As far as I understand the specs
(which is still work in progress) this would not be a problem in
xmlns:xhtml="http://www.w3.org/1999/xhtml" would be
a predefined (they call it "undeclared") attribute-level namespace.
But it isn't: the only predefined attribute-level namespace so far
Maybe someone with insider connctions can route this input to the
ongoing specification work?
Am 26.10.2011 00:03, schrieb William F Hammond:
Christoph LANGE writes:
. . . that there is no good reason for requiring non-namespaced
attributes. As an alternative to "no namespace", one will now also
be able to put the RDFa attributes into the XHTML namespace, e.g.
I've not tried to understand what's going here.
But I want to point out that attribute names with colons (':')
won't fly in the text/html serialization of html5.
In my opinion the future of math on the web depends significantly
on the provision for mathml in html5 and, in particular, on the
text/html serialization for such documents.
Namespaces are not part of html5, but, with sufficiently far-sighted
design, I would hope there's a way to keep everything working
Let me go a bit further and point out that while formal xml namespaces
may make sense in the use of xml for data exchange, there is
resistance to using xml namespaces with xml for documents.