It's amusing that TSVWG gets such a humungous room assignment.
", "time": "2024-11-08T09:30:40Z"}, {"author": "Jonathan Morton", "text": "I suppose it's necessary to contain certain egos. :)
", "time": "2024-11-08T09:31:20Z"}, {"author": "Jason Livingood", "text": ":big_smile:
", "time": "2024-11-08T09:32:51Z"}, {"author": "Jason Livingood", "text": "Each presentation shall be preceded by a song and dance routine on the stage.
", "time": "2024-11-08T09:33:38Z"}, {"author": "Jason Livingood", "text": "This is GREAT work and I really like the chart - first time I have seen all this in one place. Great aid to help folks understand how to actually deploy ECN.
", "time": "2024-11-08T09:38:48Z"}, {"author": "Joseph Beshay", "text": "Agreed. This is very valuable to have. I would have loved to have this while implementing L4S support, especially seeing both the send and receive sides in one place.
", "time": "2024-11-08T09:42:32Z"}, {"author": "Julien Maisonneuve", "text": "Could this draft actually recommend a \"proper way\" instead of leaving differences persist unmodified ?
", "time": "2024-11-08T09:42:51Z"}, {"author": "Brian Trammell", "text": "taaaaaaaaaaaaps
", "time": "2024-11-08T09:45:16Z"}, {"author": "Jonathan Morton", "text": "Windows appears to have been an \"interesting experience\" in the sense of \"it's completely broken and buggy\".
", "time": "2024-11-08T09:45:57Z"}, {"author": "Jonathan Morton", "text": "So on Windows you should use the POSIX sockets API directly, as well as on Linux.
", "time": "2024-11-08T09:46:49Z"}, {"author": "Jonathan Lennox", "text": "Brian: what does TAPS actually say about this?
", "time": "2024-11-08T09:47:39Z"}, {"author": "Jonathan Lennox", "text": "(And is it the same thing Network.Framework does?)
", "time": "2024-11-08T09:47:55Z"}, {"author": "Brian Trammell", "text": "heh was just looking that up after I said that :)
", "time": "2024-11-08T09:49:02Z"}, {"author": "Brian Trammell", "text": "https://datatracker.ietf.org/doc/draft-ietf-taps-interface/ section 9.3.3.1
", "time": "2024-11-08T09:49:25Z"}, {"author": "Jonathan Morton", "text": "1-byte vs 4-byte will work on a LITTLE-ENDIAN platform if you pass a pointer to a 4-byte int
", "time": "2024-11-08T09:49:42Z"}, {"author": "Jonathan Morton", "text": "but on a big-endian platform it will fail
", "time": "2024-11-08T09:49:51Z"}, {"author": "Jonathan Morton", "text": "so try it on PowerPC ot something
", "time": "2024-11-08T09:50:23Z"}, {"author": "Brian Trammell", "text": "that's all very abstract though, not sure how network.framework does that
", "time": "2024-11-08T09:50:45Z"}, {"author": "Ingemar Johansson", "text": "Trying to figure out how to access the ECN bits was an \"experience\" when I wanted to have in for IPv4 in Linux. Then I had a similar \"experience\" with IPv6. This document is really welcome.
", "time": "2024-11-08T09:50:49Z"}, {"author": "Craig Taylor", "text": "...surely there must be a way Martin can be comelled to do this?
", "time": "2024-11-08T09:53:22Z"}, {"author": "Craig Taylor", "text": "compelled
", "time": "2024-11-08T09:53:31Z"}, {"author": "Ingemar Johansson", "text": "During the L4S work the last few years, one big question have been around proper access to the 2 bits in the IP header. It has often (rightfully) been regarded as complicated.
", "time": "2024-11-08T09:56:27Z"}, {"author": "Eduard V", "text": "It does not make sense to trust Client (on QoS matters). Server is possible to trust only if it is directly connected to the ISP.
", "time": "2024-11-08T10:09:38Z"}, {"author": "Mohamed Boucadair", "text": "This is actualit\u00e9s pr\u00e9f\u00e9rence for packets within th\u00e9 samedi flow
", "time": "2024-11-08T10:10:41Z"}, {"author": "Martin Duke", "text": "@craig requests for my time can be directed to @ianswett
", "time": "2024-11-08T10:11:05Z"}, {"author": "Martin Duke", "text": "I don't want this info doc to be my editorial opinion or to shame people. I might send and email to the list on what I think would be conceptually cleanest.
", "time": "2024-11-08T10:12:07Z"}, {"author": "Martin Duke", "text": "Also my bug speculation to be on the record, but am happy to discuss it offline
", "time": "2024-11-08T10:13:21Z"}, {"author": "Martin Duke", "text": "Chr
", "time": "2024-11-08T10:14:01Z"}, {"author": "Martin Duke", "text": "The Chrome network stack is (IIUC) the basis of Edge and several other browsers, so once I'm done something like 99% of the browser market will send QUIC ACK-ECN
", "time": "2024-11-08T10:15:04Z"}, {"author": "Jonathan Lennox", "text": "Firefox is already doing it?
", "time": "2024-11-08T10:16:30Z"}, {"author": "Spencer Dawkins", "text": "+1 on Brian Trammell's understanding of this draft being complemenrary.
", "time": "2024-11-08T10:27:08Z"}, {"author": "Eduard V", "text": "+1 to Brian. APN is about the same.
", "time": "2024-11-08T10:28:45Z"}, {"author": "Spencer Dawkins", "text": "Brian was referring to the PANRG RFC https://datatracker.ietf.org/doc/rfc9049/
", "time": "2024-11-08T10:29:32Z"}, {"author": "Martin Duke", "text": "@jonathan based on Max Inden's contributions to the draft, the answer is either \"yes\" or \"soon\"
", "time": "2024-11-08T10:30:14Z"}, {"author": "Brian Trammell", "text": "to perhaps elaborate on my \"damped enthusiasm\" --> if I'm thinking about network cooperative approaches that actually work, the solution spaces are more complex than \"metadata flows exclusively toward the network\"...
", "time": "2024-11-08T10:34:33Z"}, {"author": "Spencer Dawkins", "text": "Brian - is that RG space or WG space work?
", "time": "2024-11-08T10:35:39Z"}, {"author": "Brian Trammell", "text": "(+many Mirja)
", "time": "2024-11-08T10:36:45Z"}, {"author": "Brian Trammell", "text": "spencer: both
", "time": "2024-11-08T10:36:58Z"}, {"author": "Brian Trammell", "text": "anything in the inward-metadata-flow space really needs to have a convincing answer to 9049 section 4.9 :)
", "time": "2024-11-08T10:38:55Z"}, {"author": "Brian Trammell", "text": "\"endpoint trust of the network\" for those who don't have the tab open
", "time": "2024-11-08T10:42:42Z"}, {"author": "Spencer Dawkins", "text": "I was waiting for Zahed to say \"I asked for this, but I was young, and easily misled\" :grinning_face_with_smiling_eyes:
", "time": "2024-11-08T10:43:17Z"}, {"author": "Spencer Dawkins", "text": "I do think this has been a useful discussion, no matter what happens next.
", "time": "2024-11-08T10:44:28Z"}, {"author": "Martin Duke", "text": "When I was AD I remember feeling personally ambivalent/indifferent about the work, but very aware that the Trammell/Eggert comments were lurking out there, and therefore a bit reluctant to move forward
", "time": "2024-11-08T10:45:45Z"}, {"author": "Martin Duke", "text": "And even more so beyond the TSVWG, where bad things could happen in IETF Last Call
", "time": "2024-11-08T10:46:21Z"}, {"author": "Spencer Dawkins", "text": "What Martin said ^^^
", "time": "2024-11-08T10:49:37Z"}, {"author": "Spencer Dawkins", "text": "Lars isn't the tip of the iceberg, he's the first iceberg in an ice flow ...so, +1 Martin
", "time": "2024-11-08T10:51:00Z"}, {"author": "Eduard V", "text": "+1 to Mirja
", "time": "2024-11-08T10:53:06Z"}, {"author": "Zaheduzzaman Sarker", "text": "@spencer.. I had no young life !!! so I can't refer to that ;-).. you all made me jump over the \"I was young\" thing.. :-)
", "time": "2024-11-08T10:56:51Z"}, {"author": "Spencer Dawkins", "text": "@Zahed, Bwahahahaha! My fiendish feedback to Nomcom WORKED!
", "time": "2024-11-08T11:00:28Z"}, {"author": "Zaheduzzaman Sarker", "text": ":grinning_face_with_smiling_eyes:
", "time": "2024-11-08T11:01:13Z"}, {"author": "Eduard V", "text": "I did vote \"No\" for both because Mirja question was not voted
", "time": "2024-11-08T11:04:42Z"}, {"author": "Eduard V", "text": "\"Use cases\" should be before \"requirements\"
", "time": "2024-11-08T11:10:48Z"}, {"author": "Craig Taylor", "text": "I think polls were 'clear' here:
", "time": "2024-11-08T11:18:29Z"}, {"author": "Craig Taylor", "text": "There's something to discus, but not here...
", "time": "2024-11-08T11:18:40Z"}]