David,
I see what you mean when you say that this is a grey area in XSD. I checked
the Recommendation and indeed this is a grey area. It is true that the
current CIQ schema does not violate any normative statement in the
Recommendation. So even though it is intuitively wrong, it is not an
outright error either.
The reason I say it is intuitively wrong is that *actually* extending a
"complex type definition with simple content" into a "complex type
definition with complex content" (i.e., by adding particles) is forbidden by
the Recommendation. So you cannot use the