This Mailinglist

[Eric Dunbar]

On Mon, 10 Jan 2005 22:31:22 +0100, Simon Ekstrand
<lists at routemeister.net> wrote:
> On Mon, Jan 10, 2005 at 03:03:08PM -0500, Eric Dunbar wrote:
> > On Mon, 10 Jan 2005 09:59:46 +1000, CB <ubuntu-users at crispin.cb-ss.net> wrote:
> > > Simon Ekstrand wrote:
> > >
> > > >My only reply in this semi-religious thread.
> > > >http://www.unicom.com/pw/reply-to-harmful.html
> >
> > Some not-so-very-convincing arguments!
> 
> I disagree, I'm sure you will also once you've sent enough private replies
> onto lists due to reply-to being set.

And think of the number of posts that are _lost_ to lists because
people use REPLY (as the VAST majority of e-mail users do) and that
means the list loses the input! If you're subscribed to a list by
DEFAULT all posts are public, THAT'S WHAT A LIST IS FOR.

> > Having skimmed the rest of the posts it seems the majority of people
> > are in support of the SENSIBLE (given what the overwhelming number of
> > lists are) reply-to-list default.
> 
> Huh? I see the exact opposite. Out of the 31 lists I'm currently
> subscribed to exactly two force a reply-to header.

Granted, I agree the idea of having a reply-to-list button in your
mailer is a nifty feature, but this requires that you enforce strict
mailer compliance and have a limited, (techno-geek) expert audience
(as I'm sure your lists tend to be... no offence intended).

> > PS Simon Ekstrand, how come your e-mail results in reply-to-list as
> > being the default whereas the other posts require reply-to-all to post
> > to the list (and clutter individuals' mailboxes with extraneous
> > e-mails)?
> 
> I have no idea, maybe your mailer was more convinced by the arguments
> than you were. :)

I was pretty convinced by his arguments too but I guess my e-mailer is
even more disgruntled with reply-to-all than I am ;P.

> And those posts that 'clutter individuals' mailboxes with extraneous
> e-mails' are quite valuable on high latency lists, they enable people to
> keep conversations running quickly even with list delays of several
> hours. Procmail is your friend, it can filter out duplicates. But that's
> a completely different discussion and one that shouldn't be brought up
> again.

Potato, potato! I find them extremely annoying. I don't WANT to have
those posts appear in my in-box and that's what happens if someone
replies to me directly as well. List traffic is separate from personal
communications -- which is why I subscribe to LISTS and don't carry on
a million and one personal conversations.

I suggest you get a real e-mailer, gmail does everything youc ould wan
and then some <fllrp> (tongue in cheek). Seriously, although gmail is
a very functional e-mailer (the way it handles threads is stunning in
its elegance and effectiveness), it isn't the only game in town but it
does mimic what most people use. In most cases reply (control-r,
command-r, whatever) is what people use.

And, to be honest, I don't have the time or the inclination to do a
systematic survey of what the reply-to behaviour is on lists out
there, but I can UNEQUIVOCALLY state that the OVERWHELMING number of
lists to which I am and have been subscribed have the list posting
address as the reply-to address. The ONE exception to this rule is a
swap (for sale/wanted) list which I track and, in that case it's
OBVIOUS that the overwhelming number of posts should go to the
individual! Occasionally there are posts that inadvertently make it to
the list but they are far and few between. Discussion lists are a
different breed of fish (or whatever the saying is). On discussion
lists the overwhelming majority of traffic is destined to be PUBLIC
(ergo, they are discussion lists ;).

Anyway, let's have a SOLICITED vote on the subject (i.e. send out
e-mails to those who have posted for e.g. (since they are the ones
affected since they do post)). I have a strong dislike for polls as
are the norm on the web (and in pop culture where people tend not to
know their butt from a hole in the ground re: statistics) where
participation is driven by participants, and results are skewed by the
most opinionated participants (and thus the most likely to
participate... which is why democratic contests where only 50 or 60%
participate are fundamentally undemocratic) or by small but well
coordinated efforts (which is why web polls are generally 100%
meaningless since you'll have multiple votes by one individual or her
friends).