Reply to list vs. reply to sender?

[Sam Livingston-Gray]

Hey, all-

Maybe it's just me, but there seems to be an uptick in replies that
might not have been intended to go to the entire list of 800+ people.

Is it time to revisit the list setting that changes the "Reply-To"
header? I know I've certainly appreciated being able to be lazy by
default, but it does seem to generate more noise—and, potentially, a
great deal of embarrassment. (Anybody want to go back and count the
number of replies to job postings in this year alone?)

For more arguments against our current setting, please see:
'"Reply-To" Munging Considered Harmful'
http://www.unicom.com/pw/reply-to-harmful.html

-Sam

[Jonan Scheffler]

For what it's worth I vote to turn it off. Somebody is likely to get fired for it eventually.

--
You received this message because you are subscribed to the Google Groups "pdxruby" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pdxruby+u...@googlegroups.com.
To post to this group, send email to pdx...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Jacob Helwig]

I can understand how it's useful to some people, but I do think that
the failure mode for forced Reply-To is more harmful than it's worth.
+1 for turning Reply-To munging off.

> --
> You received this message because you are subscribed to the Google Groups "pdxruby" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pdxruby+u...@googlegroups.com.
> To post to this group, send email to pdx...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

--
Jacob Helwig
http://technosorcery.net/about/me

[John Wilger]

I'm more than a little amused by the fact that first I get Sam to use
vim more often than TextMate and now he's even coming around to my
point of view on reply-to munging. :-D

> --
> You received this message because you are subscribed to the Google Groups "pdxruby" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pdxruby+u...@googlegroups.com.
> To post to this group, send email to pdx...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>



--

Regards,
John Wilger

johnw...@gmail.com
971-678-0999

[markus]

> Maybe it's just me, but there seems to be an uptick in replies that
> might not have been intended to go to the entire list of 800+ people.

What we need is a way to make it go out to just a random subset. :)

-- M

[Sam Livingston-Gray]

The poll I sent out got 44 responses. 73% were in favor of changing
the setting so that replies go to the original sender by default.

There were a few comments; I'd like to address one in particular:

"It's a mailing list. The default setting of the tool should
facilitate further participation in list threads. Preventing the
potential missteps of a few people should not be paid for with
continual minor inconveniences to people who want to use the tool for
its intended purpose."

While I definitely appreciate this, we've reached a size where we
can't assume that everyone shares basic assumptions about the culture
of the group or the behavior of the mailing list.

As one of several list administrators, I see a fair bit of email that
never makes it to the rest of you. Some of it is recruiters
advertising, say, a .NET job in New Jersey; I apply the banhammer.
Some of it is job posts without the literal string "[JOB]" in the
title; I send them a note asking them to resubmit with a better
subject (man, I wish Google Groups would let me edit subjects,
especially when it includes "(JOBS)" instead of "[JOB]"). ...And this
morning, it was a new list member (still sporting a shiny new
moderation bit) replying to a job post.

While my inner Spock is totally down with "the needs of the many
outweigh the needs of the few or the one," he also points out that
while the raw numbers support "leave it alone," the coefficients on
those terms can easily throw the balance the other way. As another
comment read (in part), "fail quiet is safer than fail noisy." And,
as Jonan pointed out, an accidental reply-to-list-to-job-posting can
have really really bad consequences.

Does anybody happen to know whether the difference in setting is as
simple as clicking "reply" vs. "reply all"?

-Sam

[Tyler Hunt]

On May 24, 2013, at 9:48 AM, Sam Livingston-Gray <gee...@gmail.com> wrote:

> The poll I sent out got 44 responses. 73% were in favor of changing
> the setting so that replies go to the original sender by default.

Another solution here is to split off the job postings into a separate pdxruby-jobs list. Since the majority of the accidental reply-to-list emails are in response to job postings, this would allow that list to have the Reply-To header set to the sender, while keeping it set to the list for the main pdxruby list. There may also be other benefits from a moderation standpoint.

Tyler

[Bill Burcham]

+1 on defaulting to reply-to-sender only. What follows is a related idea I've been experimenting with…

B

​ack in the days of Usenet, I used to follow my favorite groups (like comp.lang.c++)​ via my Emacs newsreader: gnus. I'd usually go into gnus a couple times a day. Did Usenet even provide an email gateway? I can't remember. But I can tell you I certainly never used it.

I've been experimenting over the past few months, with turning off all email "notifications", opting instead to visit a fairly small set of places periodically. So my email inbox(es) are just one place I visit. Google groups is another. Other examples: Calendar, (private) Google Sites, Google+, Pivotal Tracker, Asana, Engine Yard dashboard.

​There is definitely a tradeoff here. I have to be disciplined to visit those places periodically. In return, I am using, in each case, a tailored interface. The alternative is cramming notifications for everything into the (email) inbox.​ When we cram things into the inbox that are not really email, we suffer from the interface incompatibility. This whole reply-to-sender versus reply-to-all thing is one example of that. The other symptom of interface incompatibility is this workflow: I see a message in my inbox; I click on a link in the message and do something in another program; then I come back to my inbox and delete the original message. Do that about 100 times in a day and it starts to feel like a waste of time.

I can't say that my experiment has been a slam-dunk success. But I think on balance it's a better way to work.  Originating this message in https://groups.google.com/forum/?fromgroups#!topic/pdxruby/g1hdPzKIoxs it's very clear to me that Sam and pdxruby are to be "ccd" on my reply.