- From: Robin Berjon <robin@berjon.com>
- Date: Thu, 1 Dec 2011 12:09:44 +0100
- To: Harald Alvestrand <harald@alvestrand.no>
- Cc: Travis Leithead <travis.leithead@microsoft.com>, Brian LeRoux <b@brian.io>, "public-media-capture@w3.org" <public-media-capture@w3.org>
On Dec 1, 2011, at 11:58 , Harald Alvestrand wrote: > On 11/30/2011 08:02 PM, Travis Leithead wrote: >> See my other reply on the other thread [3]. I believe multi-device scenarios should be configured _after_ the user has granted general permission for audio or video or both. >> >> However, if this is necessary, the dictionary can be extended to include additional data for different device hints as well. I believe an earlier draft of WebRTC had hints for "user" and "environment" video cameras, which could easily be added into the dictionary if needed. > I believe the WEBRTC meeting in Santa Clara tended to believe that this parameter needed to be a set of attributes, and extensible. > > Will a dictionary accomplish this cleanly? Yes, a dictionary maps well to JS literal objects and it's easy to just add fields to it over time. It certainly beats the "this is an interface, but you can't ever really instantiate it or even be given an instance of it" approach that prevailed previously (and which is the NoInterfaceObject construct that is in the draft). -- Robin Berjon - http://berjon.com/ - @robinberjon
Received on Thursday, 1 December 2011 11:10:13 UTC