Re: dmarc damage, was gmail users read on... [bozo subtopic]

[John C Klensin <[e-mail]>]

--On Thursday, September 11, 2014 01:25 +0000 John Levine
<[e-mail]> wrote:

>...
> The one that's most widely used rewrites the From: line to put
> the list's address in place of the author's.  It "works" in
> the sense that it avoids DMARC rejections, but at the cost of
> screwing up the mailing lists so you can't tell who wrote what
> from the usual MUA display, and in many cases, you can't tell
> who wrote a message at all unless you put the author's address
> as the Reply-To, which has its own well known set of problems.

FWIW (and with the understanding that you don't need
convincing), I find that approach really objectionable.   In
addition to the pragmatic reasons you cite (and some others such
as the ability to prioritize mail based on origin), we've been
careful, almost since the dawn of network mail, to associate
what now appears as "From:" with the actual originator of the
message.  We've invented "Sender" and "Resent-" header fields to
preserve that distinction and make it clear.  For a mail
exploder to violate that principle, especially to make an
ill-designed protocol work better, seems problematic to me.
For the IETF, it is bad news when we cannot or will not adopt.
use, and conform to one of our own established standards-track
protocols.  But, if we have a system that does conform and we
switch it to not conform in order to accommodate an unfortunate
practice or design, that would be far worse, putting us in must
the same position we put ITU in when we pointed out that they
considered X.400 so stable, useful, and well-implemented that
they were using SMTP.

    john