Notes, October 1 2015

Den 01. okt. 2015 13:43, skrev Harald Alvestrand:
> As usual, I'm behind schedule. But I think we're doing fine.
> Only some of the items are commented by me.
> 
> 1) Overarching items
> 
> New editors' drafts:
> - mediacapture-main announced Sep 25
> - webrtc-pc announced Sep 22
> Do we need / are we able to do another round before TPAC?

We plan another round this week, and another 2 weeks after - if we have
content.


> 4) PRs - webrtc-pc
> 
> #266 Tidy up setLocal/RemoteDescription processing model (adam-be)
No feedback after last week's meeting. Harald to ping Peter.

> #273 Add a bunch of fields to RtpParameters and RtpEncodingParame (adam-be)
Needs submitter action. Adam has a PR on the PR.
Bernard thinks "InvalidParameters" should be a valid DOMException. Needs
reference.

> #280 Add extra IceTransport read-only attributes and methods. (adam-be)
Already merged.

> #291 Add PeerConnection.connectionState. (adam-be)

Merged.

> #293 Add RTCRtpTransceiver and PeerConnection.addMedia. ()

Waiting for Peter.

> #300 add API to get SSRC and audio levels (fluffy)

Merged. Base for time isn't mentioned, so it's document load time.

> #301 fix RTCStatsReport with object and maplike instead of getter (fluffy)

This is a breaking change on existing code. Still, let's merge.
Stefan to write a note to the list about it.

> #302 Removing interface use for RTCSessionDescription and RTCIceC ()

It's a shortcut for users. Let's merge.

> #307 Fix broken RTCPeerConnection's operations array to work with ()

Closed, #314 seems to have the consensus.

> #314 Update the operations queue to handle promises and closed si ()
151 comments - that's a new record!

> #315 Reorganize travis setup to use centralized check script ()

Not ready.

> #317 Make RTCDataChannels Transferable ()
> No response to pushback on mailing lists
Post 1.0.


Adam takes on pushing a new draft based on these merges - before the
weekend if possible, after the weekend if not.
Adam will add Jan-Ivar, Peter Thatcher and Peter Saint-Andre to the
contributors section.

> 
> 5) Bugs, webrtc-pc
> 
Scripting error: There are 57 open ones. My script failed to pick them up.

We assigned a number of bugs to responsible people.
> 
> #274 Add API points for creating the ICE/DTLS transport and RtpSe ()
> #275 RTCStatsReport seems broken ()
> #276 Control to turn on Client-to-Mixer Audio Level in offers ()
> #277 Side-effects of mitigations for #179 in demo pages  ()
> #278 How does the JS know if ICE is connected but DTLS isn't? ()
> #282 Constraint registration - need closure ()
> #285 Stats return object should have maplike, not getter ()
> #286 The "consumer of the session description” conflicts with ter ()
> #287 nit: "audio stream” and “video stream" mentions ()
> #288 Does addTrack require the streams argument? ()

Stopped at this point, some later ones closed.

> #294 P2P VPN ()
> Closing as "not a spec issue".
> #295 Guidance for extending objects vs extending Stats needed ()
> #296 Debugging ICE problems needs more info ()
> #299 operations array prose is broken ()
> #305 Describe what happens when media changes ()
> #308 RTCRtpSender.get/setParameters is underspecified ()
> #309 Use of DOMHiResTimestamp in Section 8.5 ()
> #312 We should not reference the public API in algorithms ()
> Haven't we seen this before? Same fix, I guess.
> #316 Don't use Interface.attribute to refer to properties of inst ()
> We know this one too.
> #319 Work through error definitions ()

>
> 2) PRs - mediacapture-main
>
> (lists are now brought to you by a short Python script and the github API)
>
> (ONE request? We've never been that low!)
>
> #235 Remove static like definitions of attributes and methods (adam-be)
> Looks ready, Jan-Ivar says LGTM. Merge?

Merged. Zero PRs!!!!!!!

>
> 3) Bugs, mediacapture-main
>
> All are assigned. Assignee should update status.
> We also have the very recent comment from PFWG (Janina Sajka). Someone
> needs to take on responding to that, and possibly turn issues into bugs.

I forgot that one. Will reply on list.

>
> #118 A practical algorithm for constraint resolution needs to be
> (alvestrand)
> Nobody but Cullen seems to care.
>
> #127 Lack of timeout / cancellation leads to UI inconsistencies (alvestrand)
> Iceboxed.
>
> #193 How is adaptive frame rate to be handled? (alvestrand)
> Still in the dark on what's a reasonable outcome on that one.
>
> #198 MediaTrackCapabilities discovery via enumerateDevices() (adam-be)
> Fixed by #211 (merged). Closing.
> Reminder: If you write "fixes #nnn" in a PR, github will close the issue
> when you merge the PR.
>
> #212 Privacy/security review questions (dontcallmedom)
> No PR yet.
>
> #227 Update onoverconstrained event handler attribute description (burnburn)
> Dan?
>
> #232 clone a track: "inherit" not defined (adam-be)
> No PR yet. Anne and Adam and Domenic seem to have a more fundamental
> discussion that really doesn't matter in this context.
>
> #233 "MediaStream.active" is wrong (adam-be)
> Should be fixed by #235
>
> #237 step 5 in clone algorithm seems confusing. (burnburn)
> On reading in context, agree it's confusing.
>
> #238 Some text on constraints needs updating. (burnburn)
> #242 Management of track members in a stream (aboba)
> #243 Confusing typo in sources and sinks. (burnburn)
> I'm confused too. Not a good sign.
>
> #244 Explain how extensions are to be done (burnburn)
> #246 track.applyConstraints() doesn't do anything. (burnburn)
> #248 Conflicting information about OverconstrainedError.constrain (burnburn)
> #249 PING: Document tradeoff for non-HTTPS usage of getUserMedia (aboba)
> #250 PING: Refer to security-arch on the MUST inside Best Practic (aboba)
> #251 PING: Mark fingerprintability of device enumeration (aboba)
>
> 
> 
> 
> 
> 
>

Received on Thursday, 1 October 2015 15:03:04 UTC