*BSD News Article 8970


Return to BSD News archive

Path: sserve!manuel.anu.edu.au!munnari.oz.au!spool.mu.edu!uunet!pipex!unipalm!uknet!edcastle!aisb!aisb!awb
From: awb@ed.ac.uk (Alan W Black)
Newsgroups: comp.unix.bsd
Subject: Re: INTERNATIONALIZATION: JAPAN, FAR EAST
Message-ID: <AWB.92Dec16072619@otter.uk.ac.ed.aisb>
Date: 16 Dec 92 07:26:19 GMT
References: <dosburn.723844277@cwis> <2485@specgw.spec.co.jp>
	<1992Dec14.185028.9757@fcom.cc.utah.edu>
Sender: news@aisb.ed.ac.uk (Network News Administrator)
Organization: Dept of AI, Edinburgh University, UK
Lines: 82
In-Reply-To: terry@cs.weber.edu's message of 14 Dec 92 18:50:28 GMT

In article <1992Dec14.185028.9757@fcom.cc.utah.edu> terry@cs.weber.edu (A Wizard of Earth C) writes:

>   I know a lot of the Japaneese 386BSD community can not email out of Japan,
>   so this is an open letter to them, and to Atsushi MURAI and other people
>   there who are active in this newsgroup.
>
>   What kind of internationalization would best enable you to use 386BSD?
>
>	   o	Kanji support?
>	   o	Kana (Katakan/Hirugana) support?

I have a keen interest (and moderate capability) in the Japanese 
language.  Since getting 386bsd 0.1 up and running I have been
running MULE (Multi-language Emacs) on my machine (as
well as on our departmental machines). Mule is distributed as a set of
patches to emacs-18.58 and is, of course, GNU Copylefted.  It offers
a multilanguage environment under X (fonts are available) or under
terminals which support those languages.

Mule supports Japanese (output and input).  Input is through Wnn
a kana/kanji convertor (which accepts romaji or kana input).  It
the input method is (I am told) compatible with ATOK which is
used on a large number of word processors.  Mule also has support 
for Chinese (with a large variety of input methods) offering
both Mandarin and Taiwanese fonts.  Also there is good support for
Korean.  Other languages that are supported in Mule are all
the major European languages, including Greek, and Russian.  There
is also alpha support for Vietnamese, Hebrew and Thai.

Mule is developed mainly by Kenichi Handa (handa@etl.go.jp) of ETL
Laboratories and there is an active mail-list (though its mainly in
Japanese).  Mule is still in beta test but version 0.9.6 (which is
quite stable) is available by anonymous ftp from
   
   etlport.etl.go.jp [192.31.197.99]:/pub/mule
   sh.wide.ad.jp [133.4.11.11]:/JAPAN/mule

Mule does not use unicode for fonts but does use international
standards for its character representation and tries to use
standard input methods (and representations) for the languages
it supports.

On another tract there is TeX support or Chinese and Japanese
freely available for ftp.

>
>   What is currently used on PCs (with their limited character sets) in Japan?
>

Well the important thing to know about the Japanese PC market is 
that there is not really an IBM PC clone market.  The most popular 
computer series is the NEC 98 series, it is not IBM PC compatiable though
it is a very similar machine.  It runs Japanese MSDOS which is essentially
the same but low-level registers etc are slightly different.  I don't
know if 386bsd has been ported to the 98 series yet but as I'm
moving to Japan next year I would be interested if anyone could
tell me its status.

On NEC 98 the most popular Wordprocessor is Ichi-Taro but most people
use a dedicated word-processor rather than a general PC.  Input is
done through a kana-kanji convertor allowing the user to type in
kana or romaji (roman alphabet).  In spite of me taking some time to
learn the kana layout, I now use romaji for input, as it seems most
Japanese people do.  This is probably because some computers do not
have kana markings on the keyboard.

In summary I think that there is little required to extend 386bsd
for internationalization, or at least only as much as any other
Unix.  If unicode is going to be standard then we do need
more tools, but currently with Mule (or the older Nemacs), kterm,
kinput, JTeX etc there is already reasonable support.  There are other
tools which may be useful (e.g. grep, sort etc) but these are
little to do with 386bsd but the wider Unix community.  

Japanese filenames I suppose may be useful but I'm not sure they are 
worth the trouble, though others may disagree.

Alan

Alan W Black                          2 Buccleuch Place, Edinburgh, UK
Centre for Cognitive Science          tel: (+44) -31 650 4627
University of Edinburgh               email: awb@ed.ac.uk