Re: [idn] UTC feedback

[RJ Atkinson <rja@inet.org>]

At 20:21 16/08/00, Keith Moore wrote:
>> In the case of (b) for example, French users may be accustomed to either
>> having accents or not in uppercase. Yet in other languages, the 
>> distinction between accented letters must be maintained -- folding 
>> them would be like folding all vowels to E in all English words. 
>> An acceptable work-around is to register both the name with all 
>> accents and the name with none.
>
>wow.  everything else from the UTC sounded entirely sound to me ..
>perhaps not the absolute perfect answer, but close.
>
>but this notion of an acceptable work-around seems like a stretch.
>
>seems like we really want to avoid the case where a name with accents 
>and the same name without accents can diverge (in languages where they
>really are accents and not completely different letters) and eventually 
>be assigned to different parties, or one gets updated without the other, 
>etc.

Keith,

        Then how do you propose to handle Vietnamese Quoc Ngu ?

        I think that UTC is quite right.  Their proposal is not ideal, 
but it certainly is an acceptable workaround.

        The root problem here is that different languages share
the same characters but have different case mapping rules for
common characters.  In a global system such as we are discussing
here, one needs a global ruleset for normalisation/canonicalisation,
not different rules for France and Viet Nam.  Different rules that
are country-dependent mean that the same input could map to different
values of IDN, which would not be interoperable.

Ran
rja@inet.org