mis, und, art (was: "mis" update review request)

[Frank Ellermann]

Randy Presuhn wrote:

> I find it very difficult to imagine a case where "mis" would
> be useful in tagging data.

Mark finds new wild and wonderful ways to (ab)use "mis" while
we're talking.  Based on exactly the same mis-conception I had
until yesterday when John shed some light on it.

There's a "high astonishing factor" in what "mis" really means,
the very minimum is to document it, preventing further havoc.

The only other alternative I see is to deprecate the "mis" tag
either by convincing ISO 639, or with brute force in 4646bis.

> perhaps we could conclude that this whole debate is really 
> academic, and that no action is needed.

Strong NAK.  Mark has also plausible ideas how to (ab)use "art",
maybe that also deserves a comment in the registry.  IMO less
critical because 4646 is very clear that it's really not about
programming languages.

Last but not least Mark found a plausible way to use "und" as
a kind of anchor for "whatever it is, I know the script", as
in und-Latn.  The SHOULD NOT in RFC 4646 is apparently wrong,
und-Latn is an excellent use case.

Frank