Re: Linux Console in UTF-8 - current state

["H. Peter Anvin" <hpa@xxxxxxxxx>]

Followup to:  <200209110239.g8B2dFt06432@xxxxxxxxxxxxxxxx>
By author:    Tomohiro KUBOTA <tkubota@xxxxxxxxxxx>
In newsgroup: linux.utf8
> 
> Issues:
> 
>  - Number of characters, few hundreds is apparently too small for
>    worldwide usage.
> 
>  - Cannot display CJK doublewidth characters, combining characters,
>    bidi, and other complex scripts like Indic.  Linux Console Project
>    (http://linuxconsole.sourceforge.net/) seems to work on CJK and bidi.
> 
>  - Input methods for languages which cannot be inputted by simple
>    keyboard mappings, like Chinese and Japanese.
> 
> I think we should say that Linux console support *a little subset of*
> Unicode.
> 

That's a good way of putting it.

> Anyway, these problems are not likely to be solved soon.  Thus, please
> go ahead packaging kernel-patch-unicode or so on if it can help a part
> of the people in the world.  In such a case, please don't forget to
> mention that the package cannot handle CJK, combining, nor bidi.

The only sane way to deal with this is to do a console daemon in
userspace.  That way it doesn't have the limitations that putting all
these huge fonts and tables in kernel space inherently carry.  I think
there already is such a package for CJK at least (called kon) --
either way it is the only sane way to deal with this.  By running it
on top of the kernel-provided frame buffer it shouldn't require tons
of configuration either.

	-hpa
-- 
<hpa@xxxxxxxxxxxxx> at work, <hpa@xxxxxxxxx> in private!
"Unix gives you enough rope to shoot yourself in the foot."
http://www.zytor.com/~hpa/puzzle.txt	<amsp@xxxxxxxxx>
--
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/linux-utf8/