Re: [Tools-discuss] [rt5.ietf.org #6435] Unable to get mail back from alias expansion - expand-mediaman-chairs

[Tero Kivinen <[e-mail]>]

Carsten Bormann writes:
> > 
> > I think that the IETF should have rejected all email from sites
> > with a p=reject policy. That would have made many of the problems
> > apparent to the senders.
> 
> The problem already is apparent.
> 
> Victim blaming is not the solution.

We are the victim, the dmarc policy people are the offenders... We are
not blaming the victims, we are blaming people who configure their
systems in a way which is incompatible with the way ietf mail systems
work. 

> Fixing the problem (doing DMARC rewrite workaround games) is the
> solution.

That is not really a solution, as it causes other problems. Changing
the both the from and envelope from address do allow emails to go
through, but that causes all kind of user interface issues and
usability problems. Using term "workaround games" just indicates that
you see that too...

Proper solution is to get rid of SPF, and make sure people use DKIM
and ARC, and configure the systems so that they do not reject emails
based on SPF etc failures, but just mark the email having unverifable
sender.

What we would need is to have better way of verifying ARC signers,
i.e., have a away of finding whether specific signer can be trusted
for certain email (not sure if that is problem that can be solved).

Rejecting emails because the SPF fails is just wrong. Marking it is
proper way to combat phishing.

> The only viable solution is not having the problem in the first
> place (i.e., not doing forwarding without rewriting). 

That is just replacing one set of issues with other set of issues. It
does somewhat work with mailing lists (where the sender is assumed to
be reachable by the sending reply to the list), it does not work
properly on the temporary aliases like draft aliases.
-- 
[e-mail]