The Politics of WebRTC

Suzanne Bowen : Monetizing IP Communications
Suzanne Bowen
37 yrs in telecom, teaching, blog & grant writing, biz development, marketing, & PR. Favorite moments in life involve time w/ family & friends, networking, IP communications industry verticals & horizontals, running, traveling, foreign languages
| 1. "Focusing your life solely on making a buck shows a certain poverty of ambition..." Barack Obama ..... 2. "One of the sad signs of our times is that we have demonized those who produce, subsidized those who refuse to produce, and canonized those who complain." By Thomas Sowell

The Politics of WebRTC

WebRTC, it provides the most potential for innovation and new business on the web, some say. The web does need a B12 shot. Wired magazine tops even Popular Science, the Geekspeak and VUC podcasts, and the Tuesday, Saturday and Sundy New York Times for some of my favorite news and information. The August 2010 issue had a highly provocative cover stating "The Web is Dead."

Dig deeper, and you would have discovered our digital world in a sort of time travel of a return to an AOL-type world where people leave "the wide-open Web to semiclosed platforms that use the Internet for transport but not the browser for display." It's not really exactly like AOL, but there are some similarities. Could WebRTC be the tool, the platform, the savior of the web for business and innovation and entertainment?


I have been eavesdropping on WebRTC conversations and studying what people are doing with WebRTC and wish to share some related issues and opportunities:

1. Standardization:

In 2012, Microsoft seemed to be quite pro - WebRTC, condoning the plugin-free technology for voice and video communications in a browser. The company wanted a different approach than that commonly condoned by other browser vendors such as Chrome and Firefox. Microsoft wanted to see a common standard. What standard that was ...?

Microsoft was pushing an approach called CU-RTC-WEB. CU-RTC-WEB provided a lower-level API that would be harder for developers to use but also would enable them to implement a wider range of applications (like Skype) with less "hacking" required. The consensus was that Microsoft made some very good points, but the WebRTC effort was too far along to basically stop and start again.

Recently Microsoft appears to have acknowledged that CU-RTC-WEB wasn't quite right either. The current consensus seems to be heading towards completing WebRTC 1.0 as-is but providing a lower-level API for WebRTC 2.0. Ideally, the existing WebRTC 1.0 API could then be implemented as a library on top of WebRTC 2.0 - making everyone happy. But the prevailing opinion (although there are some very loud naysayers) is to complete WebRTC 1.0 first.

The argument is that WebRTC 1.0 doesn't really do everything people now want it to do. But on the other hand, many of these requirements are quite new and WebRTC 1.0 does meet the original requirements.

Some say that Skype ending its app directory and that its desktop API might have enMicrosoft's attempt to steal WebRTC's momentum, and WebRTC sure does have momentum. I was at the WebRTC conference in Atlanta in June 2013. I did not see Skype or Microsoft there. It was hot, for example, that during hours of dozens of WebRTC presentations kept over 600 people glued to the edge of their seats. (Some of the many very cool companies include Mozilla, Elastix, Crocodile Talk, Augmedix, WebRTC Strategies, Dialogic, Quanta Computer Inc, Voxeo, and Digium.

BTW, Google started the whole open-sourced web in real-time communications effort in May 2011. The goal was to build a standard-based, real-time media engine into and compatible with all browsers.

2. Authentication Models and MSRP over WebRTC:

In late May 2013, Kamailio announced that Crocodile RCS's Peter Dunkley had created a new authentication model, aimed towards WebRTC authentication. The gentleman also wrote the first MSRP (session-based, data tranfer) over WebSockets. It is a natural companion to SIP and can be used for IM, file transfer or other data that needs streaming. The MSRP Relay sorts out NAT traversal issues. Innovations such as these have made it much easier for web developers with little or no telecommunications experience to work with WebRTC. 

As an aside, WebSockets and WebRTC are separate things but very complimentary. MSRP can be used over WebRTC too (various people - including Crocodile) are working on it, but it doesn't have any authentication requirements.

3. To federate (wall up) or not?

Shouldn't there be a choice? Some platforms enable federation, no federation or the choice of both. (Does anyone have a list of those that offer any of the three?)


With Crocodile, whether you federate or not is up to you ... at least they give you the option.

Aswath Rao said on July 25, 2013 "WebRTC has built-in security systems to protect the end nodes from malware apps. The browser ensures that a rogue app can not assume control of the end node ... WebRTC does away with federation is one of the important benefits and why it is going to disrupt communications industry." 

This could be termed an over-simplification. Perhaps there is no need to federate but the does not prevent federation from occuring. Some people are just against federation and some are for it. (They each have business cases that are built on one or the other just as in other areas such as wholesale SIP DID on something like XConnect vs DIDX vs Voxbone.)

4. IMS with WebRTC or no IMS? I'm actually looking for more input in this area.

IMS defense - Don't hardwire all OS behind ________________. (Fill in the blank with the word or words that makes the most sense.)

It is possible to combine WebRTC and a Rich Communication Suite. How? Why?

5. Should WebRTC and RCS be over the top or carrier-initiated? Hmm ...

6. More WebRTC ebates ...

a. "SIP is hard" debate: SIP is hard. So make it easy, just use a free library.

b. "Signalling is not required with WebRTC" debate

Argument of one side: Signalling is required. To establish communication two end-points (browsers), they must exchange some information (IP addresses, codecs supported, etc). To do this you need signalling. You can do this through a proprietary REST mechanism instead of a traditional signalling scheme if you want, but that is still signalling - just proprietary REST signalling.

c. SDP or no SDP? Some say make a decision and move on because oh my goodness, products need to ship.

Response from industry experts: This is the WebRTC 1.0 vs WebRTC 2.0 issue and at the heart of Microsoft's objection. It appears highly likely that WebRTC 1.0 will be SDP based.

d. Video codecs ...

H264 ... Sailfish ... Cisco, Microsoft, Apple, mobile operator licensing

VP8 ... quite innovative for small startups

Some say that the video codec debate is political not technical. The debate is over the Mandatory To Implement (MTI) codec. This means it is not exclusive. If VP8 were MTI, there is nothing to stop someone else adding H.264 or anything else as well. It is simply about what codec must implement everything, but it in no-way forces you to use it as you can add other codecs if you want.

This post is meant to provoke discussion, disagreement and other kinds of conversation and study.

Enhanced by Zemanta


Related Articles to 'The Politics of WebRTC'
Plivo_diagram.png

Featured Events