----- Original Message -----
From: Tomohiro KUBOTA <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, February 05, 2001 12:34 PM
Subject: Re: Unicode is optimal for Chinese/Japanese multilingual texts
> Hi,
>
> At Sun, 4 Feb 2001 19:53:31 -0500 (EST),
> Henry Spencer <[EMAIL PROTECTED]> wrote:
>
> > So? This problem is *not* specific to Han Unification. Most North
> > Americans cannot read the Blackletter (aka Fraktur) version of U+005A,
or
> > for that matter a Spencerian-calligraphy version of the same. (The
> > handwriting style I was taught was Spencerian [no relation!], and *I*
> > would have trouble with some of the weird uppercase forms now.)
>
> I don't know about Blackletter, Fraktur, nor Spencerian. Are
> anyone unhappy U+005A glyph in, for example, XFree86's
> -misc-fixed-medium-r-normal-* font? Is it impossible to design a
> world-readable glyph of U+005A? For Han Unification characters,
> it is impossible for some characters to be designed so that all
> native CJK speakers can read them.
>
world readable? strange concept .. smacks of unification principles ... :)
some practical examples of glyph variations ...
U+014A LATIN CAPITAL LETTER ENG
U+0194 LATIN CAPITAL LETTER GAMMA
These have different shapes depending on the languages ... there are also
the variant glyphs between Serbian and Russian when in the italic form ...
many other examples ...
the reality is CJK is supported .. the mechanism uses unification principles
... thats life .. deal with it ... there are quite workable strategies
available ...
at least japanese is supported ... sheez ... quite a few of the langauges i
have to deal with aren't supported by ANY operating system .... there are
many languages still awaiting solutions ... and you complain about a system
that supports your language ... because you don't like the mechanism ?
whats so wrong with a font switching mechanism? ... i could see that
complexity is added if you mix japanese text with chinese or korean text ...
but what percentage of the japanese population would ever need to do this?
What percentage of the Japanese population has a high literacy in Korean and
Chinese?
>
> > Then the incompetent suppliers who try to ignore the font issues will go
> > out of business, and the world will be a better place. Our job is to
make
> > sure that Linux does *not* ignore these issues, so it can use a font
that
> > is appropriate to the user. It is *not* our job to re-hash whether Han
> > Unification was a good idea -- that is not our decision.
>
> I want some mechanism for Unicode ordinal text (not rich text) to
> distinguish Chinese, Japanese, and Korean version of Han Unification
> character. To let (non-CJK) people understand this need, I have to
> explain why Japanese people are unhappy with Han Unification.
>
if you are working in a japanese langauge environment ... why would you need
to distinguish?
distinguishing would only be necessary if you need to display Chinese
version of Charaters within Japanese text ...
anyway .. i thought the current convention in the japanese medical industry
or example was to display Chinese characters (in chinese names) using the
japanese version of the glyphs ...
> I understand the solution cannot be modification of Unicode. It is
> against Unicode's policy that once determined characters will not
> changed in future. Thus, I'd like you to think about non-vaporware
> solution for this problem.
>
> There are some solutions suggested so far:
> - Use Japanese fonts for Japanese version of Windows
> What is the mechanism to change font?
> How a text file specify the font?
> No mechanism, responsibility of each user? No!
> - language tag
> It is likely that this will be a vaporware, since
> this is stateful like ISO-2022. However, this can
> be a solution, if developers all over the world
> would come to understand this problem.
>
> ---
> Tomohiro KUBOTA <[EMAIL PROTECTED]>
> http://surfchem0.riken.go.jp/~kubota/
> "Introduction to I18N"
> http://www.debian.org/doc/manuals/intro-i18n/
> -
> Linux-UTF8: i18n of Linux on all levels
> Archive: http://mail.nl.linux.org/lists/
-
Linux-UTF8: i18n of Linux on all levels
Archive: http://mail.nl.linux.org/lists/
