Re: [htdig] Re: List Reply-To (was Resolving hrefs in JavaScript objects)

[Gilles D. <gr...@sc...>]

According to Geoff Hutchison:
> On Monday, April 15, 2002, at 05:44  AM, Owen Boyle wrote:
> > Sorry about that - I just realised that the HtDig list just refers to
> > itself with a "CC:" and not with a "Reply-To:" so if you just click
> > "reply" you send only to the poster and not to the list.
> >
> > I'll bear that in mind for the future, but maybe the List Maintainer
> > could think about changing that behaviour?
> 
> We've tried both over the years. The problem is that with an explicit 
> list Reply-To header, often users who write to the list (but aren't 
> subscribers) will not get a reply. Almost every mailer I've met will 
> exclusively use the Reply-To header and lop off the original sender.
> 
> I guess the sort message is that I understand your complaint, but I 
> think the alternative is worse--manually pasting in a user's address on 
> almost every message.

Yeah, it gets a little tiresome hearing the same old ill-founded plea
for mucking up a Reply-To header for the benefit of those who'd rather
not think before replying.  There are very good reasons for not putting
in this header, which have been discussed before.

See http://www.unicom.com/pw/reply-to-harmful.html
and http://www.htdig.org/mail/2000/07/0303.html

What gets me is that the simple logic of the way it's done now is lost
on people.  It's so intuitive: when you hit Reply, you reply to the
sender, and when you hit Reply-to-all (or Group-reply) you reply to the
whole list.  Why is this so difficult to grasp?  Sorry, but not being
used to using the Reply-to-all capability isn't a good enough reason
to break a mailing list for everyone else.  Any decent mail program has
this capability, so why not learn to use it?  All it takes is a second
to ask yourself, before hitting Reply, "to whom do I want to send this?"
If you want everyone in on what you have to say, don't just hit Reply.

For a dissenting opinion, see Simon Hill's article at
http://www.metasystema.org/essays/reply-to-useful.mhtml

Simon makes some good points too, but here are two weak ones, IMHO.
First, in "It Doesn't Break Reasonable Mailers" he gives two examples
of Unix-based mailers that allow sending to the From address instead
of the Reply-To address.  The fact still remains that there are a
lot of mailers that don't make this easy (elm being one of them).
This argument is almost contradicted when he argues for putting in
Reply-To as a concession for broken mailers, yet doesn't give a single
example of a popular mailer that lacks the reply to all capability.
So, on the one hand list administrators shouldn't cater to what he calls
"unreasonable" mailers that don't make it easy to circumvent the Reply-To,
but they should cater to some unspecified and seriously crippled mailers?

Secondly, and this is the deal-breaker for me, he makes a passing
reference in the Addendum to the possibility of mail loops, stating it
"should be possible for the list server to detect and prevent this."
We've seen a lot of poorly behaved autoresponders that essentially
"spam" the people who post to the list.  This is annoying, but bearable.
Can you imagine the mayhem if some autoresponder decided to use the
Reply-To header as the address to send to?  I don't have a specific
example, but based on the poorly behaved autoresponders I've seen, this
is a frighteningly real possibility.  Such a program would start spamming
everyone on the list, repeatedly, as it would respond to its own messages.
Not fun, and not as easy to detect and prevent as some might think.
Much more than an annoyance, this creates an emergency situation that
the list administrator must respond to immediately.  Don't count on me
to monitor the list 24/7!  I like to enjoy my weekends.

-- 
Gilles R. Detillieux              E-mail: <gr...@sc...>
Spinal Cord Research Centre       WWW:    http://www.scrc.umanitoba.ca/
Dept. Physiology, U. of Manitoba  Winnipeg, MB  R3E 3J7  (Canada)