On Dec 8, 2004, at 11:16 AM, Ray Denenberg, Library of Congress wrote:
> I'll say this though, I suspect we'll just have to accept that it's
> not going to work the way we'd like and move on.
According to Eric van der Vlist, relying on XMLSpy for creating valid
schemas is not a reliable practice. I suspect part of the problem is
the spec itself, but Oxygen might do a better job because IIRC it's
validator is Xerces.
Of course, the best way to write a clean XML Schema is to author it in
RELAX NG ;-)