W3C home > Mailing lists > Public > www-math@w3.org > October 2011

Re: Namespaced RDFa attributes make RDFa more compatible with math markup

From: Christoph LANGE <ch.lange@jacobs-university.de>
Date: Thu, 27 Oct 2011 20:37:13 +0200
Message-ID: <4EA9A4D9.3040509@jacobs-university.de>
To: David Carlisle <davidc@nag.co.uk>
CC: www-math@w3.org
On 2011-10-27 16:54, David Carlisle wrote:
> However it would be pretty odd to use xhtml-namespaced attributes in an
> xhtml context anyway so I don't think we should ask for that.

On the elements from the XHTML namespace (or, on the corresponding
elements in the non-XML serialization of HTML) it is, of course, not
necessary to use the xhtml:-namespaced variant of the RDFa attributes.
XHTML and RDFa have been _designed_ in a way that XHTML attributes and
RDFa attributes don't clash.  If they should ever clash in future
versions of XHTML or RDFa, that wouldn't be our problem anyway.

But as you said that …

> xlink: is only allowed because of legacy usage (and only in "foreign
> content" ie mathml or svg)

… would xhtml:-namespaced attributes be allowed in that "foreign content"?

> The request to allow namespaced attributes was (as far as I
> understand it) to allow rdfa to be used with other unrelated markup
> languages that might otherwise have clashing attribute names.

The reason why I mentioned this is that Michael once, correctly,
stated, that RDFa 1.1 is flawed by design, as long as it both

* claims to be ready for integration into any XML host language


* mandates about 10 non-namespaced attributes to be offered by the
host language

and that therefore it would never seem eligible for integration into
MathML as a host language, as MathML attributes don't _currently_
clash with RDFa, but as we'd like to retain the freedom to introduce
such attributes in MathML 4.  _That_ bug in RDFa has now been fixed;
that was the point of my mail.

Now please don't start arguing that we'd never want MathML to be
annotated with RDFa anyway, when there is Content MathML and
<annotation-xml>. There is so far no best practice for RDF-compatible
annotations to MathML, which one might indeed want when annotating
formulae with arbitrary non-mathematical information for which there
happens to be a suitable RDF vocabulary. Or maybe there is a best
practice, which I am hereby happy to suggest; please open the PDF at

and jump to the first occurrence of "definitionURL".



Christoph Lange, Jacobs University Bremen
http://kwarc.info/clange, Skype duke4701

Workshop: Ontologies come of Age in the Semantic Web (OCAS)
October 24 at the Intl. Semantic Web Conference, Bonn, Germany
Received on Thursday, 27 October 2011 20:44:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 27 October 2011 20:44:55 GMT