- From: Malachi de AElfweald <malachi@tremerechantry.com>
- Date: Mon, 24 Feb 2003 07:40:37 -0800
- To: Tim Bray <tbray@textuality.com>
- cc: xml-names-editor@w3.org
On Sun, 23 Feb 2003 23:35:31 -0800, Tim Bray <tbray@textuality.com> wrote: > Malachi de AElfweald wrote: >> I am confused. No such thing as a null namespace? But the document >> (that lists you as Editor) says " The namespace name for an unprefixed >> attribute name is always null. " > > Hmm... I'm not actually in the editing loop on this version, but I agree > that that language is a bit confusing. The namespaces rec describes a > condition called "being in a namespace" and gives methods for putting > elements and attributes in namespaces. The only way for an attribute to > be in any namespace at all is for it to have a prefix. No prefix, no > namespace. I'm not sure what they mean by "null namespace" but now I'll > review the language and perhaps comment. The section most under debate is http://www.w3.org/TR/xml-names11/ section 6.2... It is there that it talks about scoping.... >> >> Regarding the xhtml part... are you saying that the only way to make it >> "correct" >> would be <a xhtml:href="http://blah">link</a> ??? Cuz, that isn't what >> any of the >> examples show. > > No, in fact, in XHTML, the "a" element is in the xhtml namespace, but its > href attribute isn't in any namespace. So <xhtml:a href="foo"> is > perfectly correct. -Tim So 'being in a namespace' doesn't mean that it defined in the dtd/xsd pointed to by the Namespace? IE: if I am pointing to xhtml.xsd, doesn't prefixing it with that namespace define that the semantics for that attribute are defined in that xhtml.xsd -- and doesn't it mean that it should be invalid if it isn't pointing to the correct namespace? That is, in fact, what Xerces does for validation. Malachi
Received on Monday, 24 February 2003 10:40:29 UTC