Re: [Doh] WG Review: DNS Over HTTPS (doh)

[Tim Wicinski <[e-mail]>]

On 9/16/17 12:15 PM, Stephen Farrell wrote:
> 
> 
> On 16/09/17 17:07, Phillip Hallam-Baker wrote:
>> We had a re-run of the same issues with DPRIV which began with the
>> assertion that a solution must be found within a year.
> 
> I don't recall any such assertion. IIRC, DPRIVE was always
> considered as the start, within the IETF, of a marathon.
> (At least by anyone credible.)
> 
> Perhaps you can provide a pointer to that assertion?
> 
> Aside from that, I'm not clear who you think is being
> ignored with the current proposal, nor what you think
> we ought wait upon, so it'd help me understand your
> objections if you could clarify those aspects. (FWIW,
> as of now, I don't share your concerns in those respects
> at all.)
> 
> S.
> 

As DPRIV co-chair, I agree with Stephen's assertion that this was always 
going to be a marathon.  Now Phillip is correct in some respects - I 
felt *a* solution for the stub to recursive resolver piece could be 
worked out within a year.  But I've always felt the IETF was all about 
iteration: Let's come up with a solution; let's write some code and 
deploy some infrastructure; let's measure the behavior and the 
usefulness of what was done; and in the interim let's keep looking at 
other ideas.

The next steps within DPRIV are around the recursive resolvers talking 
to authoritative servers; and this is something Terry (our wonderful AD) 
and myself see as a much harder problem, and one most likely that will 
end in failure.

tim