Re: QUIC re-chartering: include DNS-over-QUIC?

[Paul Vixie <[e-mail]>]

Vidhi Goel wrote on 2020-02-06 18:26:
> I support DoQ for the use cases that are based on non HTTP application 
> protocol. DoQ implicitly provides benefits of QUIC which DoT doesn’t have.

i fear this may be off-topic for this mailing list; in which case, just 
tell me. i've been watching this thread carefully, and i think the 
concern about too many dns transports with no possible guidance to 
developers as to which one to try in what order. as a server operator i 
expect to have to support everything, which means arguments about threat 
surface in one transport vs. another are effectively null.

i can definitely see why the QUIC team might want more than one proof, 
it would help assure us all that there aren't HTTP assumptions designed 
in. and all of the reasons i never pushed for DNS over DTLS or DNS over 
SCTP are certainly addressed by DNS over QUIC.

however, my reasons for not pushing DNS over DTLS or DNS over SCTP are 
also addressed by DoT. i'd be grateful if you could explain what 
problems you think DoT has, which would beg us to invent another DNS 
transport profile such as DNS over QUIC.

i see DoQ as nec'y for QUIC's success but not relevant to DNS beyond the 
fact that DNS is a convenient second example of how QUIC can be used. if 
there are use cases or failure modes not handled well by DoT, i'd like 
to learn more about those.

-- 
P Vixie