- From: Stefan Håkansson LK <stefan.lk.hakansson@ericsson.com>
- Date: Tue, 12 Jul 2016 15:53:48 +0000
- To: "public-media-capture@w3.org" <public-media-capture@w3.org>, "public-webrtc@w3.org" <public-webrtc@w3.org>
Below is a summary of decisions made at the joint media capture and webrtc virtual interim meeting of June 28 2016. The slides referred to are at [1]. We send this out to get feedback if others agree to the decisions we recorded, but also to enable those who were not able attend to provide input if they think any of the decisions are wrong. Note that other things were also discussed at the call (see [2] for minutes and [3] for recording), below is just a summary of what was decided. Sorry for cross posting, but the meeting dealt with both mediacapture-main and webrtc-pc. Stefan for the chairs [1] https://www.w3.org/2011/04/webrtc/wiki/images/1/12/WebRTCWG-2016-06-28.pdf [2] https://www.w3.org/2016/06/28-webrtc-minutes [3] https://www.w3.org/2011/04/webrtc/wiki/June_28_2016#Minutes [4] https://github.com/w3c/mediacapture-main/pull/373 Mediacapture-main ----------------- Related to Issue #350: - Temporary vs. persistent permission (to use camera/microphone): Current spec is right (i.e. option A in slide 8) - What happens at revocation: all tracks sourced by the device to which right to use was revoked are stopped (i.e. option B in slide 9) Issue #359 (slide 12): The spec currently says that the UA must clear all deviceIds at the end of the current browsing session (given that the origin has never been given the right to use any camera or microphone) to avoid a big fingerprint. However, there were some at the meeting pushing for that the UA should have the alternative to treat deviceIds like cookies. The outcome was that Jan-Ivar is going to draft a proposal (in the form of a PR) that introduces this alternative, and if the group accepts it it will be added to the spec. Note, this PR is already out for review [4] Webrtc-pc --------- Issue 644/PR 675: The proposal in slide 15 was liked and will be adopted. getParameters().encodings[].dtx will read out an enum (not a boolean) - name of it will be bikeshedded. PR 683 will not be merged (slide 17) Issue 706: Bernards proposal of updating JSEP and WebRTC 1.0 was liked; for WebRTC 1.0 section 5.1 will be updated as outlined in slide 23. Issue 555 (slides 26 - 28): option 1 (slide 27) was preferred, Martin will draft a PR Issue 678 (slide 30): we will support Id’ing the intended receiver (in addition to the sender) as Martin proposes Issue 685: A section describing multicast SDPs will be added to JSEP.
Received on Tuesday, 12 July 2016 15:54:51 UTC