Re: Unicode-based FreeBSD

2008/8/28 Oliver Fromme <olli_at_lurza.secnetix.de>

> Right now, a Chinese sysadmin cannot simply go to a FreeBSD
> console, run mutt and read mails from his co-workers that
> contain Chinese characters.  This *does* work fine with
> Linux.  This is clearly an area where FreeBSD is lacking.

Tz-Huan,
Could you comment on this? Does Linux really display and render all Chinese
glyphs corectly? Do they have video card compatibility issues? How about
top-to-bottom rendering? Is it actually used or useful in China and Taiwan?
This would mean that they use graphical mode (and I was already suspecting
this) and have complex processing layer. Unicode is not only a large bunch
of characters, there are also such things as precomposed-decomposed
characters, bidirectional text, etc.
As for my project, complex rendering layer is definitely out of scope for
now. However, the purpose of this discussion is to gather opinions for
possible future projects. I see three possible layers of unicode support in
FreeBSD:
1) Syscons in pure text mode. This needs improved Unicode support.
2) Non-X console in graphical mode (like Linux). This is not present in
FreeBSD.
3) X and related applciations. From my point of view, this does not require
modification for now.

Probably, all three solutions should be implemented in ideal. And (1)
plain-text console would be used only if running (2) is not possible.
However, implementing (2) is a very expensive task. And that's why I'd like
to receive comments. Probably, it is worth to discuss and extimate costs for
such a project.
Alexander Churanov