[tei-council] datatype loosened too far?

[Syd Bauman]

BTW, the issue is (temporarily) moot, because it turns out that 63 s
after making the change, SPQR backed it out in 12927. So it's not in 
P5, and while it's still something we have to tackle, we don't have
to fret over it for tomorrow.

SB> In r12926 SPQR changed the datatype of valItem/@ident from
SB> data.name to data.text without comment; neither bug 647 (nor 148)
SB> were closed.
SB>
SB> My instinct is that this is a bad idea: data.text is way too
SB> floppy. (It boils down to RNG:string, I think; we want at least
SB> RNG:token if not data.word+, no?)
SB>
SB> I'm not sure what, if anything, to do about this. Sebastian -- is
SB> there some reason we want RNG:string?

SR> long-standing argument, isn’t it.
SR> 
SR> values for valItem/@ident will often (sometimes) be keys in foreign
SR> databases, URLs, email addresses etc, as per discussion on 148; the
SR> latter claims for Feb 2010 that "Council reviewed the original
SR> proposal, and agreed (eventually) that @ident on vaIIem to be should
SR> be locally defined with a more flexible datatype (string was
SR> proposed)”
SR> 
SR> flogging dead horses? sleeping dogs?
SR> 
SR> bug 148 _is_ closed, by the way