Notes, June 2 2016

Den 02. juni 2016 15:15, skrev Harald Alvestrand:
> Let's depart a little from the usual.
> 
> First, let's assign or act on the decisions from the interim:
> 
> * receiver.track.stop() will not stop the transceiver, it will continue
> to be alive and send RTCP (related to [2])
> * We will add an API control whether or not DTX is used (so the app can
> decide to not use it even if DTX/CN is negotiated in the SDP O/A).
> Exactly what the API will look like is still discussed ([3]).
> * We will add a table summarizing RTCRtpEncodingParameters attributes as
> proposed in [4]
> * We will use both MIME type _and_ subType ([5])
> * We will not make addTransceiver/addTrack async, which means transport
> can in some situations be null [6]
> * Adding the same track several times via supplying it as argument to
> addTransceiver() will stay allowed [7]
> * We will not add (back) legacy methods like addStream or removeStream
> [8] or related events.
> 
> [1] http://www.w3.org/2016/05/27-webrtc-minutes.html
> [2] https://github.com/w3c/webrtc-pc/pull/662

Merged.

> [3] https://github.com/w3c/webrtc-pc/pull/675

Travis issues & ongoing discussion. Not ready for merge.
(AMR / telco codecs have in-codec CN, just to add to the confusion)
Stefan will file a bug on clarifying what CN/DTX does in the context of
AMR & Opus and other codecs where CN is within the codec.

> [4] https://github.com/w3c/webrtc-pc/pull/646

Merged.

> [5] https://github.com/w3c/webrtc-pc/pull/648

Merged.

> [6] https://github.com/w3c/webrtc-pc/pull/666

Merged

> [7] https://github.com/w3c/webrtc-pc/issues/583

Closed

> [8] https://github.com/w3c/webrtc-pc/issues/568

Closed

> 
> They were posted today, so if we have any doubts about any of them, we
> should make it ready for merge next week, but if it's clear that it's
> ready to merge, or the action is "close issue/PR", we should do that now.
> 
> Then back to our regularly scheduled programme:
> 
> Mediacapture-main
> =================
> Pulls
> -----
> #362 Remove 'User Media in an IFrame' section and use new 'user  ()
> LGTM - the new flag may not be in HTML5.1, but that may be OK.

Seems OK. Merged.

#363 Replace one png image with an svg

Dan will supply the powerpoints they originally came from, and Adam will
work
on getting all the images upgraded.

> 
> Issues
> ------
> #350 New permission definitions are wrong. (alvestrand)
> #353 IFrame access control makes problems for [CEReactions] intr (adam-be)
> Fixed by #362
> 
> #359 MUST clear requirement for deviceId (aboba)
> Stefan, do we have a proposal?

Discussion ongoing. Stefan to file a bug on Chrome for not following the
spec.

> #360 Specify relation between return from getConstraints and con (stefhak)

Assigned to Dan for review and possible PR.

> #361 Example image text is hard to read (adam-be)

See #363 discussion.
> 
> 
> WebRTC-PC
> =========
> Pulls
> -----
> #601 Specify the synchronous and queued steps for addIceCandidat (adam-be)
> Status, Adam?

No update. There will be a new PR.

> 
> #624 Upscale allowed (fluffy)
> No response
> 
> #646 Table of RTCRtpEncodingParameters for RtpSender/RtpReceiver
> (pthatcherg)
> Covered above
> #647 Clarification on RTX in Codec Capabilities/Parameters ()
> Was discussed at interim. Seems to require more work/discussion.

More discussion.

> #648 mimeType clarification ()
> Covered above
> #662 Effect of RTCRtpReceiver.track.stop() ()
> Covered above
> #666 transports can be null (taylor-b)
> Covered above
> #668 ICE Transport State Diagram (taylor-b)
> No Taylor response.
> 
> #675 RTCRtpEncodingParameters attribute to turn on/off sending C ()
> Covered above
> #676 transceiver.stop() causes negotiation-needed to be set ()
> lgtm

Merged
> 
> #677 Fix rtcpTransport description ()
> lgtm
> 

Merged.

> #679 Clean up conflict marker ()
> needs list discussion.....?

Merged

> 
> #680 Link to JSEP ICE candidate policy ()
> needs grammar fix. (previous comment was just to see if you read them.)

Merged.
> 
> #682 update JSEP link for reuse to include subsequent answers ()
> Didn't know you could have multiple targets....

it results in <section> and <section>.
> 
> #683 Add RTCRtpCodecRtxParameters dictionary (related to #548) ()
> Aboba should have opinions on this one.

Assigned to Aboba and labelled "next interim topic".

> 
> #686 update JSEP ref for incoming media in 5.1.1 ()
> lgtm

Merged.

> 
> Issues
> ------
> #253 Assurance that requests to IdP proxy originate from the use
> (martinthomson)
> #257 ICE Candidate should have accessors for protocol-relevant i
> (alvestrand)
> #295 Guidance for extending objects vs extending Stats needed (alvestrand)
> #296 Debugging ICE problems needs more info (aboba)
> #305 Describe what happens when media changes (fluffy)
> #337 Interfacing between WebRTC spec and JSEP (burnburn)
> #369 addTrack's streams parameter is unused. (adam-be)
> #457 Non-normative ICE state transition diagram (taylor-b)

See PR 668

> #518 PING review of webrtc-pc spec (dontcallmedom)
> #526 NetworkError event is not defined and might not be needed (adam-be)
> #548 RTX/RED/FEC handling (aboba)
> #550 'the process to apply candidate' should link to JSEP (adam-be)
> #551 Errors when identifying a m-line in addIceCandidate() (adam-be)
> #554 We never fire the 'connectionstatechange' event (adam-be)
> #555 Sort out requirements around IdpLoginError (martinthomson)
> #561 Normatively cite webrtc-stats for sections 8.x (alvestrand)
> #562 What to do with an RTCIdentityProvider that returns rubbish
> (martinthomson)
> #566 Separate sender and receiver sets are unnecessary when we h (burnburn)
> #568 Should we specify how addStream()/"addstream event" should  (aboba)
> Covered above
> #571 Mechanisms for populating the contents of RTCRtpSender/Rece (taylor-b)
> #578 Need to specify precisely when MID generation happens (adam-be)
> #579 Congruenting about "The negotiation-needed flag is cleared  (adam-be)
> #583 Is it OK to call addTransceiver() with a track already adde (stefhak)
> Covered above
> #585 Unclear if RTCRtpTransceiver.stop() acts right away or requ (taylor-b)
> #597 Calling RTCRtpReceiver.track.stop() (aboba)
> #600 Operations queue: What is run synchronously before the oper (adam-be)
> #644 Fob on RTCRtpEncodingParameters to turn on and off sending  (aboba)
> #645 public negotiation-needed flag as readonly (adam-be)
> #650 mimeType Clarification (aboba)
> #651 addTransceiver/addTrack: need to be async? (aboba)
> #652 RTCIceCandidate description contains some junk characters (burnburn)
> #653 Align RTCIceTransportPolicy name and links with JSEP ICE ca (burnburn)
> #654 Need JSEP reference for general RTCPeerConnection descripti (burnburn)
> #655 Update JSEP reference to 5.8 (burnburn)
> #658 Link addIceCandidate to JSEP for applying ICE candidate (burnburn)
> #661 Add informative table of all things that can cause negotiat ()
> #669 Missing destruction sequence for ice agent. (aboba)
> #671 Processing remote MediaStreamTracks without MediaStreams in
> (alvestrand)
> #674 The doc should be updated to say that transceiver.stop() ca ()
> #678 Support assertions that identify the recipient ()

Asking Martin to align with identity

> #684 Proper JSEP reference for sendEncodings in subsequent offer ()

Dan got feedback - proposal good. Assigned back to Dan.

> #685 Update JSEP reference for receipt of multiple RTP encodings ()

JSEP topic
>

Received on Thursday, 2 June 2016 15:00:54 UTC