[ietf-dkim] Are lookalike domains like parent domains?

[Steve Atkins]

On May 1, 2008, at 5:16 AM, Eliot Lear wrote:

> Steve,
>> The discussion is not over what recipient behavior in that
>> case will actually be (reject or discard it because it's
>> claiming to come from someone who doesn't exist), rather
>> it's over whether the ADSP spec needs to prescribe that
>> behavior.
>>
>>
>
> Then we are (once again) counting angels on the head of a pin.

Yup. I was summarizing where I think the group is at rather than
expressing much of an opinion.

> Many people will read this specification and attempt to discern  
> behavior for, if nothing else, diagnostic purposes.


> Nobody has answered the basic question of what harm would be caused  
> with the text being left in.

It makes the spec significantly less clean, as it stops being about  
anything DKIM related and starts redefining how SMTP works. For a BCP  
that wouldn't be a problem (as the practices it's suggesting are, IMO,  
"best" in at least that specific case) but stomping on the toes of  
unrelated standards is an issue for a standard.

>  And everyone agrees it's the right thing to do.


If you mean "reject mail from people who don't exists', yes, I think  
they do. If you mean "include that requirement in a standards spec"  
it's a bit more grey. I wouldn't be satisfied if it wasn't mentioned  
at all, but I'd be leary of anything as strong as a MUST in there too.

I think the NXDOMAIN (or "check for A or MX", if you prefer) issue is  
being caught in the crossfire of the treewalking proposal, which has  
more serious issues.

Cheers,
   Steve