The Future of IDNA

James Seng james at seng.sg
Fri Mar 20 07:31:19 CET 2009


Thanks.

Considering U+0301 (and lets be clear which codepoint we talking
about), I dont see how we can actually map that to null without having
other consequences. Just like we cannot do TC-SC mapping without
affect the others who use the same script.

We all live with bundling over here in CJK.

-James Seng

2009/3/20 Kenneth Whistler <kenw at sybase.com>:
> James,
>
> U+0384 GREEK TONOS is *not* the character in question.
> That is a compatibility *spacing* character, and like
> various other standalone spacing clones of diacritics,
> it is DISALLOWED in the idnabis-tables-05.txt document.
> Compare U+02DB OGONEK, for example.
>
> The character in question for the *actual* Greek
> tonos accent is:
>
> U+0301 COMBINING ACUTE ACCENT
>
> And the equivalences in question for bundling of
> monotoniko Greek would be, for example:
>
> U+03AC GREEK SMALL LETTER ALPHA WITH TONOS
>
>  = <U+03B1 GREEK SMALL LETTER ALPHA + U+0301 COMBINING ACUTE ACCENT>
>
>  = (by bundling for accent ignoring) U+03B1 GREEK SMALL LETTER ALPHA
>
>> Okay. Is U+0384 being used in any other context?
>
> U+0384 is irrelevant. U+0301 is the character in question,
> and of course that is widely used with Latin and Cyrillic
> (and even some other scripts) as well as with Greek.
> So we are talking about Greek-script-specific combinations
> of letters plus accents. And even then, the issue for
> bundling is for the monotoniko representation of modern
> Greek primarily.
>
> --Ken
>
>>
>> -James Seng
>>
>> 2009/3/20 Michael Everson <everson at evertype.com>:
>> > On 20 Mar 2009, at 00:21, James Seng wrote:
>> >
>> >> Is tonos still being used?
>> >
>> > Of course it is.
>> >
>> > Michael Everson * http://www.evertype.com
>
>


More information about the Idna-update mailing list