[Freeciv-Dev] Re: Reply-to-header [was: MP3 Patent Issues]

[Jules Bean <jmlb2@xxxxxxxxxxxxxxxx>]

On Sun, Jul 02, 2000 at 05:48:59PM +0200, Mathias Hasselmann wrote:
> Jules Bean wrote:
> > 
> > On Sun, Jul 02, 2000 at 04:10:36PM +0200, Mathias Hasselmann wrote:
> > >
> > > But NO! The list manager does not sets the reply-to header -
> > > as other list managers do! This a really bad behaviour, since
> > > you can't simply reply postings and mail user agents have
> > > nearly no chance to track threads.
> > 
> > First, do a web search on "reply-to considered harmful".  Once you've
> > read, understood, and thought about that information, 
> 
> First: When is the sender of a mailing list message different
> from it's author? I would say this argument only aplies, when the
> message is generated automaticlly and indicates that the mail server
> of the site which created the message is configured badly (ever heard
> about mail aliases) or when the message is spam. Both types of
> messages are messages *I* do not want to replay when appearing
> on mailing lists...
> 
> Second: Group reply makes more work: I do not want to send the
> person I replied two messages. Therefore I have to remove the
> senders mail address from the header and I (probably) have to
> change the mailing list from a carbon copy reciptent to the
> main reciptent.

Many people who participate on mailing lists are more than happy to be
cc:ed on replies. I tend to, as I am doing now, group reply to mailing
list messages (I participate on many very high volume lists, and on
those lists, authors can miss messages they may be interested in if
they are busy).  An author who does not wish to be cc:ed may cause his
MUA to emit a Mail-Followup-To: header which will send followups to
the list at least in the case that the other party is using an MUA
which respects the header (mutt does).

> 
> Third: Privacy. That's an argument. Private response to the
> themes discussed here are quiet uncommon. We normally do not
> speak about sex or politics on this list. And if we do it's
> meant public - as in the messages I've send privatly by
> accident.

But the other side of that argument is that I have the choice to reply
privately ('r', in mutt) or publically ('g' in mutt).  

> 
> But a least: If you really are convinced by this policy - in
> my opinion not usefull for this type of mailing list - it is
> no problem for me, to install a filter on my mail server ;-)

I'm not adminning this list. I'm just pointing out that this is an old
argument long gone over.

Personally I find the most convincing argument that there is a common
and correct use of 'Reply-To' when the author for some reason does not
want replies at the 'From' address (e.g. he doesn't read mail there so
often), and this use would be stamped upon in the mailing lists
rewrites that header.

> 
> > Also discover the
> > 'mail-followup-to' header and find out whether your email client
> > supports that.
> 
> This mailing list's massages contain only the headers created by
> the sender's MUA and by the involved MTAs . There are no group
> headers as 'mail-followup-to'.

Indeed.  But you can emit Mail-Followup-To headers if personal Cc:es
annoy you.

-- 
Jules Bean                          |        Any sufficiently advanced 
jules@{debian.org,jellybean.co.uk}  |  technology is indistinguishable
jmlb2@xxxxxxxxxxxxxxxx              |               from a perl script