From: Peter R. Mueller-Roemer (pmr@cs.uni-frankfurt.de)
Date: Sat May 14 2005 - 09:52:55 CDT
John H. Jenkins wrote:
>
> On May 13, 2005, at 10:33 AM, Hans Aberg wrote:
>
>> Now, to the guillemots. Assume that <<...>> are the Cyrillic
>> guillemots and that [[...]] are the French guillemots. If I write
>> [[Russian text]], is the semantic meaning from that different from
>> <<Russian text>>, assuming the quoted text is the same? The first
>> case is probably part of a French text, quoting some Russian, and
>> the second case, some Russian text quoting Russian.
>>
>
> And the standard Unicode response would be that in the former case,
> you use a font specifically designed for (or optimized for) French,
> and in the latter one designed for Russian.
I hope 'standard Unicode response' is not this oldfashioned typewriter
approach. We just had a conference in Berlin to support multilingual
communication with unicode.
Not that I am happy with the present implementations. The universal
keyboard might be a little farther away, but I plead for some small
improvements that would allow for at least most European + some
near-Eastern languages and technical Symbols to have one easy to
remember keyboard-layout.
Combining diacritical marks (CDM) point the way. Graphite-technology
schows that sequences of CDMs can be handled much more sensible than by
using a default rule of overstriking in the OT-font Arial Unicode MS.
But I want narrow Greek diacritcals (e.g. strong or weak h + accute) to
combine side by side and be centered over the narrow iota and the wide
o-mega. Similarly with some Hebrew accents, Arabic .. Etheopian (make
cononant-vowel-versions enterable as simple sequence of 2 components).
>
> This sort of unification issue is common in Unicode -- Japanese vs.
> Chinese glyph design for Han, French vs. Polish glyph design for
> accented Latin letters -- and has been from the first. Unicode
> considers these glyph issues to be addressed through some other
> mechanism. I should point out specifically that OpenType allows for
> the same font to have different glyphs for the same character
> depending on the language, if the application supports this, as do
> other high-level font formats such as AAT and Graphite.
>
> ========
> John H. Jenkins
> jenkins@apple.com
> jhjenkins@mac.com
> http://homepage.mac.com/jhjenkins/
>
>
>
This archive was generated by hypermail 2.1.5 : Sat May 14 2005 - 09:54:09 CDT