>I'll done alot of work in I18N.
I'll assume the above is a typo.
>The complexity of the issue is revealed in the number of
>implementations available.
The number of implementations is perhaps indicative of the lack of
understanding...
>In adding I18N to Inventor, I'm leaning toward mbChars - an X window
>concept which uses single byte characters with escapes to change
>encoding and to represent character sets with more than 256 glyphs.
A character set has not relation to glyphs. Sorry. Perhaps you mean
*characters*.
>Then a Font node simply has to describe a family - the encoding from
the character string takes care of the rest.
What do you do if I specify a coded character set of ISO-10646 and an
encoding of SJIS?
>But I'm not recommending the above, or any other proposal for I18N. HTML
>is surviving today with ISO8859-1 as the only supported encoding. I think
>we should start from there.
Wrong. HTML is *craving*, and being split by a need for something
other than ISO 8859-1.
>I have one addition I'd like to make to my original proposal. I'd
>like to add "MFFloat width" to the Text node
What about tati-gaki for example?
>I'd really like to see the spacing and justification fields left in.
Not the way you describe spacing. Also, again, how about tati-gaki and
bidi?
Your proposal is very ethno-centric, just as the original HTML was. We
have an easy (if very abstract) fix for the latter, not for the
former.
I'm tired. It's 1:00am here in Japan...
Your sig is just a little loud don't you think?
Oyasumi nasai.