Re: Suggestion: can we test DMARC deployment with a mailing list?

> > We've been running that experiment for at least a year.  Surprise!
> Good to hear. Obviously not the area I’m looking at hardest.
>
> If we’re having the level of problems that seem to be being reported in 
> this thread, it would appear that we haven’t learned much from the 
> experiment. I take it that the draft Doug Otis mentions is part of the 
> mitigation discussion.
The problems are occuring at the end points, not at the IETF.  For 
example, aaron(_at_)aol(_dot_)com posts to a list, where one of the subscribers is 
charlie(_at_)comcast(_dot_)net.  The list adds a subject tag and footer, as our lists 
have done since forever, and remails it to Charlie.  Comcast's DMARC 
software observes that this message has an aol.com addresss in the From: 
line, but didn't come from an AOL IP host (SPF) or has it a valid aol.com 
DKIM signature, so Comcast bounces it.  This isn't hypothetical; I've seen 
exactly this in my logs.
Before anyone says "why don't we just ...", we've been arguing about this 
for a year, and any technical change to mailing lists won't avoid the 
DMARC problem, or will change the way that lists work, requiring that 
users retrain themselves and their local mail filters.  We didn't create 
this problem, and I don't think it's reasonable to ask us to bear the 
costs of fixing it.
What would work is for the systems that implement DMARC to whitelist 
senders who send legitimate mail that DMARC can't describe.  (Not total 
whitelist, just skip the DMARC bit.)  That could work, if they're willing 
to spend their own money to do it.
Regards,
John Levine, johnl(_at_)taugh(_dot_)com, Taughannock Networks, Trumansburg NY
Please consider the environment before reading this e-mail.

smime.p7s
Description: S/MIME Cryptographic Signature