I recently spoke with Brett Shockley (Senior VP and CTO Avaya) about WebRTC, an emerging Web technology showing great disruptive potential. Brett has responsibility for corporate strategy, business development, Avaya Labs Research, and the Emerging Products and Technology division.
IJ: What are the primary WebRTC use cases for Avaya?
Brett: I would put them in three categories:
- Helping the outside world connect with users in an enterprise. Good examples include conferencing applications and multi-media customer service applications. People don’t want to install plug-ins or to download apps just to join a conversation. We have been delivering these kinds of applications with voice and video using Flash for the past couple of years, but we expect these applications to transition to WebRTC over time.
- Extending the enterprise. Remote workers want to be full participants in the enterprise collaboration and communication environments that they would use in the office. Enterprises want to enable contact center agents to be able to work from anywhere in the world. These use cases demand tight integration into rich enterprise services, with all the requisite security. These applications exist today, but WebRTC will simplify and expand these use cases.
- Collaboration. Communities of interest such as work groups or social circles want low-latency means to communicate with each other. They aren’t necessarily concerned with being connected to the rest of the company or the world. WebRTC lets you do this and avoid some of the broader enterprise integration challenges, so I think we will see this functionality integrated into both business and collaboration applications.
IJ: Customer service use cases seem to be important in the WebRTC story.
Brett: Yes. Enterprises are increasingly trying to fine-tune and deliver a consistently branded customer experience. And they want to provide that experience in whatever way the customer approaches the company. Our research indicates that 78% of consumers want to have multiple channels to reach into a company. A customer might start with a site’s knowledge base, need more help and move to a virtual agent, then escalate to a human chat. We would like to be able to provide voice and video interactions as well. WebRTC lets you bring together a number of different elements all through this customer interaction lifecycle. You can manage 50 million or more customer service calls a year on the backend with Avaya Contact Center solutions. When you marry this with WebRTC, you have a great combination that lets customer’s anywhere reach into your enterprise with just a browser. The browser now becomes a rich channel that enables the full conversation – context, screen sharing, voice, video, chat – with the contact center, enabling speedier resolution to problems.
IJ: Please tell me more about the video conferencing and collaboration use case.
Brett: Today we do voice, video and web conferencing from virtually any device using downloaded client apps or Flash in a browser. We’ve demonstrated the benefit of being able to link people on their smart phones, tablets, PC’s and Mac’s with each other and back to video conference rooms paired with rich content. But now people want more richness, more quality, and no download. WebRTC is the natural next step.
IJ: How far away are we from having those clients?
Brett: You will likely see apps in all of these use cases within the next year. Of course, it will take a while for things to spread widely. Because of our work with Flash, we can support a transition path that includes both legacy and new browsers.
Brett: Avaya has a great innovation process for bringing new technology like this to market. Our Emerging Products and Technology team starts with a small set of customers and innovates together, figuring out the required features, delivers the benefits to a group of beach head customers and then feeds it into the business unit for globalization and scale. WebRTC lends itself nicely to that approach.
IJ: What makes it so good at binding things together?
Brett: Part of it is the work we’ve done to build abstraction layers. In the old days the stack was rigid so it was hard to introduce new components. With today’s client technologies we build abstraction layers and APIs. So whether I want to build a soft client on my Mac, a tablet or a smartphone, we’ve already built the platform to support a rich set of services delivered through the presentation layer on those devices.
IJ: Do you find those APIs and abstraction layers are making it easier for people to build apps?
Brett: Our flagship Avaya Aura Communication system is built around the IP Multimedia Subsystem (IMS) standard leveraging SIP and Session Management. One of the great things about IMS is that it was designed to be tolerant of 3rd party apps — that is, apps that might be able to “bring the whole thing down” in a traditional digital or VoIP communications solution can now be safely connected to Avaya Aura. When we built our enterprise version of this IMS system to tie everything together, we faced an interesting balancing act: how to dramatically expand the ecosystem of developers to create applications, but preserve this core reliability of the IMS infrastructure. The solution is open, so in theory anyone could create a JSR289 based application and lets users subscribe to it and that might be useful to a certain segment of the market, but for most people it’s too much low-level programming. We knew that if we wanted to attract people who are writing hard-to-integrate apps we would need to abstract out the most widely applicable use cases and make them easily accessible. That is why we’ve been working on a middleware solution (“the Collaboration Environment”) that aligns better with Web developer skills.
IJ: Do you think that this middleware solution will lower costs because you can draw from the larger Web developer pool?
Brett: Absolutely. When compared with other areas of the technology industry, the communications industry has struggled to benefit from the trends that encourage startups. In communications, there can be high equipment purchase costs and complex, low-level, synchronous programming interfaces. Computer telephone integration (CTI) is not the sort of thing you would be taught in school, and that has limited the growth of apps in the communications world.
Brett: But now, we have an architecture based on IMS and a middleware toolkit, so three smart college kids can create an app and make it available in the cloud. Through our “Collaboration Environment” we can provide access to backend services via the WebRTC apps. If you want to interact in the world, you need to connect . Also, IT developers can now quickly extend business apps to include interactive communications, enabling more efficient business transactions.
IJ: Do you see this applying to the Internet of Things as well? Or a Web of Things, where the Web is the application layer for interacting with a wide range of connected devices?
Brett: The whole idea of Internet of Things fits into monitoring and managing. The telephone industry has evolved from a world of tightly coupled black boxes to a much more distributed system. A distributed system lets you build amazing things, but to diagnose issues you need to be able to monitor what’s going on. We have started to deploy sensors into the devices that plug into a heterogeneous network to diagnose problems. When we first plugged in these sensors we noted 75% of the dots were red — lots of problems had gone undiagnosed. We discovered all over the world there were issues, due to carrier upgrades for example. You would call a carrier and say “our sensors tell us you’ve got this issue” and they would resist, but ten minutes later the problem would be magically fixed. Those kinds of apps become super-charged when you think about feeding that data back into an Internet of Things infrastructure. We do all this through open standards.
IJ: Can you tell me more open standards success stories?
Brett: The monitoring example is one. Another is SIP. You can walk into a heterogeneous environment, have all the systems work together, and plug in soft clients and hard devices into session management systems. There is no way we could do that without the open standards. We provided services for the previous winter Olympics and will do so for the next one as well. We estimate 50,000 individuals from 80 countries with around 150,000 devices will show up at the events. We will see every flavor of device and we have to provide service to enable them. This will be the first BYOD Olympics. Because of the infrastructure we’ve put in place, we’ll be able to identify, secure, and support these 150,000 devices. It will also be the first time the IP protocol is used to transport video around the venues. We need to support all these devices and it’s all based on open standards (including VoiceXML).
IJ: What advice would you give W3C for our work on standards?
Brett: There’s always a tradeoff between producing a technology quickly and getting everybody to agree. I suppose one benefit of doing a good job of abstraction and not trying to build too much into any one standard is that you can move faster. In some cases, standards are not yet mature, so we have to extend them ourselves. For WebRTC, in the public-into-enterprise use cases there is already a good set of features to leverage. But if you wanted to extend the unified communications environment you would still need to build plugins to capture rich enterprise features.
Brett: Standards bodies could also do a better job ensuring that standards are applicable to business use cases. We need closer connections between engineering and business.
IJ: What technology obstacles do we need to address?
Brett: One is codec alphabet soup. In the real world transcoding happens all over the place. We could improve that situation by figuring out in the standards how to line up codecs to get the best results. Another issue is firewall traversal. And even though WebRTC allows integration, it can be challenging to build cool apps on top of services whose features evolved for decades. There are a lot of interesting challenges tying everything together.
IJ: Thank you for your time!