- From: Cullen Jennings <fluffy@cisco.com>
- Date: Fri, 15 Jul 2011 07:33:56 -0700
- To: Christopher Blizzard <blizzard@mozilla.com>
- Cc: public-webrtc@w3.org
A very common use case is that you have microphones in your computer or monitor and also have a USB headset with a microphone. You want to be able to allow the user to easily choose between these in the application and switch during the call or game or whatever it is the user is doing. YOu might want to switch from the display microphones to headset because the person at the other end complains you sounds like crap, or the guys in the sitting across from you in the airport is giving you dirty looks, or because you are about to start talking about something a bit more private. Making the sure go find the preferences dialog for the OS, or the browsers, switch, then probably restart the application is just not a great experience. Compare how easy this all works in something like Skype or facetime with something that does not do it this way. I think it is fairly desirable to allow the JS app to pick the mic and camera and be able to change them during a session - sure it should default to whatever the user as specified as the default in their OS preferences. On Jul 14, 2011, at 17:30 , Christopher Blizzard wrote: > On 7/14/2011 3:41 PM, Ralph Giles wrote: >> >> s and developers alike. >> Yes, of course. I agree those are important, exciting features for >> mobile user agents. However, as far as our API requirements go, I >> believe the question is not whether a user can switch between cameras, >> but whether web content can have a share in that decision. >> >> As Ian suggests, does a web page need to pass hints to a browser which >> camera it expects a user to prefer? Must in-page controls be able to >> switch the camera? Can that be left to the user-agent? >> >> Personally, I think user experience can be improved when the api >> offers all this data, but we have a lot to do already without >> standardizing an extensible structure for such things at this stage of >> our implementation experience. >> >> -r >> > > I think that we shouldn't be pushing so much of this off to the agent. I want people to be able to create apps that are immersive, full screen and expose that stuff through beautiful UI. An interrupt-driven UI from the browser that's popup-based (OK to use the camera? which one? you want to change cameras? oh, you're gonna have to do that from a menu somewhere) isn't a great experience. I think once you're allowed to transmit video data it doesn't matter all that much which camera you're using. > > So I certainly fall on the side of exposing a lot of this information through APIs since that puts the app in control of the experience instead of the UA, which only has a few options for jamming itself into the experience. > > --Chris > Cullen Jennings For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/index.html
Received on Friday, 15 July 2011 14:34:28 UTC