Great Circle Associates List-Managers
(August 1995)

Indexed By Date: [Previous] [Next] Indexed By Thread: [Previous] [Next]

Subject: Re: Clobbering "Reply-To" in Outgoing Mail
From: dattier @ wwa . com (David W. Tamkin)
Date: Fri, 18 Aug 1995 10:42:48 -0500 (CDT)
To: scs @ lokkur . dexter . mi . us (Steve Simmons)
Cc: list-managers @ greatcircle . com
In-reply-to: <> from "Steve Simmons" at Aug 17, 95 08:25:52 pm

Steve Simmons took some exceptions to Chip Rosenthal's anti-munging position:

(Note: I am one of the people who tried to talk Chip out of munging before he
 saw the light on his own.  That said, the lists I run offer default public
 reply as a discouraged option and one has a sublist which, by its nature,
 justifies default public replies, but as a general rule I am very strongly
 in favor of default private replies.)

R>   o  It provides no benefit to the user of a reasonable mailer.

S> This presumes of course, that the readers of the mailing list all have
S> access to a reasonable mailer.

OK, Steve, suppose a list member has a mailer with no group-reply facility.
Which is easier for him or her to do:

1. To override a default private reply by typing in the list address, which
   the member may well know by heart or for which he or she may have an alias
   installed; or

2. To override a default public reply by writing down and typing in the
   author's address, which the respondent is very unlikely to have already
   memorized or aliased?

S> Consider the following [sample] from [the General Technics/Permanent
   Floating Riot Club list, whose maintainer Steve called just "Gabe"] ... :

S> gt> From:
S> gt> Subject: Re: Part help?...
S> gt> To:
S> gt> Sender:
S> gt> Reply-To:
S> gt> Date: Thu, 17 Aug 1995 12:49:43 -0400 (EDT)
S> gt>
S> gt> {#} Replies are directed back to
S> gt> {#} To reply to the author, write to

Some HUGE points to make here:

1. A reader wishing to send a private reply still has to write down or
   memorize the author's address and type it in.  If the reader's mailer
   does have a group-reply facility, its functionality is short-circuited.

2. You left a *big* question open.  What happens when the author sends
   headers with a Reply-To: address that differs from the From: header?
   (The sample you used originated from BIX, so it probably had no incoming
   Reply-To: header.)  Does Gabe put the incoming Reply-To: information into
   the body or always use From:, losing the original Reply-To:?  If he always
   takes it from From:, then the real address for private replies is *lost*,
   just as Chip said happens when you munge.

3. If you're taking the stand that reminding people where to send a private
   reply is just as good as making it easy, then anyone else can take the
   position that reminding people where to send a public reply is just as
   good as making it easy.  With all the other arguments in favor of default-
   ing to private replies, it makes more sense to leave the headers alone and
   insert this:

   {#} Replies are directed the author at
   {#} To reply to the list, write to

4. Headers and footers don't work very well.  We've had similar discussions
   on list-managers about putting the submission and administrative addresses
   into a header or a footer of every item, and we all know the results: the
   members who need those reminders are the most likely to ignore them.  A
   large contingent worship their reply functions as omniscient gods, will
   ignore or forget what they saw in the text, will slam their `r' keys out
   of mindless habit, and will neglect or forget to change the reply address.
   Notice I spoke of using the keyboard: those who use pictorial interfaces
   are MORE prone yet to accept with blind faith whatever happens when they
   click on their reply icons and tend to be extremely reluctant to second-
   guess their infallible computers.

However, I do thank you and Gabe for that idea; until now, for those list
members who opt for default public replies despite my discouragement, I've
been saving any original Reply-To: header as an "Author-Reply-To:" header.
I may move the information to the body, where readers will have a slightly
greater chance of registering it.

R>   o  It penalizes the person with a reasonable mailer in order to
R>      coddle those running brain-dead software.

S> I would argue that penalization is small to zero, where the win for
S> letting members with brain-dead software participate can be a big
S> win for the right group.

What big win?  What *little* win?  Members can participate only if they can
write where they mean to write, and as I illustrated above it's far easier
to override a default private reply when you want to answer in public than to
override a default public reply when you want to answer in private.  Getting
your private mail blared to the world is what I call a penalty, and there is
no win at all.  Sending personal mail for one other member out to the entire
list isn't participation; it's disruption.

R>   o  It removes important information, which can make it impossible
R>      to get back to the message sender.

S> Note in the gt-pfrc example above, the important data is right there
S> at your fingertips.

The address is far from your fingertips; you have to memorize it or write it
down to have it available when you redirect your response.  And are those
the _important_ data?  Read on:

S> Gabe did it right.

OK, suppose there are special-case reasons for gt-pfrc to default to public
replies: again, one question is still unanswered.  You didn't say whether, if
an incoming submission has its own Reply-To: header, Gabe uses it in the body
or ignores it and always repeats From: in the body.  If he uses the From:
header when there is an incoming Reply-To:, he's doing it very, very wrong.

S> The group as a whole wished for the change.  ...

They were unanimous -- no dissensions, no abstentions?  Groups never function
as wholes.  That's why I made it an option.

S> My summary -- if it's what the members want and is done with care
S> and caution, it works fine.  I agree with Chip that in *most* cases,
S> automatic `Reply-To: list' is a bad idea, but there are definately
S> times when it is the right thing to do.

My summary: [1] the lists where it is wrong but done regardless definitely
outnumber those where it is right and done accordingly; [2] there is never
a monolithic preference in any mailing list with two or more members; [3] it
is far easier to respond publicly when the default is private (even without
a group-reply facility in your mailer) than to respond privately when the
default is public; and [4] it is easy to fix having sent a reply privately
when you meant to send it publicly but _impossible_ to undo having sent it
publicly when you wanted to send it privately.

S> Correlary: ...

That's "corollary".  You might be confusing it with "correlation".

This is already very long, and I'm done responding to Steve Simmons, so I'll
send the rest of my comments on the matter separately.

Indexed By Date Previous: Re: Appending "Reply-To" in Outgoing Mails
From: Gess Shankar <>
Next: Where to get help for LISTPROC
From: Dick Moores <>
Indexed By Thread Previous: Re: Appending "Reply-To" in Outgoing Mails
From: (Steve Simmons)
Next: Re: Clobbering "Reply-To" in Outgoing Mail
From: Chip Rosenthal <>

Search Internet Search