Syd Bauman wrote:
> It is probably a good idea for the
> Guidelines to discuss such distinctions and the implication on ease
> of interchange, but such should not be part of the definition of
> conformance, which should remain a yes/no distinction.
Sure, conformance is ultimately yes/no, but unless you
characterize the type, I think it's not worth
spending any effort on.
> To which write-ups, specifically, do you refer?
the one I put in the wiki, the draft which James C
submitted to the Council, and a detailed
post to the council from Laurent
> I recall discussing
> one or two of them on the Council list, at least one with the end
> result that we were agreed, what we were discussing was *not*
> conformance, but something else, perhaps "levels of
oh lord, we're back on terminology again :-}
Yes, if you like, we're discussing levels
of interchangeability, and conformance
is just some moral thing which more
or less anyone can claim. But in that
case, if I was the NEH or the AHRB, I'd
stipulate sticking to a published
and formal level of interchangeability
when I handed money to a TEI project.
So in effect it becomes conformance
in the scary sense (ie it affects your
Information Manager, Oxford University Computing Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431
OSS Watch: JISC Open Source Advisory Service