[ietf-dkim] New issue: Upward query vs. wildcard publication

[Michael Thomas]

Dave Crocker wrote:
> 
> 
> Michael Thomas wrote:
>> Dave Crocker wrote:
>>>> 2) if you don't get a ssp rr, check to see if it gave
>>>>    you a NS or SOA authority records.
>>>
>>> Michael:  Zones are not part of the user-visible DNS semantics.  They 
>>> are strictly an administrative construct.  Using anything that relies 
>>> on particular zone points breaks the semantics of the DNS.
>>
>> Dave: I don't see any reference to "zone" in what you quoted, or in the
>> algorithm I layed out, or in any other message I posted about this,
>> so I have no idea what your objection is.
> 
> 
> SOA and NS are zone constructs.  They are not constructs applicable to 
> user-visible domain name semantics.  Rather they are part of the 
> underlying distribution service.
> 
> The two are, effectively, entirely separate layers of service.
> 
> For example, see:
> 
>    <http://asg.web.cmu.edu/rfc/rfc1912.html>
>    <http://support.microsoft.com/kb/163971>

I just read both of these and I don't see how it contradicts anything
I've proposed. If the authority of a domain is a parent to the current
domain, it seems rather pointless to query all of the intermediate
labels in a treewalk. In this particular case, keeping query complexity
down is a lot more important than lots and lots and lots of unneeded
flexibility.

		Mike