The per-element-type namespace partition

Hunter, David dhunter at
Mon Dec 20 22:03:08 GMT 1999

From: Tim Bray [mailto:tbray at]
Sent: Monday, December 20, 1999 4:45 PM
> That is equivalent to asserting that
> <html:a html:href="foo"> and
> <html:a href="foo">
> must always in all languages and in all applications be considered
> identical.  Which is not an unreasonable viewpoint.  But the WG at
> the time, after lengthy consideration, couldn't swallow it. -Tim

But it seems to be a fundamental point of view of the specs coming out of
the W3C, implied or otherwise.  Take XSLT, for example.  If I declare a
stylesheet, I do it like this:

<xsl:stylesheet version="1.0"
  <!--other stuff-->

Notice that there is no explicit namespace declared for the version
attribute, it's just assumed to be the same as the <stylesheet> element.
But, there is also a shorthand we can use for stylesheets with one template
that matches against "/", so I could do something like this:

<html xsl:version="1.0" xmlns:xsl="">
<!--other stuff-->

in which case I *have* to declare the namespace of version explicitly.

Are there any specs coming out of the W3C that *don't* assume that

<whatever:element version="1.0" xmlns:whatever="whatever">
is identical to
<whatever:element whatever:version="1.0" xmlns:whatever="whatever">


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at
Archived as: and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo at the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo at the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at

More information about the Xml-dev mailing list