on language ID (was "Re: [tei-council] Glosses, glosses, everywhere, and ...")

[Syd Bauman]

JC> I don't find @ident confusing or too inappropriate. Most people
JC> won't know the RFCs use 'tag' and may get confused by other
JC> meanings of tag?

JC> Or rename those ident. ;-) Wile I have no problem with ident, if
JC> tag is more consistent, I don't really have any strong objections
JC> to that, only a mild unease because of the way 'tag' is used.

SR> if any change, make it verbose, ie languageIdentifier. its only
SR> used a few times in a document, so better if its as
SR> self-explanatory as possble.

LB> Notwithstanding what I said above (about the confusing <tag>
LB> element), I am half inclined to agree. The other half of my
LB> inclination is to apply a different general principle, keep this
LB> as @ident, and make the current @tag/s into @key/s

JC> I'm generally in favour of more @keys so don't see this as a bad
JC> thing.

I lean away from overloading key= and ident=. While the mechanism is
the same, the semantics here are quite different than those used for,
e.g., <moduleRef>.

If Sebastian's observation that <language> will not occur often in
any given TEI document applies to <langKnowledge> and <langKnown>
too, then perhaps all three should be renamed together.

I mildly prefer "langTag" or "languageTag" to "languageIdentifier",
but don't really care very much. (They all have the disadvantage of
repeating part of the semantic information of the element name, but
such is life, I suppose.)