From xemacs-m  Fri Apr 18 11:39:35 1997
Received: from master.control.att.com ([135.205.52.13])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id LAA00756
	for <xemacs-beta@xemacs.org>; Fri, 18 Apr 1997 11:39:34 -0500 (CDT)
Received: from i.control.att.com by master.control.att.com with esmtp
	(Smail3.1.29.1 #3) id m0wIGgj-002ix5C; Fri, 18 Apr 97 12:39 EDT
Received: by i.control.att.com (Smail3.1.29.1 #1)
	id m0wIGgj-000mUWC; Fri, 18 Apr 97 12:39 EDT
Message-Id: <m0wIGgj-000mUWC@i.control.att.com>
Date: Fri, 18 Apr 97 12:39 EDT
From: Larry Auton <lda@control.att.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: xemacs-beta@xemacs.org
Subject: Re: Packaging criteria
In-Reply-To: <199704181607.KAA10475@branagh.ta52.lanl.gov>
References: <m2ohbdz4ey.fsf@altair.xemacs.org>
	<kign2qxouar.fsf@jagor.srce.hr>
	<m2d8rs6dgd.fsf@altair.xemacs.org>
	<kigiv1kok6z.fsf@jagor.srce.hr>
	<m24td468p4.fsf@altair.xemacs.org>
	<199704181607.KAA10475@branagh.ta52.lanl.gov>
X-Mailer: VM 6.27 under 20.1 XEmacs Lucid

>> One point of packaging things separately is to remove restrictions like
>> this.  Or to put it another way, Barry does a better job distributing
>> cc-mode than we do.
>> 
>> Having something as a separate package also does not imply that testing
>> of it stops.

Right.

> I think I have to agree with Steve on this.  I know we may be atypical 
> here, but *many* XEmacs users I know here can start an XEmacs and use
> it for days or weeks without *ever* editing a file in cc-mode (C, C++,
> Java, whatever).  At some points this has been true for me as well
> (although I'll soon be *living* in cc-mode).
>
> I realize that is hard to believe, but it's true.

I'm a case in point.  I have used XEmacs as my principle mail/news
interface for several years.  I have, to my knowledge, never used cc-mode.

> Anyway, I know this isn't a strong argument for packaging cc-mode
> separately, but folks should realize that people use XEmacs in very
> different ways.
> 
> I think Steve's points about maintenance and upgrading are stronger
> and quite valid.

agree.

(Pardon the analogy)

XEmacs is like a Shop-Smith (to nest analogies, the Shop-Smith is like
"the Swiss Army Knife of woodworking machinery): a well-engineered
base with a powerful motor.  The packages are attachments.  Everyone
agrees that without attachments, a Shop-Smith is pretty useless.  This
doesn't take away from the essentiality of the base.  It also doesn't
mean that the base unit is tested in isolation - that would be silly.
Most everybody who buys a Shop-Smith gets the "table saw" attachment,
but it's an attachment nonetheless.  There's no negative stigma
implied by saying that cc-mode is a "package" - just that it's NOT
part of the core/base XEmacs engine.

-lda

