Comments in-line.
-----Original Message-----
From: ietf [mailto:ietf-bounces(_at_)ietf(_dot_)org] On Behalf Of Hector
Santos
Sent: Monday, July 21, 2014 4:24 PM
To: ietf(_at_)ietf(_dot_)org
Subject: Re: DMARC and ietf.org
Mike,
There is no "pretending" here. We actually IMPLEMENTED and DEPLOYED the
consensus built MAILING LIST recommendations and it works.
The fact that you implemented and deployed does not mean that there is a
general consensus in the IETF sense.
So I disagree 100% with the erroneous suggestion there has been "no
consensus at all." To suggest there is no guidelines whatsoever has been the
real disservice being promoted. Its not true.
Guidleines != consensus.
It doesn't matter if its DMARC or ADSP. It the same design guidelines and if
you actually implemented it as list developer, you might see that its really
that simple.
The devil is always in the details. ADSP!=DMARC.
The old argument that List developers are too old to change doesn't wash
anymore and in reality, once you roll up your sleeves and implement the
consensus built suggestions, you will see it really has nothing to do with
list
services. It has to do with the VERIFIER.
Even John isn't arguing that list developers are too old to change. He is
questioning whether there is a consensus as to what changes or sets of changes
are appropriate. The Verifier is the driver of the changes but even
representatives of (some) large mailbox providers have indicated that as
Verifiers they are interested in how problems in this space might be addressed.
As a 3rd party mailer, I got my organization to make changes in 2007 that
accommodate issues in this space. While I think the changes we made are the
optimal long term solution (including for MLMs), I recognize that there IS NOT
A CONSENSUS and that others have a preference for other approaches. It may be
that the community settles on a single "best" approach or that it recognizes
several alternative approaches.
So in short, I take slight offense to your suggestion that I have no
understanding of the total issues involved as a product developer, its product
offerings and also addressing the support needs of its customers which
represents a wide horizontal spectrum of applied list needs. There are
solutions and I speak as a developer of a commercial integrated mail list
server product line:
I never said that you don't have an understanding of the issues involved. I did
say that you are incorrect when you claim that there is a community consensus
on how mailing lists should address problems in the context of DMARC assertions
by domains. I do not see a rush of others posting to support your assertion and
I do see evidence (previous discussions on this very list within the last 4
months) that contradict your assertion.
http://www.santronics.com/products/winserver/ListServe.php
Note. This has nothing to do we have a "big data" problem (how to scale
signer authorization). Its a serious problem. But the consensus built
guidelines provided are solid and necessary for any solution development.
You still need to honor the policies at the mail entry level.
--
HLS
On 7/21/2014 3:18 PM, MH Michael Hammer (5304) wrote:
John is correct. There is no consensus on how mailing lists should deal with
DMARC problems, notwithstanding what rfc6377 says about DKIM. ADSP
never gained enough real world implementation for there to be a meaningful
consensus. One need only look at the discussion threads on the IETF (and
other) list(s) following the publication of DMARC p=reject by several large
mailbox providers to see the diverse range of views.
While I disagree with John on some things, in this case he is 100% dead on.
To pretend otherwise is to do a disservice to the mailing list community and
the mail community at large.
Mike
-----Original Message-----
From: ietf [mailto:ietf-bounces(_at_)ietf(_dot_)org] On Behalf Of Hector
Santos
Sent: Monday, July 21, 2014 3:10 PM
To: ietf(_at_)ietf(_dot_)org
Subject: Re: DMARC and ietf.org
On 7/20/2014 10:51 PM, John Levine wrote:
I thought the preferred solution was to rewrite the From for those
users only.
I think that remains controversial. ...
There is no consensus at all on how mailing lists should deal with
DMARC problems.
Not quite John.
The specific DMARC protocol aside, with any author domain policies in
general, whether it was SSP, ADSP or any DKIM author domain signing
authorization protocol (DSAP), there was a consensus RFC built document
that provided the basic guideline for mailing list operations in dealing
with
restrictive DKIM signing policies. It used ADSP as the "DSAP" of the day.
But
replace ADSP with DMARC and the design recommendations apply:
RFC6377 DomainKeys Identified Mail (DKIM) and Mailing Lists
http://tools.ietf.org/html/rfc6377
And overall, the basic guideline was to support the framework, not ignore
it
as it never existed and instead pushed for breaking the security protocol.
As a LIST developer and implementor of the "DSAP" protocol, it was
simple:
1) Deny Restrictive Domains from Subscribing
2) Deny Restrictive Domains from List Submission
3) Pottery Principle "You break it, you own it" - Resign mail
That is all at the top level that needed to be done and all the above
really
has
nothing to do with a mailing list but the mail receiver verifier and the
outbound mail server.
This is about not wanting to do a basic author domain signature
authorization
lookup for any kind of mail service.
--
HLS