From xemacs-m  Mon Mar 17 12:15:05 1997
Received: from crystal.WonderWorks.COM (crystal.WonderWorks.com [192.203.206.1])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id MAA07391
	for <xemacs-beta@xemacs.org>; Mon, 17 Mar 1997 12:15:03 -0600 (CST)
Received: by crystal.WonderWorks.COM 
	id QQchhg13779; Mon, 17 Mar 1997 13:05:35 -0500 (EST)
Date: Mon, 17 Mar 1997 13:05:35 -0500 (EST)
Message-Id: <QQchhg13779.199703171805@crystal.WonderWorks.COM>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
From: Kyle Jones <kyle_jones@wonderworks.com>
To: xemacs-beta@xemacs.org
Subject: Re: VM and 20.1-b7-MULE
In-Reply-To: <m2zpw3m4qe.fsf@altair.xemacs.org>
References: <QQchfh02043.199703170517@crystal.WonderWorks.COM>
	<m2bu8jnkhi.fsf@altair.xemacs.org>
	<QQchfj02784.199703170545@crystal.WonderWorks.COM>
	<m2zpw3m4qe.fsf@altair.xemacs.org>
X-Mailer: VM 6.20 under 19.15 XEmacs Lucid (beta99)
X-Face: /cA45WHG7jWq>(O3&Z57Y<"WsX5ddc,4c#w0F*zrV#=M
        0@~@,s;b,aMtR5Sqs"+nU.z^CSFQ9t`z2>W,S,]:[+2^
        Nbf6v4g>!&,7R4Ot4Wg{&tm=WX7P["9%a)_da48-^tGy
        ,qz]Z,Zz\{E.,]'EO+F)@$KtF&V

Steven L Baur writes:
 > Kyle Jones writes:
 > 
 > > The only tricky bit is that 20.0 is already out there.  If I roll
 > > out VM 6.20 with the patches, 20.0 users lose because 20.0 has the
 > > old variables.
 > 
 > I guess that's why we have obsolete functions and variables.  I'll
 > obsolete the old names in the next beta if that will help.

I imagine users of the old variables will Not Be Happy at seeing
them made obsolete in one month!  Although it could be argued
"better now than later" when more users would be affected.

Syncing with an incompatible MULE version that _isn't even on the
street yet_ is abusive of the people who are trying to write code for
your version of Emacs.  "What's the point of writing for XEmacs,
they'll just change things to sync with FSF Emacs and break your code
anyway."  I haven't reached this point yet, but this variable change
really does bother me.  When you release a version to the public, we
need to have confidence that the variables and documentation have a
half-life of more than a few months!  Changes during the (really
alpha) cycle are fair game, but 20.0 is in the hands everyone now.
Also, since FSF Emacs/MULE isn't on the street yet, they could change
the specs for their variables again.  This would mean that VM and
anything else that has been MULE-ized might have to go through another
round of modifications.

I believe the old variables should be supported for a grace period.
Mark them obsolete but support them.

