[{"author": "Jonathan Morton", "text": "

@Meetecho: There is no camera on the presenter

", "time": "2022-11-10T13:08:17Z"}, {"author": "Jonathan Morton", "text": "

tks

", "time": "2022-11-10T13:08:59Z"}, {"author": "Vicky Risk", "text": "

can the speaker perhaps get closer to the mic?

", "time": "2022-11-10T13:15:55Z"}, {"author": "Vicky Risk", "text": "

yes, better, thx

", "time": "2022-11-10T13:16:27Z"}, {"author": "Benjamin Schwartz", "text": "

FYI, Chrome DoH has been using DoH/3 since launch, when used with Google Public DNS or Cloudflare

", "time": "2022-11-10T13:42:01Z"}, {"author": "Benjamin Schwartz", "text": "

So comparing do DoH/2 is perhaps not the most logical alternative

", "time": "2022-11-10T13:43:06Z"}, {"author": "Jireh Jam", "text": "

Please may i have the QR code for the paper

", "time": "2022-11-10T13:43:49Z"}, {"author": "Jireh Jam", "text": "

ok

", "time": "2022-11-10T13:48:03Z"}, {"author": "Mirja K\u00fchlewind", "text": "

slides are in the proceedings: https://datatracker.ietf.org/meeting/115/materials/slides-115-maprg-dns-privacy-with-speed-evaluating-dns-over-quic-and-its-impact-on-web-performance-00

", "time": "2022-11-10T13:49:11Z"}, {"author": "Dave Plonka", "text": "

Link to PDF herein: https://dl.acm.org/doi/10.1145/3517745.3561445

", "time": "2022-11-10T13:50:00Z"}, {"author": "Andrew Campling", "text": "

Others have reached similar conclusions about the performance benefits of DoQ to DoH See for example a presentation by Andrey Meshkov, CTO of AdGuard, at https://419.consulting/encrypted-dns/f/doq-in-the-wild

", "time": "2022-11-10T13:50:10Z"}, {"author": "Chris Box", "text": "

Just to expand on what I said at the mike: encrypted DNS has lots of potential to be highly performant, e.g. using 0-RTT QUIC with session resumption. Further data would be most helpful in understanding this.

", "time": "2022-11-10T13:50:10Z"}, {"author": "Tommy Pauly", "text": "

DoH3 with 0-RTT should have identical performance to DoQ, modulo a handful of bytes.

", "time": "2022-11-10T13:51:50Z"}, {"author": "Dave Plonka", "text": "

holding the queue so we can check for time when Francois finishes

", "time": "2022-11-10T13:57:22Z"}, {"author": "Chris Box", "text": "

It makes sense that from a client perspective, DoH3's ability to fall back to DoH2 is an advantage over DoQ, assuming a non-zero percentage of networks with UDP blocking. From a network's perspective where we have certainty that UDP is allowed, DoQ appears more attractive. Maybe DoH3 is the overall middle ground.

", "time": "2022-11-10T13:58:14Z"}, {"author": "Jonathan Morton", "text": "

I believe the term would be \"zenith\" for when the satellite is overhead

", "time": "2022-11-10T14:01:56Z"}, {"author": "Jonathan Morton", "text": "

\"apogee\" means \"furthest from the Earth\"

", "time": "2022-11-10T14:02:08Z"}, {"author": "Erik Nygren", "text": "

For me, DoQ makes sense primarily as the \"MTI\" for recursive-to-authoritative. DNS authorities aren't going to want to have to deploy HTTP stacks, and DoQ seems strictly better than DoT for many reasons if we had to pick one between them. DoH3/DoH2 as the primary mechanism clearly makes sense for client to DNS recursive. For OS resolver stub to DNS recursive, I can see more arguments in various directions (primarily to keep HTTP stacks out of the OS stub resolver).

", "time": "2022-11-10T14:03:39Z"}, {"author": "Tommy Pauly", "text": "

Yes, DoQ for authoritative makes great sense. But client access should be DoH3/DoH2. Our stub resolver has no issue doing HTTP there \u2014 you can have a very minimal HTTP stack for DoH.

", "time": "2022-11-10T14:05:30Z"}, {"author": "Dave Plonka", "text": "

apologies again; queue temporarily closed for time (in session) to be fair to remaining contributors

", "time": "2022-11-10T14:15:30Z"}, {"author": "Jonathan Morton", "text": "

@Meetecho: camera pan back to presenter pls

", "time": "2022-11-10T14:23:37Z"}, {"author": "Chris Box", "text": "

No doubt that IoT security is a large and growing problem.

", "time": "2022-11-10T14:37:55Z"}, {"author": "Robin Marx", "text": "

It bugs me so much that we don't applaud after presentations like this at the IETF. Is there a historical/cultural reason for this?

", "time": "2022-11-10T14:38:20Z"}, {"author": "Mirja K\u00fchlewind", "text": "

just do it (but we are a bit short on time); we also decide to rather leave the time for the presentations and questions then introducing speakers lengthly

", "time": "2022-11-10T14:44:59Z"}, {"author": "Mirja K\u00fchlewind", "text": "

but I should have said Gautam is one of the winners of the ANRP and he gave another nice talk on another topic in irtfopen

", "time": "2022-11-10T14:45:43Z"}, {"author": "Mirja K\u00fchlewind", "text": "

Risky BIZness: Risks Derived from Registrar Name Management
\n Gautam Akiwate
\nhttps://dl.acm.org/doi/10.1145/3487552.3487816

", "time": "2022-11-10T14:47:48Z"}, {"author": "Dave Plonka", "text": "

Thanks, Mirja!

", "time": "2022-11-10T14:48:07Z"}, {"author": "Mirja K\u00fchlewind", "text": "

And as we are talking about it. There is another measurement paper presented by another ANRP holder David Wagner yesterday in irtfopen:

", "time": "2022-11-10T14:49:33Z"}, {"author": "Mirja K\u00fchlewind", "text": "

United We Stand: Collaborative Detection and Mitigation of
\n Amplification DDoS Attacks at Scale\u201d
\n Daniel Wagner
\nhttps://dl.acm.org/doi/10.1145/3460120.3485385

", "time": "2022-11-10T14:49:34Z"}]