RTCWEB@IETF100
20171015

https://datatracker.ietf.org/meeting/100/session/rtcweb

Magnus Westerlund: Simulcast discussion is happening in MMUSIC

Sean: Cluster 238 is in progress.  Please help it get done.
-fec is in WGLC but needs a comment to be addressed
-jsep is headed to IESG

EKR: Is -sdp-examples full of mistakes?
Cullen: -sdp-examples may have mistakes because it was partly hand-generated, but it's not full of mistakes.

Harald: Is there a way to extract the SDP?
Cullen: No, but I will be happy to help with that if we decide to publish it.

Magnus: We should have Simulcast in -sdp-examples and/or in the Simulcast document.
Cullen: I'm happy to help with that, in this document or a new one.

----------------------------

EKR
On -security
If there is no objection we will go to WGLC on -security.

On -security-arch
Call for objections: can we restrict getUserMedia to secure origins?
Jonathan Lennox and Richard Barnes: What about debug mode?
EKR: Insecure access in debug mode is fine.
No objections voiced.

No objection to banning MKI.

On Identity providers:
Decided: allow a port number in the domain field and document that it is not authenticated.

Decided: Write "some crap" regarding SDP containing multiple identities.

Decided: Require one DTLS handshake per component.  Overrides 5764.

---------------------

Peter Thatcher / Harald Alvestrand

Split the DSCP control?  Requires pulling the draft from the RFC-editors queue and redoing a (targeted) WGLC.

Hum for incorporating this change: rough consensus that the change should not be incorporated at this time.

Decided: This change will not be incorporated.

---------------------

Should Opus FEC be MTI?

Conclusion: Cullen will help to write an e-mail clarifying the intent of this change.