UTC Agenda Item: IDNA proposal

[John C Klensin]

--On Thursday, 23 November, 2006 22:23 +0000 Michael Everson
<everson at evertype.com> wrote:

> At 02:37 +0530 2006-11-24, Patrik Fältström wrote:
>> New version can be found as 
>> http://stupid.domain.name/idnabis/table-latest.html
>> 
>> - Accept all LDH in US-ASCII
>> 
>> - Do not accept IPA extensions for latin
> 
> Um. Patrik? Some of those are used in African orthographies.
> We CANNOT ban things based simply on what UCS Table the appear
> in. Is that what you have done?
> 
> (I have not been able to follow all of the discussions.)

Michael,

Some of the IPA characters are problematic because the
conventional glyphs look identical to base Latin/ASCII
characters, or identical to them modulo standard style settings
(e.g., italic).  It won't do to include all of the IPA group as
"Latin" for that reason.  It won't do to exclude all of the IPA
group because a few of them are used in African orthographies.
Someone --and I would be delighted if you would take it on-- is
going to have to go through the block and identify which
characters are needed for those orthographies (and, obviously,
are not glyph-equivalent to basic Latin characters) so that they
can be included.

This sort of issue, which we have been aware of since at least
the time that you pointed out the use of some IPA characters in
African orthographies many months ago, is precisely the reason
some of us have had the hypothesis that, by the time this effort
is over, an additional character class is likely to be needed
that makes the appropriate distinctions.

regards,
     john