From xemacs-m  Tue May 13 21:29:23 1997
Received: from mail1-im.etl.go.jp (mail.etl.go.jp [192.50.105.9])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id VAA03619
	for <xemacs-beta@xemacs.org>; Tue, 13 May 1997 21:29:12 -0500 (CDT)
Received: from etlpom.etl.go.jp (etlpom.etl.go.jp [192.31.200.9]) by mail1-im.etl.go.jp (8.8.3/3.5Wpl1-96112918) with ESMTP
	id LAA07934; Wed, 14 May 1997 11:01:16 +0900 (JST)
Received: from etlken.etl.go.jp (etlken.etl.go.jp [192.31.197.11]) by etlpom.etl.go.jp (8.8.5/3.5Wpl4-ETL_MASTER) with SMTP id LAA17885; Wed, 14 May 1997 11:01:14 +0900 (JST)
Received: by etlken.etl.go.jp (SMI-8.6/6.4J.6-ETL.SLAVE)
	id LAA03093; Wed, 14 May 1997 11:02:48 +0900
Date: Wed, 14 May 1997 11:02:48 +0900
Message-Id: <199705140202.LAA03093@etlken.etl.go.jp>
From: Kenichi Handa <handa@etl.go.jp>
To: morioka@jaist.ac.jp
CC: steve@miranova.com, mrb@Eng.Sun.COM, xemacs-beta@xemacs.org, Kazu@Mew.org,
        ksakai@netwk.ntt-at.co.jp
In-reply-to: <199705131341.WAA02029@mikan.jaist.ac.jp> (message from
	=?ISO-2022-JP?B?GyRCPGkyLBsoQiAbJEJDTkknGyhC?= / MORIOKA Tomohiko on
	Tue, 13 May 1997 22:41:31 JST)
Subject: Re: emacs-mule as a coding system name for Emacs' internal format
References:  <199705131341.WAA02029@mikan.jaist.ac.jp>

   From: =?ISO-2022-JP?B?GyRCPGkyLBsoQiAbJEJDTkknGyhC?= / MORIOKA Tomohiko <morioka@jaist.ac.jp>
   Date: Tue, 13 May 1997 22:41:31 JST

     This incompatibilities are very inconvenient.  In particular,
   no-conversion has different meaning in both emacsen.  I discussed
   about this problem with HANDA Ken'ichi, YAMAMOTO Kazuhiko (author of
   Mew; a MIME MUA for Emacs) and SAKAI Kiyotaka, then RMS and HANDA
   Ken'ichi agreed to use following name:

	   no-conversion:	don't convert every bytes
	   emacs-mule:	convert only line break code

Mr. Stallman also told me that the name automatic-conversion is not
good.  It can be acceptable on decoding, but if we consider it can
also be used on encoding, the name `undecided' is much better.

I also think the name `undecided' is reasonable.  For instance,
	(detect-coding-region 1 2) => undecided
makes sense much more than returning `automatic-conversion'.

Another matter is that he also requested to have coding systems of
more informative names.  For instnace, as for the current name of
"CODING", make "REGION-CODING" as the real name and make "CODING" the
alias of it.

   From: Richard Stallman <rms@gnu.ai.mit.edu>
       In this case, real name is "japan-shift-jis", and the aliases are:
	       shift_jis and sjis

I'm not sure which is better, REGION or LANG/SCRIPT, but anyway, this
seems to be a good idea.  He suggested to use "europe-" prefix for
coding systems of names iso-8859-X.

---
Ken'ichi HANDA
handa@etl.go.jp

