[ietf-dkim] Re: New Issue: Problems with Scenario 4: Resent

[Michael Thomas]

Dave Crocker wrote:

>
>
>>>>>>> Whatever SSP does (and the more interesting case is a "Bob"
>>>>>>> who is completely DKIM-unaware), the mail should not be rejected 
>>>>>>> by the next receiver(s) ...
>>>>>>
>>
>> Dave, I'm completely confused. The constraint of requirement 12 is a 
>> constraint on the protocol design in the form of "don't provide 
>> this". It's not telling the receiver to do anything.
>
>
>
> Definitely a good distinction to make clear.
>
> Unfortunately, I do not see how the language "the mail should not be
> rejected" is a constraint on protocol design, only.  It strikes me as an
> explicit directive to an operational receiver.

I'm fairly certain that we're in violent agreement here, with the likely
confusion being in Frank's phrasing ("should not be rejected" was his
words, not the draft's). The draft definetely doesn't phrase this in terms
of the receiver doing anything. Indeed, the draft's requirement here was
intended to reinforce that the protocol is an information service, not a
"Dear Abby" advice line.

       Mike