Hi All,
I am restarting afresh the thread
http://mail.nl.linux.org/linux-utf8/2005-01/#00020 titled "Unicode and
the Linux Console (again)". Looking back in the archives of this mailing
list, I noticed a similar thread, thus I am reusing the title
(http://mail.nl.linux.org/linux-utf8/2000-03/#00036)
That thread was asking in March 2000 the same thing we are asking now...
Please tell me if I make any mistakes below.

The clear outcome of the thread
(http://mail.nl.linux.org/linux-utf8/2005-01/#00020) was that in order
to support well Unicode on the Linux console, one should not touch
anymore the kernel. Rather, one should write a user-space program to do
the work.
Markus Khun describes very well this direction at
http://mail.nl.linux.org/linux-utf8/2005-01/msg00061.html
and effectively closes that thread.

The user-space program should use then the framebuffer device of the
Linux kernel. A rather outdated HOWTO on the framebuffer can be found at
http://www.tldp.org/HOWTO/Framebuffer-HOWTO.html
There are several drivers for the Linux Framebuffer, depending on the
graphics card one has. One may use vesafb which is the lowest common
denominator, or hardward-accelerated versions (such as intelfb) if your
card is supported.
There is an issue that some Linux distributions do not enable/use the
Linux framebuffer device but rather prefer the emergency terminal (panic
terminal). One reason why this happens is because most users do not use
the console; using an accelerated framebuffer and X at the same time
might lead to resource conflicts.
It has to be figured out how to get the Linux distributions to provide a
basic framebuffer by default.

Now, are there examples of such user-space software that allows you to
use Unicode on the framebuffer?
They have been mentioned in different sources, to sum it up there are
two that look (to me?) very promising.

A. jfbterm, http://jfbterm.sourceforge.jp/ by Fumitoshi UKAI.
"FBTERM/ME takes advantages of framebuffer device that is supported
since linux kernel 2.2.x (at least on ix86 architecture) and make it
enable to display multilingual text on console. Is is developed on ix86
architecture, and it will works on other architectures such as
linux/pcpc."
While searching for jfbterm, I noticed that it was used for some time in
the Debian distribution to display Unicode in the console, but now it is
not in use as the Unicode support in the emergency console is in favour.
I could not find mailing list archives and I believe that if they exist,
they will be in Japanese.
Last release, May 2004. 

B. uterm
http://members.ispwest.com/hanpaul/uterm.html
by John Paul (could be John Palmisano, not sure).
When you browse the Website and Firefox say "No data", click again the
link until you get it.
uterm looks really promising.
There are two screenshots demonstrating a login screen and a test page
covering a few ranges of Unicode (Boxdrawing, Korean, Cyrillic, Greek,
Graphics characters).
The FAQ (http://members.ispwest.com/hanpaul/uterm.faq) is quite
descriptive and mentions that the latest development was in Jan 2005.
The link to the source code
(http://members.ispwest.com/hanpaul/uterm.src.tgz) is not available (one
can download a binary though).

All in all, 
1. I feel that uterm is quite promising but it needs more people to work
on it (currently it looks like a one-man show).
2. I do not know how to get in contact with the author. Ideas?
3. Do you feel that uterm is the way to go?

Cheers,
Simos Xenitellis



--
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/linux-utf8/

Reply via email to