Re: ignoring the CoC in regards to cc:s (Re: Can we ship sources of a PDF file in the Debian diff?

[Miles Bader <miles@gnu.org>]

Ben Finney <ben+debian@benfinney.id.au> writes:
>> You're arguing that a Reply-To header is "harmful" (not that I am
>> convinced)
>
> That field is very useful. What's harmful is mailing-list software
> munging that field, which is for the author to set and for nothing else
> to fiddle with.

Yup.   Reply-To is for the _original sender's_ use!

If mailing list software were to start setting Reply-To, what is it
supposed to do if it gets a message with Reply-To already set (by the
original sender)?  It could (1) overwrite the original Reply-To header,
breaking personal replies to the sender, or it could (2) refrain from
setting Reply-To for such messages, completely confusing the readers who
have become accustomed to depending on the mailing-list's setting.

I think there's no perfect solution to the general problem, because
there's too wide a variety of MUAs in use, which support different
feature sets.  But it's much better to get "duplicate" messages in some
cases than to break things in a way that leads to _lost_ messages.

My experience is that in practice, it's not such a huge problem anyway;
a combination of MUA list-followup commands + Mail-Followup-To: headers
+ MUA duplicate suppression seems to keep duplicates in check reasonably
well...

-Miles

-- 
If you can't beat them, arrange to have them beaten.  [George Carlin]