Posts

When will WebRTC be ready for prime time?

No matter what camp you are in, pro-SDP or Object API, WebRTC feels like it’s here to stay. The fact that there is so much push-back on an Object-centric API in the WebRTC working groups tells us that there are many who are concerned about delaying v1 any longer than is necessary, there is a feeling of urgency, which seems like a good thing.

As an application developer, one looks at the current state of WebRTC as a proposed standard and it soon becomes apparent that it is not trivial. Building a commercially viable application or service will take much more than a few JavaScript APIs leveraging the WebRTC stack in Chrome and Firefox.

WebRTC is in its early infancy, building a commercial service using WebRTC in the browser today is not only unlikely it might even be a little silly. That’s not to say that the work being done in and around WebRTC today isn’t valid, far from it. It’s just that WebRTC “in browsers” is far from being ready for any service to deploy commercially and scale.

Where are we most likely to see traction early on?

What WebRTC, or specifically Google, has given us is a great media stack that we can leverage in apps.

Mobile is where WebRTC can be put to work easily and quickly. Any iOS, android or even BB10 application developer can now make use of new mobile developer toolkits that include the WebRTC media stack plus other features like social identity federation, p2p signaling and the like. Mobile is where WebRTC technology shines today.

Building for mobile first is not just convenient it’s smart. By building on mobile ahead of the browsers the developer somewhat “future proofs” the life cycle of the product being developed. When WebRTC is ready for prime time, those mobile developers will be in a perfect position to deploy to the browser, potentially edging out their competition.

When will the browsers be ready?

It’s going to be the better part of a year (some say years) before some of the hard problems are solved in the respective WebRTC working groups and even then the v1 features supported will be somewhat minimal. Even I thought this would have been done by now, but we are a long way from done.

Questions remain unanswered:  Where is Apple? What will Microsoft do with CU-RTC-Web?  What video codec will be chosen as MTI?

Regardless of how much work there is left to do, the bureaucracy and politics inherent in the standards bodies will ensure that we will not see this in production across all major browsers anytime soon.

As an example, let’s take a look at XMPP. Before it became a standard XMPP was the Jabber protocol, an open technology used primarily for instant messaging and presence. Since Jabber was already defined and deployed heavily all across the globe you would think the standardization would go rather quickly, right? Nope. That process took more than 2 years in the IETF and very little was changed.

If you are a mobile developer looking to leverage WebRTC for a mobile project, you’re in luck! There are some good mobile SDKs available today.  You might want to hold off on deployment plans for all the major browsers anytime soon, especially not if you are focusing your efforts in the Enterprise markets where MS Windows and Internet Explorer still rule the roost.