For the record, I disagree with the first half of Eliot's thesis here.
I think that there is an *excellent* chance of getting consortium and
leading vendors to coalesce in support of a schema proposal which attains
the notorious MPRDV (Minimum Progress Required to Declare Victory) level
but does not rule out downstream extensibility.
MPRDV components IMHO are
1. does what DTDs do in as intuitive as possible a way
2. uses XML syntax
3. is compatible with the RDF data model
4. does basic lexical data typing of character data and attribute
values
The other big thing you need, of course, is data types which are
inheritable and abstracted, to facilitate schema engineering.
It can be (and has been) argued that we have that already with
architectural forms. I think there is a practical determinant as
to whether we are doing OK: if we can build an MPRDV schema facility
which rules out neither AFs nor RDF schema types (two very different
approaches to related but not identical problems) then that's a
step forward.
On the other hand, when Eliot points out that doing this will be
technically difficult, he is correct. -T.