xml-dev - Re: Schemata are not just constraints

[Martin Bryan <mtbryan@sgml.u-net.com>]

Nice to hear from you Rick. Glad to hear you are not "overdone" at present
(in Wulai). Some of us "hard-boiled" old eggs take longer to get up steam!

> Why should XML Schemas have
> key constraints and uniqueness, but not multiple inheritance?

With the relational database bias of RDF showing itself at W3C do you really
need to ask this question :-) Interestingly the abstract element concept
shows the influence of the UML modellers. It maps nicely to UML stereotypes,
but as this is a feature of UML that is little understood it will take a
while for people to wake up to what is happening here. (AFs for classes!)

>Until we have
> a broad range of different experimental and commercial languages for
> constraining certain things in relation to other things,  we cannot know
> what is powerful and convenient.

Agreed. The real problem is expressing constraints that apply across
classes, rather than inheritable property constraints. (If class X > 12 do
not require class Y, but do check that class Z is restricted to values A or
B.) This is one of the big "holes" in ebXML. We know its boundaries, but are
having difficulty filling it in.

Martin