Re: [dmarc-ietf] Revisiting the Race Condition in draft-crocker-dmarc-sender-01

[Dotzero <[e-mail]>]

On Wednesday, August 19, 2020, John R Levine <[e-mail]> wrote:

> On Wed, 19 Aug 2020, Dotzero wrote:
>
>> Then Ericcson as an organization has made a decision regardless of the
>> objections of those employees. The correct thing for Ericcson as an
>> organization to do is to publish an internal policy that employees should
>> not use company mail for participating in mailing lists. An alternative to
>> that would be for them to hire someone to help them come up with a
>> workable
>> approach. We both know plenty of people who could help them.
>>
>
> No doubt, but they're not going to do that.  They want their employees to
> work on the IETF, the publish p=reject and they apparenty believe the
> contradiction is not their problem.  So we're stuck with it.


For some definition of stuck. I happen to believe that validators/receivers
SHOULD generally respect policy assertions. Mail list operators should as
well. Absent a functional and meaningful mechanism for authorization of
specific intermediaries by domains, that is the way to go. If things break
and users complain then those users should be told to contact their mail
administrator. Just saying.

>
> Of course you personally know what DMARC policies mean and what they
>>> imply.  But every time a list has to rewrite a From line, we have
>>> evidence
>>> that someone else doesn't or at best doesn't care.*
>>>
>>
>> I'm going to guess that more often than not it is the latter. *I heard the
>> same. Too big to care?
>>
>
> Yup.  They want phishes, or in Yahoo's case expensive user complaints, to
> go away, they don't care about discussion lists one way or the other
>

Perhaps they would care if users/customers complained and or walked. Again,
just saying.

Michael Hammer