- From: Chris Lilley <chris@w3.org>
- Date: Tue, 18 Mar 2003 12:05:38 +0100
- To: Masayasu Ishikawa <mimasa@w3.org>
- CC: www-tag@w3.org, swick@w3.org, em@w3.org, steven.pemberton@cwi.nl, dave.beckett@bristol.ac.uk
On Tuesday, March 18, 2003, 11:52:03 AM, Masayasu wrote: MI> Dan Connolly (connolly@w3.org) wrote: >> All are invited to review the notes from last week's meeting. >> Note that this was something of an ad-hoc meeting, and >> the 'minutes' are just whatever anybody thought was worth >> putting in the IRC channel... I organized the logs >> and presentation materials just a bit... MI> A little follow-up from that meeting ... one of major issues about MI> embedding RDF in XHTML was how to allow RDF inside XHTML without MI> sacrificing validation. DTD is hopeless on this front, and several MI> people expressed that ##any in XML Schema would not be the desirable MI> solution. Clearly. This is what SVG is counting on, anyway. MI> I mentioned Modular Namespaces (MNS) [1] as one of possible MI> approaches, so as an example I wrote an experimental MNS schema which MI> allows RDF/XML inside XHTML2, among other things [2]. MI> A sample document [3] mixes XHTML2, MathML, SVG, EGIX, ContactXML, MI> HLink and RDF/XML together, and can be validated against this MNS MI> schema using Jing [4]. MI> This experiment only handles the syntax problem of embedding RDF in MI> XHTML, and it doesn't solve the semantic problem of interpreting such MI> a document, I don't think a schema language can be expected to solve that. MI> nor does it solve entity problem or ID issue. why do the xml folks hate scoping so much? scoped ids and scoped entities would be so helpful .... MI> Still, MI> it could demonstrate that directly embedding RDF/XML without sacrificing MI> validation is to some extent possible. MI> I hope this experiment could help the TAG to address RDFinXHTML-35, As I see it the main problem with RDF and XHTML is a problem from the human readable prose - the stuff about ignoring tags and processing the element content. This was the original reason for the 'hide it in attributes' syntactic variant of RDF. So SVG for example does not have a problem embedding RDF inside SVG, at the semantic level. MI> and possibly mixedUIXMLNamespace-33 as well. Possibly. MI> [1] http://www.thaiopensource.com/relaxng/mns.html#example MI> [2] http://www.w3.org/People/mimasa/test/schemas/rng/hybrid.mns MI> [3] http://www.w3.org/People/mimasa/test/xhtml2/hybrid MI> (for testing validation, not quite suitable for rendering) MI> [4] http://www.thaiopensource.com/relaxng/jing.html MI> Regards, -- Chris mailto:chris@w3.org
Received on Tuesday, 18 March 2003 06:05:52 UTC