Re: [tlug] i-mode i18n

[Stephen Lee <sl@example.com>]

Stephen Lee <sl@example.com> wrote:
> Technology-wise, a single font can support variant glyphs of a single
> character, although I have yet to see a good implementation.  This is
> probably because I don't buy fonts, but considering how even a large
> company like MS have problems with having a consistent look between
> JIS-0208 and JIS-0212 characters, it will take a lot of effort to
> achieve a single CJK font.

I mean a single CJK font with appropriate variant glyphs for each
country, of couse.

> Here's proof that a font can contain variants, from someone you know
> well :-)
> 
> http://www.unicode.org/iuc/iuc13/a10/slides.pdf

And this actually gives a case for the need of a "glyph code" (in
contrast to "character code"), thus somewhat validating TRON's case. 

I have not make up my mind about which code, glyph or charater, is
better.  It does feel kind of silly to code two glyphs of the same
charater at different code points (thinking in terms of coding the two
variant 'g' glyphs), but this is exactly want is done with full-width
roman chars it is not completely unnatural.

Stephen