, The RDFa Task Force and the XHTML 2 working group are aware that you had some discussions about RDFa at your 20 May f2f meeting. Thanks for considering our document, and also our input during your meeting. After reviewing your minutes, we have had more discussions about it in the RDFa task force (with XHTML 2 people present) and wanted to provide you with some additional information. First, we believe that what you are trying to do is tightly define how engines on the semantic web can discover semantic relationships in a deterministic way. We agree that this is critical to the continued evolution of the semantic web, and want to ensure that we participate fully. Thanks for bringing this important issue to our attention. Second, we consider that XHTML documents have ALWAYS contained relationship information, but without a well-defined mechanism for extracting that information. Therefore, it is reasonable to define the way that this relationship information can be expressed as RDF. However, we think this issue is independent of the media type used to deliver the containing document. Sections of XHTML may be embedded in other namespaces in multi-namespace documents, and delivered using a non-XHTML related media type, and yet the RDF relationships should still be extractable. With these things in mind, we feel the best course of action is to declare that all documents using the xhtml namespace http://www.w3.org/1999/xhtml are capable of being interpreted to produce RDF triples. In order to support the follow-your-nose use case and support your semantic web model, we propose that we update the document at the end of the namespace URI to indicate there is a GRDDL processor (as described in [1]) AND that we update the prose to indicate that RDF can be extracted using the rules defined in the RDFa Syntax document [2]. We would like to know 1) is our assumption about your concerns correct, and 2) if this course of action would help address those concerns. [1] http://www.w3.org/TR/2007/REC-grddl-20070911/#ns-bind [2] http://www.w3.org/MarkUp/Drafts#rdfa-syntax Best wishes, Steven Pemberton For the XHTML2 WG and the RDFa TF
Received on Friday, 6 June 2008 10:39:05 UTC