[tei-council] Roma, ODD-processing in XSLT 1.0

[Sebastian Rahtz]

On 1 Feb 2011, at 16:56, Syd Bauman wrote:

> 
> Umm ... this may be a stumbling block then, no? I mean, part of the
> point (if not the entire point) of all the work you did packaging
> Roma(1) up as a Debian package is so that we in the big TEI world can
> run the service separate from depending on (connections to) Oxford.

Well, sort of. I never packaged up eXist-with-TEI properly,
and I'd be surprised if anyone really _does_ that locally. If you just install
Roma, I think it defaults to accessing www.tei-c.org, even though it
processes locally. Its true that by using OxGarage as a backend,
I am moving the goalposts again, and making another dependency.

Frankly, I don't see a good way out of these dilemmas.

Obviously some Debian folks can use command line as needed :-}


>> if they have managed to get web Roma working locally, and thats it,
>> they will not be able to move to TEI 1.9.0 + (because at that point
>> the handling of att.global changes)
> 
> So you're saying Roma(1) will not (properly) process any TEI 1.9.0+,
> even if the ODD does not use new features like include=?

Correct. because of the change from att.global being implicit to
being explicitly included in all <elementSpec>.  Roma^1 does not
know about that nicety.

--
Sebastian Rahtz      
Information and Support Group Manager, Oxford University Computing Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

Sólo le pido a Dios
que el futuro no me sea indiferente