What if Microsoft was Really Serious about WebRTC
IE11 would have some WebRTC implementation already.
Do you know that MailChimp already uses WebRTC? I recently opened there an account to manage my email subscription on research I do. Once I did, I took the time to fill in my profile – a bad habit I had. MailChimp gave me the option of taking my photo for the profile – and that invoked the getUserMedia API of WebRTC in order to access the camera and capture the image. No Flash. No plugin. Just pure HTML5.
It was a wow moment to me. It is that kind of a time when you see how the banal aspects of WebRTC can assist in areas that you don't really care about. I am a video conferencing person, so it is easy to forget such details as camera capture and what you can do with that thing alone.
Then I started thinking. Why isn't this supported by all browsers?
How much would it hurt Microsoft or Apple to admit that acquisition of camera feed can be important and relevant in itself? It poses no real threat to Lync or Skype or FaceTime. It just gives one more capability to those implementing their service.
If either Microsoft or Apple were serious about WebRTC, then they would have implemented that simple capability and pushed it to market already. Nothing to it. A show of faith. An indication that they really want this.
When both of them just whine about VP8 and suggest H.264 instead, and Microsoft just trying to find an SDP alternative – it looks like delaying the standard. If they would have done that alongside taking such a position, then I'd be more willing to take them seriously as well.
Microsoft – give me a show of faith. There are useful parts to WebRTC that aren't just video. Figure out which parts of it you're comfortable with and go implement it.
Tsahi Levent-Levi
Independent WebRTC analyst. I help companies ship real-time communications they can actually monitor. 20+ years in the comms space, last 13 focused on WebRTC.
More about Tsahi →
Nice article, I share the spirit of the post, but I think mentioning the SDP stuff is not fair. There are more people (in fact mostly independent developers) supporting the ORCA effort.
Gustavo,
I've been in standardization meetings in my past. The usualy way these are handled is that when one vendor is putting a suggestion on the table, its competitors will have the knee jerk reflex of opposting it - with good or bad reasons it doesn't really matter. It costs nothing to do that, and doesn't really put any commitment on the one making the objection.
Yes. There are problems with WebRTC that needs to be solved, but waiting for them to get solved in order to adopt WebRTC as an atomic component instead of starting to offer the parts of it where there is little to no objection has no merit.
Good post.
I didn't realise about Mailchimp, but I mused a while back (can't remember if it was a Tweet, a blog post or a report) about the potential for Facebook using GetUserMedia for the camera - seems ridiculous you can't do a "selfie" with a PC webcam directly from the page.
It's the sort of thing that would make me revise up my WebRTC active-user forecasts by 100million+, almost overnight.
As far as rumors go, Facebook are using WebRTC in their VoIP client in the US. So not something for a selfie but rather the whole shebang.
I am a little concerned, Dean, that you can no longer distinguish between a tweet and a report - ahh the travails of digital age! More seriously, it would seem that, if you are not careful, your WebRTC adoption curve is likely to asymptote to the HTML5 adoption curve as web-developers start to see all the various bits of WebRTC as just other composable pieces in the same toolbox along with WebGL, WebAudio, Canvases and any other HTML5+ construct. Then WebRTC-specific numbers/reports/tweets will, perhaps, face an existential crisis?
:-)
If you think that ORTC is an "alternative" to SDP then you understand nothing. Document yourself before posting.
I think ORTC is an attempt to fix how SDP is used in WebRTC and the headaches it causes developers. I also think that playing the standardization game is far from adopting a technology.
What headache? 99.9% of developers just transmit the SDPlookalike given to them.
SDP is just good for all that stuff already done with SDP (i.e. SIP and XMPP/Jingle), and those playing with WebRTC currently does basically that or something similar to that. And for telcos (which understand WebRTC as a new interface to their PSTN networks) that's good enough.
But when you start playing with server implementations to do something else than a boring WebRTC to SIP gateway, then you understand those "headaches".
Of course, if you just develop a web application where a web user can "call" another web user then you are fine with SDP. I hope we all understand WebRTC must be beyond that telephony stuff.