From: vunzndi@vfemail.net
Date: Thu Nov 01 2007 - 23:14:36 CST
Quoting James Kass <thunder-bird@earthlink.net>:
>
> Andrew West wrote,
>
>> If you were going to ask me what the "best" way to represent kanji
>> ligatures such as <U+2FF5 U+9580 U+9F8D> would be under an ideal
>> Unicode model, I would say as <U+9580 U+200D U+9F8D>, using ZWJ to
>> indicate the ligation, and smart fonts would ligate the two components
>> into a single glyph if they could. ...
>
> Your suggestion to use ZWJ turns out to actually be working in the
> real world. I've successfully tested a font which makes a substitution
> for <U+9580 U+200D U+9F8D>. A glyph from the PUA appears in the
> display instead of the string.
>
Yes. Though this is proably a better way to describe characters like
U+9584 閄 <U+9580 U+200D U+4EBA>
and
U+21B89 𡮉 <U+9580 U+200D U+5C0F>
In fact it might be possible to map to a glyph in the font that does
not actually have a unicode point atall, not even a PUA one.
> Trying the same kind of set-up with the ideographic description
> character, however, does not work here. Unfortunately.
>
I am sure that there should be a way to map the IDS to a PUA glyph
> Should the use of ZWJ in running CJK text to promote the formation
> of "ligatures" for the purpose of representing such personal characters
> be encouraged or discouraged? Do the Unicode mavens have any strong
> objections to such a practice?
>
> Best regards,
>
> James Kass
-------------------------------------------------
This message sent through Virus Free Email
http://www.vfemail.net
This archive was generated by hypermail 2.1.5 : Fri Nov 02 2007 - 06:13:31 CST