[tei-council] Comments on the conformance document

[Syd Bauman]

[Council -- my apologies. I wrote this back in late September, but for
 some reason it never got posted. The recent discussion on TEI-L had me
 look for it, and finding that it never got posted here it is ... I'm
 not sure there is much, if anything, new here.]


LB> The TEI's claim is that has sufficient generality and coverage in
LB> its semantics that it can be customized to meet most needs
LB> without too much effort. ... we need to have a metric to assess
LB> the interchangeability ... of the resulting documents. That's
LB> what conformance is.

While it may be a very good idea for us to have a metric to assess
interchangeability, I don't think it's a requirement. After all, P3
and P4 did not have such a metric, or at least not at the same level
of nuance. Nonetheless, I agree that it's a good idea to discuss such
a metric in the Guidelines. But I think (*very* strongly) that the
metric of ease of interchange should not be confused with a
definition of conformance. After all, I can (and should be able to)
create TEI documents that are 100% conformant, but because they rely
on my own taxonomy of type=, my own method for indicating rhyme=, my
own method for indicating rend=, etc., are not easily plopped into
your system. On the other hand, I can create a document that has an
added element which your system could accommodate with only the
smallest of tweaks.

That is, all of the degrees of interchangeability discussed at
http://www.tei-c.org.uk/wiki/index.php/Conformance describe conformant
documents (correction: #5 does not describe a TEI document at all,
and I do not think it should be on the list).

I think that we should focus this discussion on the degrees of
interchangeability, and stop mis-using the term "conformance".
Conformance is a separate (although related) issue about making what
you've done with respect to the Guidelines explicit in a specified,
predictable, way.