Re: [dns-privacy] WG Call for Adoption: draft-pauly-dprive-oblivious-doh

[Wes Hardaker <[e-mail]>]

"Martin Thomson" <[e-mail]> writes:

TL;DR: don't adopt

> I would question the value of publishing the experimental
> approximately-but-not-quite-O-HTTP version then.  Unless we fail
> majestically in chartering and executing there, we're not talking
> about significant delays.

It seems silly to me to write up a document that is decoupled from the
parallel O-HTTP work when it would be better off depending on the
results of that work.  Otherwise we'll have potentially two different
protocols that are subtly different enough to increase code complexity
in stacks trying to offer support for both.  I'm not sure where that
intersection will exist (browsers?), but I doubt it's zero
implementations.

IMHO, ODoH is a useful goal and protocol but should be a small document
describing the additions needed beyond the eventual O-HTTP.

If the current proposal does get adopted, I'd argue for experimental
being a much better track.  I strongly doubt, without evidence, that
this will be the final solution to this newly targeted problem.

------

(as proof that I'm not opposed to the technology proposed in general:
https://datatracker.ietf.org/doc/draft-hardaker-dnse-split-key-dns/ from
2014)

-- 
Wes Hardaker                                     
USC/ISI