[{"author": "Lorenzo Miniero", "text": "We're 5 in the room now, we're getting warmer<br/>", "time": "2022-03-21T12:00:17Z"}, {"author": "Dedlyne Cloiseau", "text": "Hi everyone / Bonjour tout le monde / Greetings from Haiti<br/>", "time": "2022-03-21T12:00:49Z"}, {"author": "Murray Kucherawy", "text": "Can I do a quick audio test?<br/>", "time": "2022-03-21T12:02:23Z"}, {"author": "Juliusz Chroboczek", "text": "Wow, there are actual *people* in the room.<br/>", "time": "2022-03-21T12:03:06Z"}, {"author": "Adam Roach", "text": "We need Sean and Nils mannequins to put in the chair chairs. :)<br/>", "time": "2022-03-21T12:04:56Z"}, {"author": "Sergio Garcia Murillo", "text": "a bit of echo from the room<br/>", "time": "2022-03-21T12:06:32Z"}, {"author": "Spencer Dawkins", "text": "Adam - but certainly not Sean and Nils blow-up dolls, right?<br/>", "time": "2022-03-21T12:07:10Z"}, {"author": "Juliusz Chroboczek", "text": "Famous last words :-)<br/>", "time": "2022-03-21T12:07:26Z"}, {"author": "Jonathan Lennox", "text": "Sergio is very backlit<br/>", "time": "2022-03-21T12:08:09Z"}, {"author": "Juliusz Chroboczek", "text": "@Meetecho, the echo from the room is pretty bad.<br/>", "time": "2022-03-21T12:08:18Z"}, {"author": "Lorenzo Miniero", "text": "I only see chair slides and my slides in the meeting materials<br/>", "time": "2022-03-21T12:09:43Z"}, {"author": "Ted Hardie", "text": "That matches what I see, and I just reloaded the meeting materials.<br/>", "time": "2022-03-21T12:10:11Z"}, {"author": "Jonathan Lennox", "text": "I see two decks - chair slides, and impelementation in janus.&nbsp;&nbsp;Same in the datatracker meeting maeterials<br/>", "time": "2022-03-21T12:10:12Z"}, {"author": "Simon Romano", "text": "They should be there now...<br/>", "time": "2022-03-21T12:10:12Z"}, {"author": "Ted Hardie", "text": "These are still not visible in the room.&nbsp;&nbsp;Perhaps just talk to the points, given that we are now 15 minutes into the hour?<br/>", "time": "2022-03-21T12:14:41Z"}, {"author": "Meetecho", "text": "Juliusz, echo should be reduced now, please let us know if it's still bad<br/>", "time": "2022-03-21T12:15:12Z"}, {"author": "Lorenzo Miniero", "text": "Slides need to be reimported, if they were updated<br/>", "time": "2022-03-21T12:15:57Z"}, {"author": "Lorenzo Miniero", "text": "I can try to do it now, if they were updated on the datatracker too<br/>", "time": "2022-03-21T12:16:11Z"}, {"author": "Lorenzo Miniero", "text": "I've just impoirted them again<br/>", "time": "2022-03-21T12:16:46Z"}, {"author": "Lorenzo Miniero", "text": "There should be two chair slides now, one of them should be the new one<br/>", "time": "2022-03-21T12:16:57Z"}, {"author": "Lorenzo Miniero", "text": "Yay!<br/>", "time": "2022-03-21T12:17:49Z"}, {"author": "Nils Ohlmeier", "text": "Sorry I got hit by the Chrome loosing permissions bug.<br/>", "time": "2022-03-21T12:19:02Z"}, {"author": "Sean Turner", "text": "Apologies - that was pretty much a chair problem ;)<br/>", "time": "2022-03-21T12:20:45Z"}, {"author": "Adam Roach", "text": "FWIW, this all looks right to me. The bit about not sending etags in other methods (e.g., DELETE) is more of a \"there's no reason to\" rather than \"should not,\" but everything else is fine.<br/>", "time": "2022-03-21T12:24:02Z"}, {"author": "Jonathan Lennox", "text": "Is there a difference between no if-match and if-match: \"*\"?<br/>", "time": "2022-03-21T12:24:34Z"}, {"author": "Lorenzo Miniero", "text": "I had the same doubt (mentioned in my slides later)<br/>", "time": "2022-03-21T12:25:50Z"}, {"author": "Juliusz Chroboczek", "text": "I still feel the restart/candidate distinction should be explicit in the request.&nbsp;&nbsp;It would allow the server to detect inconsistencies, and make everything easier to debug.<br/>", "time": "2022-03-21T12:28:10Z"}, {"author": "Adam Roach", "text": "From the HTTP perspective, there's a difference in that \"If-Match: *\" requires that the corresponding resource already exists. That seems correct for our scenario.<br/>", "time": "2022-03-21T12:28:17Z"}, {"author": "Ted Hardie", "text": "I thought if-match matched eTags with strong matching (byte for byte).&nbsp;&nbsp;Wouldn't that mean \"*\" would only match the literal *?<br/>", "time": "2022-03-21T12:28:36Z"}, {"author": "Ted Hardie", "text": "Ah, Adam's answer corrects this impression.<br/>", "time": "2022-03-21T12:29:07Z"}, {"author": "Jonathan Lennox", "text": "To type up my issue in case I was unclear at the mic - if you have more than one ice-restart PATCH request in flight at once, it'll be unclear to the server which one was sent first.&nbsp;&nbsp;This ambiguity is bad; one simple solution would simply be to forbid this, but as Sergio points out that might be bad if your device fell off its network (because it might not get the response to the first ice-restart).&nbsp;&nbsp;If so we'd need another way to resolve the ambiguity.<br/>", "time": "2022-03-21T12:34:10Z"}, {"author": "Juliusz Chroboczek", "text": "Jonathan, Ted, the behaviour of \"*\" is specified in RFC 7232 Section 3.1.<br/>", "time": "2022-03-21T12:34:53Z"}, {"author": "Adam Roach", "text": "I am stealing the word \"umptysquat\" for my future use.<br/>", "time": "2022-03-21T12:35:02Z"}, {"author": "Jonathan Lennox", "text": "I've raised my ICE restart issue on the github repo<br/>", "time": "2022-03-21T12:38:40Z"}, {"author": "Nils Ohlmeier", "text": "Thank you Jonathan<br/>", "time": "2022-03-21T12:39:20Z"}, {"author": "Juliusz Chroboczek", "text": "Fine with me.<br/>", "time": "2022-03-21T12:42:54Z"}, {"author": "Adam Roach", "text": "To be fair, the issue that Julisz raises has bedeveled HTTP users pretty much since the early '90's. I don't think we can fix it in here. :)<br/>", "time": "2022-03-21T12:43:32Z"}, {"author": "Lorenzo Miniero", "text": ":D<br/>", "time": "2022-03-21T12:44:25Z"}, {"author": "Adam Roach", "text": "Lorenzo's mic is way to quiet here<br/>", "time": "2022-03-21T12:46:28Z"}, {"author": "Adam Roach", "text": "s/to/too/<br/>", "time": "2022-03-21T12:46:34Z"}, {"author": "Juliusz Chroboczek", "text": "Error handling in general.&nbsp;&nbsp;It's always depressing when you finish a first prototype and then start doing the error handling.<br/>", "time": "2022-03-21T12:46:48Z"}, {"author": "Juliusz Chroboczek", "text": "There's a whole language designed around the premise that you should crash instead of handling errors (Erlang).<br/>", "time": "2022-03-21T12:47:08Z"}, {"author": "Adam Roach", "text": "As a footnote to what Lorenzo says about OBS -- they're planning on adding WHIP support this year, with a target of Q3; see <a href=\"https://github.com/obsproject/obs-studio/discussions/4021#discussioncomment-2127281\" rel=\"nofollow\">https://github.com/obsproject/obs-studio/discussions/4021#discussioncomment-2127281</a><br/>", "time": "2022-03-21T12:56:07Z"}, {"author": "Adam Roach", "text": "(This is the main OBS project rather than the Millicast fork)<br/>", "time": "2022-03-21T12:57:09Z"}, {"author": "Sean Turner", "text": "great!<br/>", "time": "2022-03-21T12:59:19Z"}]