[tei-council] Looking ahead

[James Cummings]

Sebastian Rahtz wrote:
> Apropos of the future, I have been corresponding with folks
> working on some putative standard called TBX, involving ISO TC 37;
> it doesn't really matter what it is, but they have (somewhat
> gingerly) picked up ODD to write their spec in.
> 
> I will assist them as much as I can, obviously, but today
> they raise a point to consider, about modelling
> co-occurrence constraints in ODD ("this element can
> only have value X if attribute Y has value Z"). I am
> wondering on what timescale, and how, to revisit ODD
> with a view to making a 3rd generation version of the language
> in which we could consider such things.
> 
> Does the idea horrify or excite you all?

Sounds like it is important to me.  How does modification of ODD fit in with our 
promises to break backwards compatibility though?  As an ODD document is a TEI 
document like any other, so any TEI P5 1.0 ODD must validate against later 
versions of TEI P5, any modification to ODD to handle co-occurrence must do so 
in a way that doesn't break existing ODDs.  That in itself shouldn't be too hard 
to accomplish, I would have thought, and updating ODD to allow this kind of 
thing seems like a reasonable step.  I was today explaining the concept of ODDs 
to a couple people up in Edinburgh, and maybe the 4th question out of their 
mouths was about co-occurrence constraints.  (Ok, they didn't really understand 
that this was what they were asking for, but if they were built into ODD instead 
of needing something like schematron, my answers would certainly have been 
different.)

-James

-- 
Dr James Cummings, Oxford Text Archive, University of Oxford
James dot Cummings at oucs dot ox dot ac dot uk