[idn] prohibit simplified code points

["Adam M. Costello" <idn.amc+0@nicemice.net.RemoveThisWord>]

Someone proposed prohibiting Han code points until the Chinese community
decides what to do about the traditional/simplified matching issue.
Obviously that would be very onerous on the Japanese and Korean
communities.

A similar idea, though, would be to prohibit only the simplified Han
code points until the Chinese community decides what to do about the
traditional/simplified matching issue.  This would not affect other
languages, right?  (If I'm wrong, then disregard this message.)

Let me say up front that I don't advocate this idea.  I think IDN is too
low a layer for this sort of matching, and I advocate IDNA as-is.  But
if there is great opposition to IDNA, prohibiting simplified Han code
points for the time being might be an acceptable compromise.

It would leave the door open for a future version of IDNA that maps
some simplified characters to traditional characters and allows the
other simplified characters (the ones that have no one-to-one mapping)
to stand as they are.  Mixed-case annotations could be used to indicate
which traditional characters want to be mapped back to their simplified
counterparts, but this T->S mapping would be enabled only if the IDNA
prefix contained both an uppercase and lowercase letter, to reduce the
chance that T->S mapping is accidentally performed when not intended
(like on Japanese names).

The door would also be left open for a future version of IDNA that
simply removes the prohibition on simplified Han code points, resulting
in the same IDNA that's now on the table.

AMC