[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Why does reply not reply to the list?

From: Quinn Taylor <quinntaylor_at_mac.com>
Date: Thu, 7 Aug 2008 21:09:50 -0600

PREFACE: Sorry for feeding the trolls...

On Aug 7, 2008, at 5:02 AM, Colin Alston wrote:

> Quinn Taylor wrote:
>> I'm on many mailing lists, and this is the only one I remember that
>> doesn't have a Reply-To header that sends replies back to the list
>> email.
> Those mailing lists are doing it wrong.
>> Is there a rationale for this, or has it just not been set up? I'd
>> much rather that replies go to the list by default.
> The fundamental similarity between people who request Reply-To
> munging is that they are terrible at operating their email client. I
> still have hopes for the human race, and part of that hope is that
> people who are capable of joining a list as well as operating
> subversion are competent at the simplest task of pressing "Reply
> All" (the button on Thunderbird). This is similar to people who
> complain about heated debates "filling their inbox with junk"
> because they are incapable of setting up mail filters and folders.
> Feel free to read
> http://www.unicom.com/pw/reply-to-harmful.html
> --
> Colin Alston <colin_at_thusa.co.za>

Both that article and its rebuttal talk about "The Principle of Least
___". Anyone in favor of "The Principle of Least Pretension"?

In my original post, I asked for rationale (politely, I thought),
which was provided satisfactorily. I stated my preference, but didn't
blatantly state that this list's policies were wrong. Both articles
cited in the FAQ are somewhat subjective and partially right. Folks
who are adamant about certain behavior being "wrong" would do well to
distinguish between inherent superiority and dogmatism.

For example, this article states that "Reply-To munging destroys the
'reply-to-author' capability", but this is untrue. What is meant is
that the normal behavior changes, in that if you wanted to reply,
you'd have to replace the group email with the original author's
address. Hmm, guess what we on the other side must do when the Reply-
To is *not* munged, and we want to avoid duplicate messages?
Similarly, accusing list admins who decide to munge this header of
being arrogant is ridiculous. More likely than not, they're serving
the wishes of the list members. (Forcing unpopular policies on list
members who dislike them would be far more arrogant.)

I know perfectly well where "Reply All" command is in my email client,
but the habit of pressing Cmd-R for a normal Reply is deeply
ingrained. I'm trying to break the habit exclusively for this
list. :-/ Calling out users who prefer a different behavior and
assuming "that they are terrible at operating their email client" is
similarly hypocritical; if the tables were turned, you'd complain
about having to retype a private email address. "Brain dead" is in the
eye of the beholder, and dependent on the context. For example, I
suppose we could choose to mock those whose clients' copy-paste
functionality aren't helpful enough to prevent re-typing addresses
that are already in the message... ;-)

Personally, I like Mail.app, while others like Outlook, Thunderbird,
Eudora, pine, mutt, elm, etc. The fact that your mail client works the
way you want when a list is set up the way you like does not make all
other options categorically wrong. And let's be honest, using only
examples from text-based mailers is not scientifically convincing
evidence that munging makes all users miserable.

>> I don't think I've ever seen the topic even mentioned on the lists
>> that
>> set the reply-to back to the list - it just does what people
>> expect. I
>> can understand the concept that the default should be private if the
>> list topic involved things of a personal nature. It doesn't make a
>> lot
>> of sense on a technical list where public responses help everyone.
>> But,
>> as long as there are any lists that don't force the reply-to, the
>> canonical way to reply to list messages is going to be reply-all.
> Come to think of it you are right. I've never noticed a discussion
> about the reply-to if it goes back to the list. Both this list and
> the mailman list do not automatically go back to the list, so there is
> discussion every now and then. Both lists have web page sections
> regarding the reply-to issue. If the reply-to was to the list,
> neither section would be necessary :-)

Ditto. I don't remember ever seeing a "catastrophic failure mode" on a
munged list. The closest I've seen is when someone means to mail off-
list about a business deal or questions not of general interest. When
this happens, they are gently reminded that they're posting to
everyone; no lengthy explanations or debates ensue.

The examples of mistakenly-broadcast messages listed under "Principle
of Least Damage" in the article (boss' personal hygiene, roommate's
sex life, etc.) are ludicrous in the context of a technical email
list. Furthermore, adding a layer of protection to shield oneself from
sending a damaging email to unintended parties is not the best
solution. Sure it's cliché, but "If you can't say something nice...."
Your failure to control what you write and send it to the correct
parties should not be my problem. In fact, such behavior might be more
deserving of the description "Coddling the Brain-Dead, Penalizing the

If I ever set up a mailing list for my easily-offended extended
family, I'll be careful not to munge the Reply-To header. However, for
lists that benefit most from replies to the list, I'll continue to
feel strongly that munging is a good choice. You may feel otherwise,
but please, let's not get pretentious or dogmatic about right and
wrong, especially over an issue that obviously has two valid arguments.

  - Quinn Taylor

  • application/pkcs7-signature attachment: smime.p7s
Received on 2008-08-08 05:10:22 CEST

This is an archived mail posted to the Subversion Users mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.