On 25 Jan 2013, at 14:54, John P. McCaskey <[log in to unmask]> wrote:
> Implementing cRefPattern would be straightforward enough. But what about the more general question: How should one encode what cataloging (er, cataloguing) system @type refers to?
by extension, of course, this applies to @type everywhere.
there is a lot of work to do here. for starters, the technical council has an ongoing action to trawl through (again) all the _ad hoc_ versions of @type
on individual elements and try and make those elements members of att. typed instead (there are over 50 such cases to look at, from memory). If that was
done, and we had isolated all use of @type, we could then have a clean field to look at.
I like the idea of a formal place to record @type in a <taxonomy> for those people who want to, though. Trouble is, we can't change the default
datatype of @type at this stage. obvious people _have_ been using taxonomies, but the recommended way to link would be nice.
Director (Research Support) of Academic IT Services
University of Oxford IT Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431