[tei-council] Re: on spec grp 2, datatypes normalized

[Syd Bauman]

> Does it strike anyone that instead of
>   foo="true|false|unknown"
> we could just say that the absence of the foo attribute
> means "unknown"? then all the boolean
> attributes stay as true|false, but some are required
> and some are optional.

But then we'd lose the lovely, if opaque to many, distinction
between "unknown" and "unspecified". 

Seriously, I've oft thought about recommending a different set of
"not-true, not-false" values, but never got around to it. If we can
agree that distinctions between things like
 * unknown (I didn't try to find out)
 * undetermined (I tried, but failed)
 * unknowable (not only did I fail, but you will too)
 * unspecified (answer was not in the source I'm transcribing)
 * indeterminate (I'm sure I could find out, if I had more
                  information)
 * unwilling (I know, but I'm not telling)
can all be collapsed into one non-value, then it may make a lot of
sense.

Alright, it wasn't entirely seriously.