Re: T.61 -> Unicode conversion

[Hallvard B Furuseth <h.b.furuseth@usit.uio.no>]

Leif Johansson writes:
> This thread gives me the heeby-jeebies! We cannot possibly spend this
> much time on T.61.

If we are going to publish a T.61->Unicode mapping, we should try to
avoid errors in it.  In particular when the strprep appendix with the
mapping is normative.  I have no idea if we need the appendix, or if
T.61 is as dead as you suggest.  The URL Howard posted:

  http://www.mail-archive.com/openca-users@lists.sourceforge.net/msg04365.html

recommends to use T.61 in some circumstances, but concludes with "In
2004, you can switch to UTF-8 strings and forget about this entire
section of the guide."  So maybe it's still in use but fading fast?

> I suggest that we say that usage of T.61 is a SHOULD NOT (...)

That sounds nice (from my completely-out-of-touch viewpoint:-), with or
without the appendix.  We should certainly not publish the appendix in a
form which might prolong the life of T.61.

Also, the above URL mentions shift states, Japanese and other encodings
which the appendix does not cover, disagreements on a level I never
dreamed of about which character sets a T.61 string is using by default.
So it looks like the appendix at least needs to mention a number of
caveats, and I'm unsure how much sense it makes that it is normative.

-- 
Hallvard