[tei-council] outstanding issues in pure odd

[Syd Bauman]

My recollection on this is fuzzy at best, but IIRC we had thought
that the advanced schema designer would appreciate a <textNode>
element so she could precisely describe where text nodes are and are
not allowed. Of course, buyer beware, there are lots of patterns you
could describe that modern schema languages could not verify, or even
worse, would choke on. We wanted to provide <mixedContent> as a
mechanism that any fool could use in schema design, because it would
only produce a content model that XML 1.0 DTDs would support:
    ( #PCDATA | a | b | c )*    [http://www.w3.org/TR/REC-xml/#sec-mixed-content]
(and thus all other schema languages would support).

My inclination is that is still a good fork to have. That said, a)
I (for one) am not married to using <mixedContent> as opposed to
<content type="mixed"> or whatever, and b) I haven't read the
"<content> vs <mixedContent>" thread yet.