Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uunet!comtst!mdv
From: mdv@comtst.UUCP (Mike Verstegen)
Newsgroups: comp.windows.x
Subject: Chinese User Interface in X
Keywords: Chinese NLS X
Message-ID: <135@comtst.UUCP>
Date: 6 Jun 90 20:41:59 GMT
Organization: Domain Systems Inc., WPB, FL
Lines: 46


We are planning to convert an existing multi-national application from a
character based environment to X. We currently use PCs running a a terminal
emulator on top of a special graphics package that interprest 2 bytes
sequences as a single Chinese character (called the "Big-5 code"). We don't
have access to the fonts in these PCs, only the ability to display selected
characters. Additonally, the key codes are grabbed and specially processed
to allow the input of Chinese characters using one of several methods (which
I believe are phonic and pictographically based).

We are currently running X11R3 on ISC 3.2 which does not include any Chinese
fonts, though I have heard that R4 was supposed to have Chinese. I cannot
verify this on our system obviously.

I would like to know from those of you with infinite knowledge and experience
with X and access to R4...

1) Does X11R4 have Chinese fonts? Please note that this is NOT the same
	as the Japanese Katakana characters.

2) If the fonts are not part of the standard distribution, are they avaiable
	anywhere else?

3) Is there any support anywhere for the input of Chinese characters under X?
	The methods used usually involve toggling the character set to be
	entered (English/ASCII), then selecting a mode of Chinese input
	(phonetic, etc), then one keystroke "pops up" a list characters
	(usually a single line of 2 to 9 in the lower left corner of the
	screen), the next keystroke either selects one of those characters,
	or pops up another row... until the character is fully selected
	(usually 2 to 4 keystokes). The selected character (which is two
	bytes) is then returned to the calling read().

4) Has anyone had any experience with the load on the server when using
	languages with large character sets (such as Chinese or Japanese)?
	I am guessing that a larger amount of server memory would be required
	to hold all the fonts for 15,000 possible characters. In a 24 by 24
	cell (which we are currently using) that's over a megabyte. Is this a 
	correct extrapolation, or are fonts loaded on the fly? We plan to use
	X-terminals for some users, so memory requirements are a big concern.

Thanks for you assistance. Please reply by mail since I'll be back on the road
shortly. Mail will be forwared to me, but news won't. I'll summarize responses
if there is any interest.

MDV