Sebastian Rahtz wrote:
> Martin Holmes wrote:
>> P5 is now (as far as I
>> understand it) closed to further proposals, so that would mean that
>> nothing would be done in P5 at all.
> your proposal is on the table now for P5. That doesn't
> mean it cannot be refined without being chucked out.
> there is a commitment to dealing with it one way
> or another.
I'm not really trying to shut down debate. I'm just a little worried
that it might get lost in a welter of discussion, given the limited
amount of time left before November.
>> have something for the purpose in P5. We will all be using P5 for years,
>> and I think on balance it will be better for tool developers if we have
>> something, rather than nothing.
> It does not have to be part of the TEI schema, though.
> The prose of the Guidelines could say "This sort of thing
> is a good idea, we recommend you use this namespace
> to extend element <XXX> in the header for the purpose
> of your tool", and stop there. Would that just be annoying?
I think it might be counter-productive, because the idea of the proposal
is that there should be a standard way of recording this information, so
that all authoring tools can find information added by other authoring
tools. The information will be useful if it's always encoded in a
standard way. I'm not sure what the difference would be between the TEI
recommending it be encoded in a certain way, and the TEI including it in
the schema, other than that the former makes it slightly more difficult
On the issue of whether some of this information belongs in
revisionDesc, I think we've discussed this before, off-list, but my
feeling is that revisionDesc is where humans record the changes they
made to a file, along with the reasons for making them. An application
can't know the reasons a change was made; it can't explain itself in a
way that's analogous to what humans place in revisionDesc. I envisage
someone using a tool like the Image Markup Tool adding their own entry
in revisionDesc (which they can do in the IMT, by editing the
teiHeader), to record changes they made to the file along with the
reasons for making them, while the tool saves information about itself
in the appDesc tag automatically.