Tables and contextual rule for Katakana middle dot

John C Klensin klensin at jck.com
Mon Apr 6 11:04:06 CEST 2009



--On Monday, April 06, 2009 10:47 +0200 Patrik Fältström
<patrik at frobbit.se> wrote:

> On 6 apr 2009, at 09.36, Yoshiro YONEYA wrote:
> 
>> Dear John-san and Patrik-san,
>> 
>> In reality, in Japanese context, KATAKANA MIDDLEDOT (U+30FB)
>> is used to compose names and to concatenate words, so it is
>> used various   places.
>> And sometimes preceeding and/or succeeding character is
>> alphabet or   digit
>> ([a-zA-Z0-9]).  Furthermore, KATAKANA MIDDLEDOT is sometimes
>> placed at the beginning or the ending of names.
>> 
>> Therefore, it is very difficult to define rule set for
>> KATAKANA   MIDDLEDOT.
>> What I can say is:
>> 
>>    (KATAKANA MIDDLEDOT) MUST be used in Japanese context.
> 
> 
> Yoneya-san,
> 
> Can you express this in more specific terms that can be
> included in the draft? Do you with "japanese" imply it has to
> be one of the japanese scripts (as John said) for example?

Specifically, while removing the "between" restriction should be
ok, and I guess we could say something equivalent to "there must
be at least one other Hiragana, Katakana, or Han character in
the label", I think we should try to avoid permitting this
character in labels that are otherwise, e.g., Latin script-only.
Would that work?

    john

> 
>      Patrik
> 






More information about the Idna-update mailing list