[tei-council] limited phrase, take 2

[Syd Bauman]

> >I want to be able to say that the <p>s in my corpus headers have a
> >content model which is a specific subset of the <p>s in my corpus
> >texts (only PCDATA, not PCDATA or <s>, for example). How would
> >each of these proposals help me with that problem?

> not at all. but was it supposed to?

No, it wasn't. Which is why I have, at times, expressed doubt as
to whether or not this particular change is worth the effort. (I've
come to think it is, but if it solved the above problem it would
obviously be worth the effort -- *way* worth the effort, as my
daughter would say.)


> this is what god gave you Schematron for, though people are amazing
> reluctant to make use of it.

I humbly disagree. This is what Mrs Clark & Murata gave us
co-occurrence constraints in Relax NG for.