Date: Wed, 6 Mar 2013 22:08:49 +1100
Message-ID: <20130306220849.GB5730@hili.localdomain>
From: Peter Wang <novalazy@gmail.com>
To: Jani Nikula <jani@nikula.org>
Subject: Re: [RFC PATCH 3/3] cli: support Mail-Followup-To: in notmuch reply
In-Reply-To: <871ubwdcy7.fsf@nikula.org>
References:
 <5eaa3acc22ee5513bdce5ab931b7a79ade880e06.1362254104.git.jani@nikula.org>
	<f9c7258be72bec0b174553db0410c29eba2745fa.1362254104.git.jani@nikula.org>
	<20130303120745.GA4884@hili.localdomain>	<871ubwdcy7.fsf@nikula.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
Cc: notmuch@notmuchmail.org
Precedence: list

On Sun, 03 Mar 2013 11:56:00 +0200, Jani Nikula <jani@nikula.org> wrote:
> I have added the reply.honor_followup_to configuration because, although
> widely used, it's not a standard, and therefore some people might not
> like to respect that. Although I'd argue if the sender of the message
> has added that header, it's the sender's wish it should be respected.
> 
> If people think our reply-all should always and unconditionally respect
> Mail-Followup-To if it's present, I'd be happy to throw out the config
> option. It's ugly.

How about making it it optional on the command-line rather than through
configuration?  Then you don't lose the functionality where you can
initiate a reply to all visible From/To/Cc addresses (ready for pruning),
regardless of the value of some header you probably don't even see, even
if typically you *do* want to respect that header.

(Bringing us back to the three reply functions.)

Peter