Quoting Bill de hÓra <[EMAIL PROTECTED]>: > > I think we need to update the draft and stress that (X)HTML content is > > not subject to xml:base processing.
I didn't read this carefully enough when I replied earlier, I didn't see that Robert was suggesting that xml:base shouldn't apply to (X)HTML at all (which is what I was suggesting too). Don't forget that the URI of the Atom document also sets the base URI; xml:base is just one way that the base URI is set. A step further would be to say that content has no defined base URI, so relative references must not be used. > Why would we would mandate short-circuiting xml:base processing for > XHTML via spec text, but default namespace processing via markup > signals? Consistency would suggest some kind of wrapper/marker for xml:base. I don't really understand what you mean by a wrapper/marker, can you give an example? -- Dave
