[tei-council] TEI Conformance

[James Cummings]

Sebastian Rahtz wrote:
> Two important new things here:
> 
> a) the suggestion of a standard namespace for TEI additions. It's
>   a very attractive idea, and I'd welcome it. It both differentiates
>   these from normal elements, and groups them for processing. However,
>   is the burden for the casual user too great?

I would argue that it isn't.  A casual user, de facto, does not need to be
adding things which are not able to be defined with <equiv>.  If they aren't,
they are not a casual user. ;-)

They could always use a very simple xslt (I think I've already put it on the TEI
wiki) which just removes all namespaces, before processing, before outputting
HTML.  While I would not recommend that, it does lower the problems (though they
have to be careful to avoid element collision.

> b) raising the profile of <equiv>, to make it the mechanism
>   by which one legally adds new elements to the TEI namespace.
>   this is the perfect opportunity to sort out the use of <equiv>,
>   which makes me very happy

I think we should have a clear mechanism for documenting such equivalents as
<list type="gloss"> is what my <glist> element is, and then encourage (enforce?)
use of it whenever possible.  If it isn't able to be documented with <equiv>,
then surely it isn't really part of the TEI.  If it isn't really part of the
TEI, then it should be in a separate namespace.  (See, I could have made my
recommendations much more draconian....but I thought I should avoid that.)

> My real problem with James' offering is the naming. Its not
> catchy enough.....

I also worry because of the commonly used meanings of 'extension' dating from
TEI P4.  I agree they aren't very catchy, and I'm happy to take any suggestions
for improvements from anyone who agrees with the categories. ;-)


> I also don't think we should insist on a source description in the header.

That is a requirement from the current guidelines, I didn't feel it was
important to remove it.  That the source for a file is 'born digital' is
important, and better than having an absence of that information.

-James

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