From xemacs-m  Thu Jul 24 02:01:16 1997
Received: from mercury.Sun.COM (mercury.Sun.COM [192.9.25.1])
	by xemacs.org (8.8.5/8.8.5) with SMTP id CAA09595
	for <xemacs-beta@xemacs.org>; Thu, 24 Jul 1997 02:01:15 -0500 (CDT)
Received: from Eng.Sun.COM ([129.146.1.25]) by mercury.Sun.COM (SMI-8.6/mail.byaddr) with SMTP id AAA24704; Thu, 24 Jul 1997 00:00:46 -0700
Received: from kindra.eng.sun.com by Eng.Sun.COM (SMI-8.6/SMI-5.3)
	id AAA10027; Thu, 24 Jul 1997 00:00:42 -0700
Received: from xemacs.eng.sun.com by kindra.eng.sun.com (SMI-8.6/SMI-SVR4)
	id AAA02787; Thu, 24 Jul 1997 00:00:35 -0700
Received: by xemacs.eng.sun.com (SMI-8.6/SMI-SVR4)
	id AAA16491; Thu, 24 Jul 1997 00:00:31 -0700
Date: Thu, 24 Jul 1997 00:00:31 -0700
Message-Id: <199707240700.AAA16491@xemacs.eng.sun.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
From: Martin Buchholz <mrb@Eng.Sun.COM>
To: Jens-Ulrik Holger Petersen <petersen@kurims.kyoto-u.ac.jp>
Cc: xemacs-beta@xemacs.org, William Perry <wmperry@aventail.com>
Subject: Re: charset's, coding-systems and w3
In-Reply-To: <lbu3hl0y3c.fsf_-_@orion.kurims.kyoto-u.ac.jp>
References: <199707220817.RAA11955@orion.kurims.kyoto-u.ac.jp>
	<m2lo2zbgte.fsf@altair.xemacs.org>
	<199707220920.CAA14630@xemacs.eng.sun.com>
	<lbu3hl0y3c.fsf_-_@orion.kurims.kyoto-u.ac.jp>
X-Mailer: VM 6.32 under 20.3 "Sofia" XEmacs  Lucid (beta9)
Reply-To: Martin Buchholz <mrb@Eng.Sun.COM>

>>>>> "J-U" == Jens-Ulrik Holger Petersen <petersen@kurims.kyoto-u.ac.jp> writes:

>>>>> "Martin" == Martin Buchholz <mrb@Eng.Sun.COM> writes:
Martin> Whether charset-registry of ascii should ever be anything
Martin> other than "iso8859-1" is highly controversial, and I have
Martin> been flamed for setting it to "jisx0201" in Japanese mode.

J-U> At the time I didn't pay attention to that discussion: could you give
J-U> us briefly both sides of the argument again?

Basically, it's very important for most Japanese users to have jisx0208
characters exactly twice as wide as ascii characters. This is much
more likely when using jisx0201 instead of iso8859-1. 

Contrary, one can argue that ascii and japanese-roman (displayed via
jisx0201 fonts) are actually 2 distinct character sets, and should be
treated as such.  In particular, backslash ("reverse solidus") appears
as a yen sign with my default. Japanese users have been using this
peculiar version of ascii for years.

J-U> Also I want to ask, what is the reason for setting

J-U> 	(set-default-buffer-file-coding-system  'euc-jp)

Japanese EUC is the standard native encoding used by Solaris, and, I
believe, other commercial Unices.  Linux may be a different situation.

J-U> in the definition of the Japanese language environment
J-U> ("language/japanese.el")?

J-U> In my init file I do:

J-U> (set-language-environment 'japanese)
J-U> (set-default-buffer-file-coding-system 'iso-2022-8)

J-U> which seems to work fine: could that cause me any problems or
J-U> inconsistencies?

Not as long as you never use any of the other utilities supplied by
your OS vendor :-)

Actually, euc-jp and iso-2022-8 are sufficiently compatible that
everything might just work out.  Or maybe things will only go wrong
when using jisx0212?

J-U> As I wrote before, with

J-U> 	(set-default-buffer-file-coding-system  'euc-jp)

J-U> binary fetching under w3 is broken, but with 'iso-2022-8 it is fine.
J-U> Should that be fixed in w3 or is it an XEmacs/mule problem?

Unknown.  Could be luck.

Martin

