Reply-To (was: piercing corporate)

[Derek Martin]

Please note that originally, I heeded Mike's request to move this
discussion off-list.  However, since you continued it here, and raised
points which I feel require response, I will post here on this topic
one last time.  I will not post in this thread again (and I'm sure the
list will thank me).

On Sat, Feb 07, 2004 at 10:53:43PM -0500, bscott at ntisys.com wrote:
> On Sat, 7 Feb 2004, at 6:00am, invalid at pizzashack.org wrote:
> > I therefore advocate that you not set a reply-to header. 
[SNIP]
> All too often, I find private replies on public lists did not need
> to be private, and the community loses something by making them
> private.  Case in point: The message you sent contained valuable
> information and nothing I could see as needing to be "private".  So
> why send it private?

In this particular case, I asked Mike about his new situation.  I
regard this as a private question, and expected a private reply.  I'll
agree (as I said before) that nothing else was private or sensitive,
and that some of it was even germane; but I feel that I asked a
question to which the recipient may not want to respond to publicly,
and as such it should have been private.  I certainly could have
separated my responses, generating two messages, but I did not feel
that what I contributed really warranted the extra effort.  In fact,
most of what I said was (more or less) said by another respondent.

Additionally, I think you should consider that maybe some people don't
WANT to post publicly.  Maybe they are shy, or concerned that they
will be flamed or laughed at.  This list tends to avoid that kind of
thing, but people who feel that way certainly have precedence for it.
Or maybe they have concerns about their e-mail address appearing in
public archives.  Or, maybe they simply want to communicate with
only YOU, for whatever reason...

But, by setting reply-to, you make it difficult for them to avoid
replying publicly.  Human nature being what it is, many such people
will not notice what you have done, because it is most unusual and
unexpected.  In my opinion, this constitutes a form of violation
(albeit a very minor one).  I believe that you don't have the right to
override someone's preference to reply privately.  But beyond that,
there are also practical considerations, which I'll get into
momentarily.

> So, as far as *I'm* concerned, my headers are correct.  I want
> replies to go to <gnhlug at discuss.org>.  That's what Reply-To is for.

No it isn't; it's to route mail to you in the event that the address
in your headers will not get mail to you, for some reason.  Not
everyone agrees with me, and IIRC there may even be some suggestion of
other uses mentioned in some RFC or other.  But IMO, any other use is
bogus.

> I don't ask you to change yours; please don't ask me to change mine.

You may recall a time maybe two years ago when the list added these
headers by default.  There was a big debate about this, and it was
eventually decided that the headers should be removed.  A fair number
of people agree that such things are evil...  At that time of course,
we were only debating the behavior of the list management software;
however whether it is done by the list management software, or by
individual posters, the problems it introduces are identical.  

Now, admittedly, this problem affects only a relatively few people who
either simply don't want to post publicly, or who fit into a very
narrow category of people who:

  - feel that, because the computer is a tool which exists to make
    life easier, one should not ever need to manually edit the
    "envelope" headers on e-mail to which one is replying;

  - understand that there are 3 possibilities when replying to mail or
    mail-like communications:

      1. replying to the original sender ONLY
      2. replying to the mailing list or news group(s) ONLY
      3. replying to all recipients
      [Any of these may also include new recipients not previously
      mentioned in the headers (which should be CC:'d), which can only
      by necessity be added by some manual process.  Or, in some
      cases, one may want to reply to only a subset of addresses.
      Clearly this also can not be automated, as the mailer can have
      no way to guess which ones you want.  Though, that would be
      cool!]

  - have specifically chosen their e-mail client (at least in part)
    expressly to handle these three cases automatically, after
    researching of clients and their features;

  - have carefully configured said client to automatically handle
    these cases

  - occasionally want to send private mail in response to mail sent to
    a mailing list or news group

  - some of said recipients set reply-to headers to go somewhere other
    than to themselves

I am such a person.  I realize that I am in a (probably very small)
minority on this, which is why I used phraseology such as "I
advocate..." instead of such phraseology as, "it is poor netiquette..."
or, "you are a moron because..."  ;-)

As it happens, the only two people I am aware of who set reply-to on
this list are people I occasionally want to send private mail to.  You
basically know this; you're one of them.

So, despite my care in selecting a very intelligent mail client, and
the hours of painstaking configuration I have done to prevent this
problem, those who set reply-to have managed to thwart all my efforts
brilliantly and painlessly  with the setting of one single (IMNSHO
misused) header.  Perhaps with my description, you can understand the
frustration this causes me.

You are also aware of at least one case where this kind of problem
(public reply which was intended to be private) has caused me some
amount of embarrassment; there have been others which, with management
whose sense of humor was lesser, could have cost me my job.  These
events are (in part) what led me to search for a solution to this
problem in the first place.  These events are not so uncommon...  They
(in part) have led to the development of mailers which make it easy to
reply to exactly who you want to reply to.  

[Granted, you must make that decision and press the correct reply key;
however I will assert that after a short time of not mindlessly
pressing the "reply" key, one becomes accustomed to having to make that
decision, and does not make a mistake with it.  Many people scoff at
this idea, having not tried it themselves; but as yet, I have found no
one who has actually tried it who disagrees.]

Given that I have found such a solution, which works with zero thought
on my part (that is, after having decided exactly which reply case I
want) except for maybe one reply in about 1000, it is essentially not
possible for me to notice when it has been circumvented by someone or
something which has set reply-to.  Human nature is what it is; when so
rare an exception of this sort occurs, we invariably fail to process
it.  This is why all the prompts in Windows which ask, "Are you really
really really sure you want to do this dumb thing that you're about to
do?" are completely ineffective...  Clicking on OK becomes habitual,
and we never notice that we accidentally just deleted our whole
hard drive until it's too late...

The effect of setting reply-to makes it possible for only 2 of the
3 reply cases to be handled automatically by your mailer, no matter
what else you do.  A far, far better solution is for the list to set
Mail-followup-to (best solution), and/or List-* headers; and to use
clients which honor it/them and/or which can be otherwise configured
to recognize mailing lists.  In this way, it remains possible for all
3 of the cases to be handled automatically.  This solution is most
logical, and most efficient.  There will be people who do not choose
to use clients which do this, and must manually edit their headers to
get it right; but I think the burden of what to do about that problem
should be on THEM.

The cases where this presents a problem may be rare, but its occurence
will almost certainly go unnoticed, and the effects of it can be
embarrassing or even devastating...

> > Yes, yes, I realize I should check my headers before I hit the send key
> > ...
> 
>   Yes, you should.

But I already addressed the reality of that problem.


>   You should also change the "Subject" line when hijacking a thread.

I'm sure this isn't the first time I've forgotten; but if you look at
the archives, I believe you'll see that I do so much more often than I
don't.  It was an oversite.

>   There's also the whole "email address in a public email forum" flamewar
> that happened a few months ago.

Is this here to suggest that I don't follow netiquette?  If so, I fail
to see how it supports that idea.  I tried to bring to light the idea
that many people may want to participate, but at the same time not be
required to have their e-mail addresses posted in public archives in
order to prevent at least one method of spam harvesting, and that
therefore attribution or any other posting of individuals' addresses
in a public archive should be considered rude.  You may not agree, and
you're certainly entitled to your opinion.  But I don't see how you
can say this supports a pattern of not following etiquette.

If you're trying to suggest that attempting to convince people of my
point of view on that issue "involves no effort on [my] part," you're
out of your mind.  I spent hours and hours doing so.  And, FWIW, I
spent a lot more than that, including actual cash money, to set up an
anti-spam strategy which still to this day is extremely effective,
which was the reason I brought that to light in the first place.  To
say that there was no effort on my part in any of these cases is
completely spurious.

>   I note that you, personally, Derek, seem to advocate netiquette and
> community involvement only when it involves no effort on your part.  

This notion strikes me as bizarre, not only because of the
aforementioned effort, but also since I have probably spent more time
configuring my mailer to comply with the rules of netiquette than most
people spend configuring their entire systems in their whole lifetime.
You, of course, have no way to know this, but I assure you I have
spent weeks of my life on exactly that problem.  Probably far to much,
to be honest.  But whatever...

[And of course, there are always points of contention and
controversy...]

> If it means *you* need to do more work, then it's okay to ignore it.
> That *REALLY* pisses me off.  Maybe it's just a bad coincidence, but
> I really can't help but notice a pattern here.

Please provide evidence of a pattern of me ignoring netiquette.  If
you can do so, I will earnestly endeavor to correct the problem.  But
I think that you will not find any...

> I say this mainly because I value your knowledge, insights, and
> opinions, on this list and elsewhere, and I would hate to lose them.
> If you were just some net.random, I would just tell you to take a
> hike.

Thank you!  And I yours...  I really meant what I said above; if you
find that I'm ignoring rules of netiquette on a regular basis, I want
to know about it.  OTOH if you find individual cases (and I'm sure you
will) which can be attributed to oversite, then please do so.

<takes a big breath>

-- 
Derek D. Martin
http://www.pizzashack.org/
GPG Key ID: 0xDFBEAD02
-=-=-=-=-
This message is posted from an invalid address.
Replying to it will result in undeliverable mail.
Sorry for the inconvenience.  Thank the spammers.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.gnhlug.org/mailman/private/gnhlug-discuss/attachments/20040209/b30decad/attachment.bin