Re: [webrtc-pc] Treat rejected m= sections (port of 0) as having "inactive" direction.

See [this blog](https://blog.mozilla.org/webrtc/rtcrtptransceiver-explored/) for a refresher on what events we fire today (examples work in Firefox). In short, `ended` and `muted` are orthogonal. This works out somewhat ok in practice for playback at least, since video elements know to ignore `ended` tracks, so it doesn't matter much that they stay in their stream(s).

With this change we'd be firing not just `ended`, but also `muted` on remote tracks, as well as removing them from any remote streams and firing`removetrack` on said streams. Is that the intent?

This may make sense, but we'd need to update Firefox to comply, and we just shipped it.

Also, if we we want to do this, I think it would make sense to preserve the symmetry of doing this on the offerer side as well. Currently, this is done in prose under [stop()](http://w3c.github.io/webrtc-pc/#dom-rtcrtptransceiver-stop):

 *  *" In addition, since the receiver's [[ReceiverTrack]] has ended, the steps described in track ended MUST be followed."*

We probably want to extend this to do `muted` and `removetrack` then too, right?

-- 
GitHub Notification of comment by jan-ivar
Please view or discuss this issue at https://github.com/w3c/webrtc-pc/pull/1845#issuecomment-383648553 using your GitHub account

Received on Monday, 23 April 2018 17:02:49 UTC