Newbie Q: Namespace Handling

[Michael Kay]

> That is because many of Google's tools (and other people, not 
> to pick on Google alone here) have no clue of real XML 
> namespace handling (well there are exceptions but this is a 
> prime example).

Yes, I think it's an example of a tendency I've seen elsewhere: there are
some companies with lots of resources and lots of clever programmers who
think it's smart to do their own XML parsing; and they inevitably get it
wrong.

It might be, though, that they are applying some kind of DTD-based
validation even though the file you generated has no reference to a DTD.
DTDs are not namespace-aware, so they often impose constraints such as the
use of a particular prefix or none. That's one reason XQuery and XSLT 2.0
were changed to be more prescriptive about the prefixes that are generated
in the result document. It's also a good reason for abandoning the use of
DTDs.

Michael Kay
http://www.saxonica.com/