Re: ML in WebApps

On 2018-01-26, 23:14, "Justin Uberti" <juberti@google.com<mailto:juberti@google.com>> wrote:

I think this is very interesting but likely outside the scope of a v2 WebRTC charter.

+1 to high probability much lying outside WebRTC API but will there be dependencies? I’d like to keep the door open for proposals for extensions to the API coming the next 2-4 years (I do not expect we’ll finish the next version much faster than v1. Do you, ☺?).

Isn't there some effort for this already underway in W3C?
You mean ML for webrtc use cases? I am not aware of that. Probably Dom knows more… Guess we could check with the Web Platform people.
There are other stuff discussed in W3C (no, haven’t got a complete inventory, ☺!).
But there are other ML in browsers discussions/efforts ongoing, outside W3C, and in native device environments there are much happening in my opinion. W3C has it discussed in the strategy funnel…

And if it has not been clear before- my intention is to have this topics discussed and understand how the WG would like to handle this, not pushing for a particular approach in the charter at this point. I can live with, and handle, the different possible outcomes but clarity is always more efficient.

Also, the other topics- “SDP-free”, “Transport evolution”, “low level ICE++” are more important to fix in the short/mid-term.

But the key question is how to handle WebRTC ML support in the charter. Leave outside for future charters? Perhaps….

Be well!
Göran