A common refrain among video conferencing advocates is that making a video call should be as simple as making a voice call. That is, users don't pause before dialing a number to consider if the person they're calling has a Cisco or Avaya phone or whether they can call a mobile phone from a landline. But such has been the general state of video conferencing over the years.
Although there have been notable advances toward interoperability, the challenges have only grown more complex as different types of endpoints continue multiplying and vendors scramble to support the latest operating systems and device manufacturers. But there is one endpoint that practically everyone with an Internet connection has access to: a Web browser.
Polycom, the October winner of SearchNetworking's Network Innovation Award for its RealPresence CloudAXIS Suite, recognizes the browser as the "lowest common denominator" among all users, making it an ideal client application for business-to-business and business-to-consumer video conferencing. Ensuring an enterprise-grade experience across a browser is no small feat, however.
Polycom CloudAXIS is a software product that runs on commodity servers and supports video conferencing and collaboration across any type of endpoint, including Web browsers, for users inside and outside corporate firewalls. Polycom has previously sold (and continues to sell) mobile and desktop software clients. CloudAXIS expands video access by enabling users to send external contacts, imported from Facebook or Google, a link to a webpage where the only requirements to join the video call are a webcam and a downloadable plug-in.
SearchNetworking features writer Jessica Scarpati spoke with A.E. Natarajan, Polycom's executive vice president of worldwide engineering, about CloudAXIS and the future of the video conferencing market.
Where did the idea for Polycom CloudAXIS come from?
A.E. Natarajan: The advent of our vision is video ubiquity, which essentially means you should be able to have multimedia, communications and collaboration capabilities with any device, in any place and with anyone. When we started embarking on this and started putting video conferencing software on iPads, iPhones, Android tablets and everything else, it became even more evident that the lowest common denominator was the Web browser. The ability for us to enable [those sessions] with anybody who has access to the Internet and a Web browser, to get into multimedia communications in a secure fashion and engage with all of the capabilities of Polycom video conferencing and collaboration from, and to provide that level of quality -- that was the vision and the driver for CloudAXIS.
Once you reach to a browser, you want to be able to support millions and millions of users, and … to do that, you need to have this infrastructure that is virtualized in software and scalable.
What does the 'cloud' in CloudAXIS refer to?
Natarajan: There are two parts to it. The first [thing that cloud refers to] is the video conferencing capability -- we want to be able to reach anybody with a browser, to embrace any of your contacts that are not necessarily in Outlook or [a personal contacts list] but are in Skype, Google Talk or any of those things, and to be able to actually connect and drag them into your conference.
The second part [of the reference to cloud] is the infrastructure behind the scenes that can enable, in a scalable way, that software in a virtualized environment so it can [support] this at a much larger footprint than what [is possible with software] today. Because once you reach to a browser, you want to be able to support millions and millions of users. To do that, you need to have this infrastructure that is virtualized in software and scalable. That brought us to taking our infrastructure, which is our RealPresence platform, and moving it over to a cloud entity. And that's how the cloud [portion of CloudAXIS] came about -- the RealPresence platform and infrastructure can be placed in a cloud in any data center in any geography, and it can be provisioned and orchestrated both in private clouds and public clouds and in hybrid models.
Were there technical challenges in getting browser-based video conferencing to perform like room-based systems?
Natarajan: First and foremost, there's the [task] of taking what is done on specialized chips -- video signal processing [chipsets], or VSPs, that encode, decode and do fancy things with video and audio as a pile of bits -- and [replicating those capabilities] in software on commodity [servers] like x86s so that you can actually deploy [video conferencing] in software.
On the client side, another big challenge is the ability for you to understand the various APIs [application programming interfaces] and protocols from which you can derive your contact list -- from Skype, Google Talk, LinkedIn or Facebook -- and unify them and give the user who's using CloudAXIS an ability to connect any one of them into a multimedia conference.
Does CloudAXIS use WebRTC? What does or would it enable?
Natarajan: When we embarked on CloudAXIS, the W3C standards body had not embraced multimedia communications inside of a browser. It's still a work in progress with Google and others who are still working through it. Since they had not embraced the standards or enhanced the browsers at that point in time, we took our client and plugged it into the browser.
What we find now is WebRTC gives the browser a standard framework [for knowing] how to actually look at the environment and embrace devices that can help with multimedia communications -- like a mic, speaker, display [or] camera. These are embedded frameworks within the browser that is provided by the standard browser in WebRTC. Somebody still has to write additional code on top of this framework to build a portal that can do multimedia communications. Now, with the browsers actually providing those capabilities, those things will now become part of the browser, and we will embrace WebRTC technology as we move forward.
[However,] the standards for WebRTC [vary] slightly from the standards that we're using. The encoding standards for WebRTC are VP8 and VP9, though the signaling [we use] is the same. But the encoding and decoding mechanisms are slightly different.
Where will further innovation occur in the video conferencing market?
Natarajan: There are multiple places where innovation will happen. First, we need to have a core site that can embrace all these endpoints that are not 100 or 200 rooms, but millions and millions of users. The scalability that you need to have [for that] global footprint, the ease of doing it and ease of provisioning it from the core site is [all] one part of it.
The second part of it is … [figuring out] how you can actually have a collaborative session with both synchronous as well as asynchronous video communications. Let's say we have this video that we record and then we store it away. You pick up and say, 'I want to listen to what A.E. said about WebRTC. Can I go back to that video and replay just that section?' [That requires] the heuristics, the analytics, the ability to search, the ability to do speech-to-text, the ability to do content management, the ability to provide a large variety of collaboration tools. What if you could take that same video, replay it, but mark it up with your commentary on top of it and then ship it off to a client who analyzes it differently? Those are a whole bunch of spaces where a lot of research and innovation can happen.
This was first published in October 2013