>  Syd's suggestion of adding a new special purpose element is a good
>  alternative. But it should be added to the divBot class, ...

It is (here @ WWP), and more importantly it was (in my post).


>  and its TEIform= should be P, not NOTE.

While it would be pretty easy to convince me that the TEIform should
not be NOTE, it will be quite hard to convince me it should be P.
After all, a postscript can have multople paragraphs within it.


>  ... and specialPara is a content model to be avoided if at all
>  possible

Fair enough. I chose specialPara for the posting as it does allow
a series of multiple paragraph-level elements or, alternatively,
PCDATA with phrase-level elements, which I think would appeal to a lot
of people's sense of what postscripts are made of. But in fact, in
SGML terms the content of PS becomes much easier to manage once you
give up on the idea of having PCDATA directly inside PS, and concede
that every PS must, therefore, have at least one child P (or similiar).

The content model I'd want for prose would look something like
     (
         ( head | opener | label )*,
         ( bibl | cit | l | lg | list | listBibl | p | q | quote | seg )+,
         ( respLine | closer | trailer )*
     )
plus all the things that can go anywhere: ANCHOR, NOTE, PB, FW, etc.
if needed (i.e., if using XML).

The content model we use for PS here at the WWP has some stuff that
you probably wouldn't want in a postscript; this is because it's the
same content model that we use for NOTE.


>  A div is by definition incomplete. A complete letter should be
>  tagged as a TEXT.
A letter here at the WWP is encoded as a DIV if it is part of a
collection of letters or part of the prefatory material of a larger
work; it is encoded as a TEXT it is a complete letter plopped in the
middle of a larger work. (It would be a TEXT if we were only encoding
the letter itself, too, but we are not encoding individual letters at
this time, just books.) You (Lou) may think we're daft or at least
inconsistent, but in all cases we think of what we're doing as a bit
of a hack. What we'd really like is an element that is a cross
between DIV and TEXT -- one that could indicate "this might stand on
its own" the way DIV does not, but would not carry the "this is a
whole text" implication of TEXT; one that would have content similar
to a DIV, but could appear wherever a P might appear, i.e. would not
need to tessellate its parent.