[Swfdec] shall i write to the list or to the person who answered?

[Dennis Heuer]

On Thu, 01 May 2008 16:17:22 -0400
Pavel Roskin <proski at gnu.org> wrote:

> "Reply-To" Munging Considered Harmful:
> http://www.unicom.com/pw/reply-to-harmful.html
> 
> Not everyone agrees, but I do.

Please read this nonsense again. A five-years old child can read out
the "illogic" in his statements. The problem of this guy is that he
can't differ between the general case--for which the mailinglist is
made--and the special case--which appears to him and is to be managed
by himself. His anti-arguments (drawing panic with terms like
"dangerous" and "harmful", though there's no security issue evolved)
are based on cases that cannot appear if not he himself does something
fully wrong. But, to save him from that, he expects the mailinglist to
go paranoid and break the base of it's reason of existance.

There's also no sense in putting a secondary *feature* onto the
*primary* reply-to button. This is there for the general case and for
everybody. It is not there for people who want to define weird
communication-chains/spreads via a general communication media.
Actually, those headers belong to be filtered. This is abuse of
mailinglists!

If one wants to write to an author privately, one can click on his
address, which is displayed by the email. So easy it is. I don't know
if "Elm" can do this. However, the special case is to be managed by
oneself; it is not a problem of the mailinglist and nothing to be done
primarily.

The same if a sender wants the replies to be sent to someone else. This
is nothing the mailinglist should care about. The mailinglist sends to
and receives from *members*, and nothing else it should support. Also,
if the sender is not the original author or just writes for a friend to 
send replies to, he shall type this at the top of the message that every
reader is informed immediatly about where his answers shall go to.
It is the reader who decides if he wants to answer to there or to the
mailinglist or to both, not the author. This is a matter of respect and
not of the mailinglist. The client can help by supporting a feature to
create a reply-to message to the clicked address in the text. This is a
helpful *feature* but nothing else.

The really weird in his arguing is that he states that every mail one
receives via a mailinglist represents some private communication and
that, in contrast to this, the reply to the mailinglist is something
special to choose extra via the group-reply button. However, this is
plain wrong. And, this causes every mail to be sent twice to the
author--at least with normal configured mailinglists.

> Please use "reply to all".  It's the safest approach, as it will include
> the persons who are not subscribed.

If I'm not subscribed, I don't expect replies (What a silly idea! If
you cancel your newspaper subscription, do you still expect to receive
the sports part?) If I still want to see possible replies, I watch the
archive. This is nothing the mailinglist is responsible for. Here's
again the real misunderstanding. Your private way is not of interest to
a general-purpose mass-communication media.

I agree that programmers don't want to stay logged in to high-volume
lists. They'd rather like to get mails to their threads while logged
off. However, this is a *wish* for which there can be a *feature*. This
is not what mailinglists were defined for and first brought in with
threading, which allows to organize sub-communication within the
general stream, which is a critical overload of the principle and would
rather make an IMAP account at the server mandatory. This isn't offered
but no excuse for the abuse of the reply-to button. It's rather a sign
that there should be more than one mailinglist for the project.

Otherwise, the mailinglist server can provide an option to generally
reply to all authors of mails in the same thread--no matter if they are
still registered or not (their addresses are still remembered in the
thread's messages). Bugzilla--though not a mailinglist server--does so,
for example. The author of the bug message always gets the reply and
others can register to get it. But that's a feature and not a must; and
it is, again, done so in general and not by case. The individual case
is not of matter to the mailinglist, bug tracker or whatever other
general-purpose service.

Last but not least, his example of people wishing a third button is not
an example of the failure of the common configuration of mailinglists
but just a *feature* people would like to have *by case*. It is useful
but nothing mandatory. It doesn't belong onto the *primary* reply-to
button, as he states.

This guy rather breaks the idea of a mailinglist and overcomplifies
the general communication for some occuring cases than that he has
anything senseful to say. I refuse to follow this.

And, if you switch over to private communication, this communication is
lost for the server's archive. Why the hell should this be the default?
This could make sense in the case of a sex-agency but not in the case
of a mailinglist. However, a sex-agency will not have a public archive.
So easy it is!

Regards,
Dennis