On Fri, Jan 21, 2005 at 09:18:15PM -0600, Callum Lerwick wrote:
> (Someone else said, but the attribution did not appear in Callum's post:)
> > I happen to think it's very broken behavior myself; but users who aren't
> > used to mailing lists seem to expect that when they click "Reply", it will
> > go to the person, and when they click "Reply All", it will go to the list.  
> > *shrugs*

I'm quite used to mailing lists, yet I still expect "reply" to reply
only to the sender and "reply to all" to send a copy to each and every
address found in the to:, from:, or cc: headers (possibly excepting mine,
depending on my MUA configuration), regardless of whether the message
came by way of a list or not.  I dislike it when list management programs
break this expectation as a matter of course.

Now, if we were to get all common MUAs to support separate "reply
to sender" and "smart reply" functions where "reply to sender" would
always do exactly that and "smart reply" would detect mailing lists,
check reply-to:, etc., then I would have no problem with that.  Setting
reply-to: on lists tries to turn "reply to sender" into "smart reply",
which is just bad because a) it's not very smart and b) if I say "reply
to sender" I mean "reply to sender".  But, considering the failure
of mutt-like "reply to list" commands to appear in all common MUAs,
I'd say it's a pretty safe bet that a widely-implemented "smart reply"
won't be happening any time soon, either.

> Yes. One point of irritation is EVERY other list *I* happen to be on,
> mangles reply-to. And its how the old list worked for many years. So its
> deeply ingrained habit now to just be able to hit Reply, no matter how
> "incorrect" it may be.

Every other list I happen to be on does not insert reply-to: headers,
with the exception of the half of them that are run through yahoo groups.
My deeply ingrained habit, then, is to use mutt's reply-to-list function
when that's what I want to do, so that I don't have to remember how the
list I'm on at the moment does it.  And, yes, I had completely forgotten
that the old list had set reply-to: on messages because of this; if I
had remembered that detail, I probably wouldn't have commented on the
topic in the first place.

> > Personally, it annoys the heck out of me to get a direct copy of the
> > message instead of just receiving it through the listserv. Procmail rules
> > on List-Id don't work very well on direct replies, after all!
> 
> This is the other point of irritation.

Why is that, assuming that you're getting just the direct reply and
not one direct and one via the list?  A direct reply isn't list mail,
so list filters really shouldn't apply to it, should they?  (If you
are getting two copies, aren't there procmail rules that can be used to
recognize and suppress the duplicate?)

In any case, my big beef with lists setting reply-to: automatically
is that, as you say, getting a duplicate copy is irritating, but, if
someone wants to reply to the list and inadvertently replies privately,
then getting duplicates is the worst thing that will come of it and it's
just irritating, no more.  Inadvertently sending a response to the list
which you intended to be private is much more likely to have significant
negative consequences, ranging from embarrassment to lawsuits.  (Oops...
I didn't mean to break that NDA...)  Not only is leaving reply-to alone
more consistent with non-list mail, it's also safer for the list's
subscribers.

> The fact that weird and annoying issues like this exist in the first
> place, are the reason I think mailing lists are an ugly kludge created
> to solve a problem that no longer exists.
> 
> Another reason? The bounce problem.

That's really more of a problem with email itself rather than lists
in particular.  Different MTAs handle bounces differently, which can
make it very difficult (if not impossible - some MTAs bounce mailing
list messages back to the original sender instead of the list server)
to reliably recognize them.

And then there are incorrectly-configured MTAs to contend with...
I have one user on a Mailman list I run who is apparently on a dialup
connection and using something similar to fetchmail.  I guess he tried to
cleverly handle multiple accounts in multiple domains and was outsmarted
by his MTA, because whenever he retrieves messages, I get a bounce at the
list owner address (fortunately he's in digest mode so nobody else gets
them) complaining that there's no user on his machine with the list's
name (presumably derived from the to: header instead of the envelope
information) and the bounce claims to be coming from my list server
(his MTA seems to be pretending that it's the host in the to: header as
well), right down to HELOing as my domain.  I tracked down his ISP in an
attempt to find out what the actual subscribed address was so I could
remove it, but, of course, they said that the IP address the bounces
came from was their NAT host and they couldn't even try to identify the
user and contact him themselves unless he's doing something illegal.

A long digression, I'll admit, but that's an SMTP/MTA issue that even
people who know what they're doing aren't able to cleanly fix.  How is
a piece of software supposed to deal with it?

> Mailing lists made sense back when everyone wasn't on the net full time,
> or even directly connected to the net at all, (UUCP, FidoNet, Bitnet,
> Compuserve...) but it really doesn't anymore.

Dial-up users still exist.  Even if we assume that everyone on this
list is online full-time (which I think is highly unlikely), we're a
highly-technical segment of the population here and can't be expected to
represent the average joe in that aspect.  Not to mention that, until we
have ubiquitous public wireless coverage, my laptop isn't (and won't be)
on the net full time, even if my other machines are.

I don't think that anyone believes email or mailing lists are perfect,
but I have yet to see anything better.

-- 
The freedoms that we enjoy presently are the most important victories of the
White Hats over the past several millennia, and it is vitally important that
we don't give them up now, only because we are frightened.
  - Eolake Stobblehouse (http://stobblehouse.com/text/battle.html)