IRC log of webtransport on 2019-09-18
Timestamps are in UTC.
- 05:43:00 [RRSAgent]
- RRSAgent has joined #webtransport
- 05:43:00 [RRSAgent]
- logging to https://www.w3.org/2019/09/18-webtransport-irc
- 05:43:39 [JRR_Tolkien]
- Youenn: how would 'dropping' data happen?
- 05:44:15 [JRR_Tolkien]
- Victor_??_Google: currently no way to ignore, UDP sockets allow ignoring by not reading from buffer (overflow)
- 05:44:44 [nicktr]
- rrsagent, set logs public
- 05:45:12 [jianjunz]
- jianjunz has joined #webtransport
- 05:45:27 [JRR_Tolkien]
- Peter_Thatcher_Google: one of the goals is to apply back-pressure
- 05:46:02 [JRR_Tolkien]
- Youenn: want to send datagram above max size, what would happen?
- 05:46:32 [JRR_Tolkien]
- Victor: exception. API call will give you max size.
- 05:46:51 [minami]
- minami has joined #webtransport
- 05:47:44 [JRR_Tolkien]
- Victor: closing stream semantics can help. there are a lot of higher level apis you can provide to deal with larger messages.
- 05:48:53 [jungkees]
- jungkees has joined #webtransport
- 05:49:08 [JRR_Tolkien]
- Peter: example in explainer shows message of arbitrary size
- 05:49:28 [JRR_Tolkien]
- ???: is protocol name set in constructor?
- 05:50:06 [JRR_Tolkien]
- Victor: it's done using interface mixin. constructors may vary, but object must support WebTransport interface
- 05:50:45 [JRR_Tolkien]
- Alex_Apple: WebSocket has handshake to prevent misuse. does this have a similar mechanism
- 05:51:48 [JRR_Tolkien]
- Victor: Quic transport uses ALPN to match a specific value, without a valid value it won't work and fail the connection. Browser verifies list of domains can talk to before sending data
- 05:52:22 [JRR_Tolkien]
- Jan-Ivar Mozilla: does getWriter get a different writer every time?
- 05:52:55 [wonsuk__]
- wonsuk__ has joined #webtransport
- 05:53:07 [JRR_Tolkien]
- ??: once stream is closed it will never be used again.
- 05:53:20 [ricea]
- That was me
- 05:53:36 [JRR_Tolkien]
- full name would help :)
- 05:53:39 [ricea]
- Adam Rice
- 05:53:40 [nicktr]
- s/??/ricea/
- 05:54:09 [inamori_]
- inamori_ has joined #webtransport
- 05:54:37 [JRR_Tolkien]
- Jan Ivar: calling send datagram twice gets two streams, how do they interact (multiplex, etc.)
- 05:54:55 [JRR_Tolkien]
- Victor: should only be one stream in datagrams. they should be either sent or not sent (UDP semantics)
- 05:55:15 [JRR_Tolkien]
- Victor: Goal is to use UDP semantics with flow/congestion control
- 05:55:41 [JRR_Tolkien]
- ????: timing data was missed with WebSockets. can metrics/stats be added to API?
- 05:56:16 [JRR_Tolkien]
- Victor: some stats are there, need users to provide feedback on which stats are interesting/needed.
- 05:56:44 [JRR_Tolkien]
- ?????: ...fallback when you don't have UDP?
- 05:57:04 [JRR_Tolkien]
- Emad_Omara_Google: will it support 0-RTT?
- 05:57:24 [JRR_Tolkien]
- Victor: no support for 0-RTT. API design is hard, more complex state machine, etc.
- 05:57:38 [JRR_Tolkien]
- Youenn: will WebRTC stats be reused (applicable ones)?
- 05:58:13 [JRR_Tolkien]
- Victor: added the ones that we know are already useful
- 05:59:12 [JRR_Tolkien]
- Steve_Anton_Google: WebRTC has a lot of stats because you are not on the data layer, this should be different because you are using the networking layer
- 06:03:38 [JRR_Tolkien]
- ??: how will debugging work? what about dev tools?
- 06:03:57 [JRR_Tolkien]
- Victor: dev tools is planned to be supported.
- 06:05:12 [JRR_Tolkien]
- Amit_Hilbuch_Google: how will this interact with p2p scenario
- 06:05:23 [JRR_Tolkien]
- Peter: currently focused on Server-Client but planned
- 06:05:35 [JRR_Tolkien]
- Jan Ivar: is congestion control required?
- 06:05:57 [JRR_Tolkien]
- Victor: yes all streams are required to use congestion control.
- 06:06:54 [JRR_Tolkien]
- ??: Implemented in quic using datagram frame, data will be blocked if window is full (adding more data will cause it to be dropped)
- 06:07:35 [JRR_Tolkien]
- ???: will Keep-Alive behavior be part of the transport?
- 06:08:26 [JRR_Tolkien]
- Victor: connection will be kept open as long as it isn't explicitly closed (or garbage collected)
- 06:08:58 [JRR_Tolkien]
- David_Google: Quic has a keep-alive mechanism.
- 06:09:16 [JRR_Tolkien]
- s/???/Mark Foltz_Google
- 06:09:27 [JRR_Tolkien]
- 06:10:42 [JRR_Tolkien]
- Victor: ... multiplexing mechanism is not yet clear. this is a learning from WebSockets on url leading to http assumption.
- 06:11:06 [JRR_Tolkien]
- ????: it should be easy for Http3
- 06:11:20 [JRR_Tolkien]
- Victor: for http it is easy, but for quic transport it's not as clear
- 06:11:56 [nakakura]
- nakakura has joined #webtransport
- 06:12:24 [nicktr]
- s/????/AdrianHB/
- 06:12:57 [JRR_Tolkien]
- David: next steps for this work - do people have opinions on the best plan forward?
- 06:13:06 [JRR_Tolkien]
- Jan Ivar: considered forming a WG?
- 06:13:27 [JRR_Tolkien]
- Victor: eventually it will hit a WG, but it's not ready for that state yet (no real dev feedback yet)
- 06:14:08 [JRR_Tolkien]
- Jan-Ivar: mozilla would support a working group
- 06:14:36 [JRR_Tolkien]
- Youenn: would node js or similar frameworks use the same pattern?
- 06:14:49 [JRR_Tolkien]
- Victor: have not reached out about that yet
- 06:15:19 [JRR_Tolkien]
- Youenn: node js can be used for both client and server
- 06:17:32 [JRR_Tolkien]
- Peter: any feedback for technical discussions? CC or BWE?
- 06:17:45 [JRR_Tolkien]
- ??: what happens during connection?
- 06:18:01 [JRR_Tolkien]
- Victor: we need to define how it interacts with extensions
- 06:18:44 [JRR_Tolkien]
- ??: is the expectation that the connection initiation will carry cookies?
- 06:19:20 [JRR_Tolkien]
- Victor: explicitly no, considered exposing channel bindings
- 06:19:42 [JRR_Tolkien]
- ??: is there expectation that a user gesture is required to initiate a connection?
- 06:20:04 [JRR_Tolkien]
- Victor: currently not.
- 06:20:21 [JRR_Tolkien]
- Jan Ivar: thoughts on putting this in service worker?
- 06:20:36 [JRR_Tolkien]
- Victor: should be supported in service workers.
- 06:21:21 [JRR_Tolkien]
- ?????: is this meant to be protocol-agnostic?
- 06:21:48 [JRR_Tolkien]
- Victor: there is a websocket based fallback
- 06:22:01 [JRR_Tolkien]
- ?: how does this interact with prioritization (multiplexing scenario)
- 06:22:52 [JRR_Tolkien]
- Victor: prioritization is a hard topic in http. with http3 it behaves similar to regular http streams (scheduler provides fairness). dos + fairness are concerns
- 06:24:16 [nicktr]
- rrsagent, draft minutes
- 06:24:16 [RRSAgent]
- I have made the request to generate https://www.w3.org/2019/09/18-webtransport-minutes.html nicktr
- 07:57:21 [Zakim]
- Zakim has left #webtransport
- 08:02:36 [wseltzer]
- wseltzer has joined #webtransport
- 08:02:42 [wseltzer]
- rrsagent, pointer?
- 08:02:42 [RRSAgent]
- See https://www.w3.org/2019/09/18-webtransport-irc#T08-02-42
- 09:20:57 [hyojin]
- hyojin has left #webtransport