[itvt] ITV Interview: Jaison Dolvane, Co-Founder, President and CEO, Espial

At the TelcoTV show in Dallas earlier this month, Canada's Espial unveiled a product called Evo Server, which, according to the company's co-founder, president and CEO, Jaison Dolvane, puts it "directly in competition with companies like Microsoft" by allowing it to offer an end-to-end, client-server IPTV architecture. [itvt]'s Tracy Swedlow recently spoke to Dolvane about Evo Server and about the "Future-Proof Framework" that Espial also unveiled at TelcoTV; about the company's strategy for competing against Microsoft; about his assertion that Microsoft's IPTV solution has inherent scalability problems; about Espial's future IPTV plans and more.

[itvt]: At TelcoTV you announced a product called Evo Server. What does this new product do, and how does it relate to the Evo product that you launched at the IBC in September?

Dolvane: What we launched at the IBC was a new version of what we're now calling Evo Client. Evo Client consists of our client-side applications and middleware. It's a full-service client-side middleware solution, in which a browser is baked in, so to speak, as an embedded application, allowing you to do things like VOD, portals, walled gardens, etc. We've offered the browser element of Evo Client for some time now, though branded as Espial Escape. It's now been rebranded Evo Browser and is an application within Evo Client.

Now Evo Server, which is the new product we launched at TelcoTV, is something that puts us directly in competition with companies like Microsoft, in that it means we're offering an end-to-end client/server modular architecture. Evo Server is an IPTV middleware solution that integrates and efficiently manages data communications between the headend and set-top box, data that comes from EPG services, VOD servers, CA/DRM, softswitches, etc. In addition, Evo Server does pricing and packaging and integrates with billing and OSS systems. It also helps to manage CPE and monitors IPTV performance and service-impacting faults. It is a true carrier-grade and reliable system, fully redundant with no single points of failure, in-service upgrades, and it scales up to 10,000 set-top boxes per server.

[itvt]: Now I've always thought of Espial as offering various individual components of an IPTV system that could be integrated into third-party IPTV platforms. Why did you decide to start offering a whole end-to-end platform?

Dolvane: Well, the Evo Server is actually still another component in our portfolio of components. What potential customers in the market are actually looking for now is middleware--client middleware and server middleware. But what we did was to take a different approach to the marketplace than most of the other middleware vendors out there, in that we've actually stuck with our componentized approach, and added another modular component to our portfolio.

So, we can provide a componentized approach, where we can sell one or more components to telcos and OEM's. Or we can provide a pre-integrated solution, which we can go in and combine with DRM, VOD servers, encoders and set-top boxes. We've actually also created a pre-integrated reference architecture for those kinds of deployments.

At the end of the day, we are all about listening to our customers and partners. Several of the pain-points that IPTV operators feel--scalability and reliability, openness and modularity, and fast end-to-end performance--were just not being met. We consulted many of our customers and partners and decided it was the right thing to do to build Evo Server. As you said, many partners and operators use Evo Client and Evo Browser today and many of them have developed their own server middleware components or integrated with third party components, so not all of them will upgrade to Evo Server right away or ever. But we believe in the value that Evo Server brings to the industry and we remain committed to supporting all of our customers and channel partners with whatever IPTV components they may need.

[itvt]: I'm assuming that this is an open environment…

Dolvane: Yes. It's an open architecture. As part of this open architecture, we launched something called "Future-Proof Framework," which is basically a set of tools and SDK's and API's that enable third parties to build services and to actually integrate into various different networks. These toolkits provide streamlined ways to integrate VOD and DRM servers, port to new set-top boxes, create new services, and create user-interfaces we call "SkinTones," which offer incredible flexibility for service-provider branding and differentiation.

Now I should stress that Evo Server is not an immature product. We've actually had it for quite some time and it has already gone into trials. We've completed successful pilots with a couple of customers. Those customers are Auroras in the US and UPC Sweden, which is now called Com Hem. Auroras is a wholesaler and integrator--they're sort of an ASP/integrator for IPTV services. They're going after the tier-two and tier-three markets, and they're providing content as well. They were one of our launch partners at TelcoTV: together we demonstrated Evo Server and Evo Client integrated into a full, end-to-end--i.e. headend-to-home--IPTV system, which streamed content via a satellite feed that was set up at the show. So that demo was of the full, headend-to-home solution, including content ingestion. Com Hem, meanwhile, is the largest cable operator in Sweden, with 1.7 million subscribers. In addition to Espial's Evo Client and Evo Server products, UPC Sweden/Com Hem and its partners used the open and modular toolkits of the Future-Proof Framework to significantly reduce the cost, complexity, and time of the deployment--which only took four months. Other ecosystem partners included set-top boxes from Tilgin, encoders from Tandberg, VOD servers from Kasenna, and a proprietary DRM system. System integration with ecosystem components and backoffice billing and OSS was completed by local Swedish integrator Kentor in less than four months. UPC used the service-creation environment of the FPF to create SkinTones customized and localized to Swedish look-and-feel, language, and programming data, all of which made their IPTV service identical to their existing cable TV service.

[itvt]: You mentioned that Auroras is a wholesaler of IPTV services: have they sold Espial Evo-based IPTV services to anyone yet?

Dolvane: Yes, they have three customers that they've already been in trials with for some months now.

[itvt]: Can you say who those three customers are?

Dolvane: We can't announce them because they're still in trials right now and really, it's up to Auroras when they want to announce who the specific customers are. Based on the traffic at their booth at TelcoTV, I expect Auroras will be signing up many more customers looking for an end-to-end IPTV middleware solution that is pre-integrated with content, STB's, and VOD.

[itvt]: You say that the launch of Evo Server puts you in direct competition with Microsoft in the IPTV space. Could you talk a little about how your end-to-end IPTV solution is differentiated from Microsoft's?

Dolvane: Given Microsoft's ambitions to own a large part of the IPTV value chain, it is difficult not to compete with Microsoft at some level. Evo Server is a different approach in that it's an open and modular componentized architecture, compared to the model of a "one-size-fits-all" monolithic architecture that Microsoft has. Using our Future-Proof Framework, it enables telcos to create their own services, and integrates with various IPTV components so you can build your own best-of-breed ecosystem for VOD, DRM, ad servers, billing, VoIP, etc. The FPF also provides a set of tools that allow set-top box manufacturers or integrators to port it across multiple devices on the client side. This platform portability and open modularity has been really lacking until now.

So its open architecture is one of the key differentiators. Another key differentiator is scalability. The way we've architected this product, it supports 10,000 to 20,000 subscribers on one server, whereas Microsoft is more in the range of 200 to 300 subscribers per server (or two), and so it requires multiple servers even for a small deployment. The way our architecture is set up means that it's basically infinite: you can keep adding new servers without bottlenecking the system. Since it's modeled after fully redundant and highly available telco networks, each new server you add to the cluster backs up all the others so it actually becomes more reliable as the network grows.

We understand the importance of redundancy because of our pedigree in the telecommunications space. A lot of people who work here used to work at companies like Nortel, Siemens, and Alcatel. So we understand the carrier market--the carrier world--whereas Microsoft's history has been tied to the enterprise world.

Our client middleware is also a huge differentiator from Microsoft. On the client side, we've heard that Microsoft's client middleware environment is a very difficult environment to integrate with: it takes really long and requires a lot of handholding from Microsoft. A recent article in the Wall Street Journal suggested that Verizon had to put hundreds of their developers to work in order to fix or rework Microsoft's client applications to meet Verizon requirements. Compared to this, Espial's Evo Client middleware has a very easy porting layer that we have hardened, and already enabled over 20 different set-top box manufacturers to do the ports themselves. We also provide an open and standards-based service-creation environment to enable creation of set-top box independent services. This SCE has already been used by several application developers to create new services like games, messaging, karaoke on-demand and other interactive applications.

On the server side, we also have a very open environment: the Future-Proof Framework provides a set of open API's, or "adapters" as we call them, that allows integration with third-party best-of-breed VOD servers and DRM technologies. This is in contrast to Microsoft, which seems to be trying to build all that stuff into what you might call a monolithic block of code. So, in other words, they want to control all that, versus letting best-of-breed components come together to build a solution that is not only going to be technically feasible to deploy, but that is truly going to hit the cost points that carriers are looking for. And remember, openness is not a meaningless concept. Without it, carriers have no leverage to negotiate the best available technology at the right price. Just look at what Windows did to the PC market.

[itvt]: Yet Microsoft has secured deployments with many of the world's largest telcos…

Dolvane: Yes. Telcos are looking for a single source with the scale to deliver and support an end-to-end IPTV solution. Until recently--and even though it's not the best technical solution--Microsoft has been the only provider in the market with the scale and demonstrated commitment to provide a solution. In addition to this, Microsoft has deep pockets and has won business by subsidizing the economics. To offer an alternative to Microsoft, Espial is working with several partners now with the scale to provide a single-source, end-to-end IPTV solution to telcos that includes Espial's strong technical solution, with all the benefits of an open, modular and standards based approach.

All these large telcos have been taking a very long time to roll out their IPTV services, and that's been because of issues with Microsoft's platform. Now some of those telcos are finally beginning to deploy some Microsoft-based IPTV services, but those services are going to have a difficult time scaling, once they get to a certain threshold. There are some large architectural changes that Microsoft is going to have to make to its platform over time to enable those telcos to break the barriers of mass market. On the other hand, a lot of the smaller telcos who are not using Microsoft's platform have already been offering IPTV services for some time now with good success. Don't get me wrong: I do believe Microsoft will get parts of the solution right in time. But to believe you can incrementally tweak performance and scalability gains out of a heavy client architecture and non-scalable server architecture is just not realistic.

At the same time, remember that there are some very large and successful IPTV deployments around the world that aren't using Microsoft's platform. For example, Espial already has 500,000 IPTV subscribers in deployment. We also have several tier-one customers who have exceeded 100,000 subscribers, including NTT Communications in Japan. In fact, there are several large incumbents in Asia that are already deploying IPTV services: PCCW in Hong Kong isn't a customer of ours, but they're already delivering IPTV services to around half a million users. There's also Belgium: Belgacom, which is that country's incumbent telco, now has 100,000 subscribers. They're using our Evo Browser product in that deployment, which, as you know, was integrated by Siemens and uses Siemens' Myrio middleware.

[itvt]: Let's talk about some of the interactive TV applications that Espial Evo supports…

Dolvane: Yes, we've got an EPG. We've got VOD. We've got DVR on the near horizon. We've got interactive Web services, which are part of the Evo browser. We offer simple walled-garden content applications--we don't develop those ourselves, but we provide a platform for them. You're probably already familiar with some of our partners for those applications: companies like Iacta and Accedo Broadband. They're currently providing us with various small applications for our platform, and we plan to expand our developer community over time. Also, our Web browser provides full Internet-browsing capabilities with Flash--and, if required, Java applets.

[itvt]: Now I understand you're claiming your EPG offers some unique capabilities…

Dolvane: Yes, we've done some pretty intelligent things around how we manage the data, how we refresh the data and make it available to the end-subscriber, between the client and the server. We've basically removed the latency to get to that data--a latency that in the past has been a problem with the client/server model in the IPTV space. I can't disclose exactly how we manage the data because it is special sauce, but it is a combination of how we use our IP data carousel to coordinate the data from various applications, how we separate data management from UI graphics and files, and how we manage the local data environment in the set-top box, that enables super-fast navigation and searches. The quality of the subscriber experience is instantaneous in most cases. As you know, the slowness of the interface has been a real drag on the TV industry in general, and the stakes are raised a lot higher with the increased level of interactivity that IPTV entails.

Another thing that's pretty unique and that's an inherent part of the entire Evo IPTV Service Platform is something I mentioned earlier: what we call our "SkinTones" concept, which allows you to replace the entire look-and-feel of the subscriber interface on the fly, without interrupting the video: so a family can have one guide for the kids, with parental controls baked into it, and another guide for the parents. Or operators can use this feature to insert certain kinds of advertising into the interactive elements of the user interface. How we've accomplished this SkinTones concept is by some more secret sauce, whereby applications and user interfaces are still written in HTML and JavaScript, but there's no overhead associated with building applications: basically we've eliminated the latency usually present in a browser-based model. Changing the look-and-feel of the user interface is something that telcos can do remotely, without interrupting the video experience.

[itvt]: So the ability to change the look-and-feel of the UI is something that you're targeting at the operators rather than directly at the consumers?

Dolvane: You can do it both ways. It really depends on how the telco wants to offer the service. I think in the short term--and this is just my personal opinion--telcos will have a set of pre-canned guide screens that they deploy themselves, and then, over time, they may give a bit more artistic license to the subscriber to make their own selection or even add elements to the SkinTones. There is no technical reason the carriers won't be able to offer subscribers a MySpace.com-type of IPTV experience some day.

[itvt]: To what extent are your customers using your browser's capabilities to bring Web content to the television?

Dolvane: Right now, I think most of the deployments are still focused on building walled-garden content within the browser environment--which is still Web-services based. But I think there are certainly some folks who are talking about, for example, how you could have an eBay-type bar that would be available inside the user interface to provide a blended entertainment, communications, and shopping experience, so that people could even see options to purchase products in the TV they are viewing. I think people are also looking at making it possible to access over-the-top video content like YouTube, or audio content, over the Internet--through Web content or Web services that are already available.

So I think that kind of thing will happen--in fact, I think it's being done in a small way right now. But remember that telcos' primary focus is simply trying to launch a TV service. Then, once they've done that, our platform will allow them to rapidly and cost-effectively add a lot of new, interactive services over time. And that's certainly something that's very exciting to them.

[itvt]: Now, I understand that you've demo'd your Evo platform working with high-definition set-top boxes. Do you feel that the platform offers any differentiating features when used in HD environments?

Dolvane: Certainly, I think our SkinTones are especially suited for the HD environment: you'll be able to have some pretty sexy HD user interfaces that can be switched on the fly--and, in fact, you could incorporate more artwork into the different skins to take advantage of the HD environment. One nice thing is that, if the end-subscriber changes from SD to HDTV, it will be possible to switch their user interface from an SD to an HD version without either the operator or the subscriber having to go through a long process. Also, the granularity of HD allows the guide to contain more information at the top-level without looking squished, and the flexibility of SkinTones enables customization of the guide to benefit from this, which also further improves the speed of navigation and "time-to-content," which is becoming a very important IPTV metric.

Another advantage of our platform in HD environments is that the Evo Browser provides support for AJAX, Flash, Java Applets, and a lot of the new Internet standards: the kind of PC-type browsing experience that it enables will be a lot more interesting on a high-definition TV set than on a standard-definition set.

We demo'd HD at TelcoTV with HD set-top boxes from Tilgin and Scientific-Atlanta. But we're actually working right now in the Asian marketplace with a dozen other set-top boxes that are supporting HD--for example, set-tops from DASAN Networks in Korea.

[itvt]: Does your platform support single-screen enhanced TV applications--so broadcast-bound applications?

Dolvane: Triggers in the stream, you mean?

[itvt]: Yes. Triggers in the stream.

Dolvane: It will do shortly. It has the capability to support those kinds of applications. As you know, there's a lack of standards for doing this with IPTV. However, we have figured out a way to do it. We plan to enable that kind of red-button functionality in one of our future releases. In fact, we already have a couple of customers who are using some of that functionality in the form that it's in right now.

There is actually a group within the DVB that's discussing standardization of triggering mechanisms for enhanced TV apps in IPTV environments. But I think the reason why there hasn't been a greater push for this kind of thing to date is that the telcos are still just learning the industry: a year or two years ago, they didn't know how to develop content relationships. Well, they seem to have figured that out now, but they still tend not to know how to efficiently deploy TV services yet. Though some of them already seem to have figured that out, and the others will soon too. Remember that the IPTV industry as a whole is still pretty nascent: there are currently around 3 to 5 million IPTV subscribers in total around the world. So I think you'll eventually see the emergence of enhanced TV services on IPTV: it's just a matter of the evolution of the marketplace, and of the learning curve that telcos have to go through.

[itvt]: Can you say which of your customers are already using your proprietary triggering technology?

Dolvane: I'm not at liberty to say who they are right now. However, in Siemens' Belgacom deployment, which uses our Evo Browser, they've built some interactive applications.

[itvt]: Speaking of Siemens, what exactly is your relationship with them?

Dolvane: Siemens, as you know, acquired Myrio, and various Espial technologies are part of the Myrio stack. So, when Siemens wins deals, they bring us in. This has been a critical partnership for us: I think that, between us and Siemens, there will be some interesting innovations that we'll bring to market. I think there's a good co-development relationship there.

[itvt]: Are you planning to announce a developer group for the Evo platform?

Dolvane: Yes. Eventually we will. The plan is to set up a formal application community around it. But we haven't formalized anything on that yet. So I would say, "Stay tuned," and that you might see a formal announcement in the second-half-of-2007 timeframe. However, we do already have an informal application community--a dozen or so partners who are building applications for the Evo platform.

[itvt]: What's your impression of the growth-rate of the IPTV market?

Dolvane: I think we're seeing good subscriber growth now in the IPTV market. We're still probably about a year from getting to the 100-150% growth rates, but there's certainly a ton of pent-up demand, and there's a lot of activity going on. It's a matter of telcos successfully doing some implementations. My general comment on the market would be that it's tracking, and that, if anything, it's going to accelerate over the next year or so.

[itvt]: What kinds of new product announcements should we expect to hear from Espial in the coming months?

Dolvane: Oh, there's a lot on the roadmap, and some priorities that are still being decided based on customer need. There are things we are doing in the triple-play and mobility space that will be pretty neat. We are gearing up for new IMS requirements that will emerge for the IPTV space, just as they are for the telecom world. Japan continues to be a huge market for us and there are special standards such as BML and IPSP that we are tracking to. You will continue to hear about things that we can do with our open and modular Future-Proof Framework, that we expect to solve even more pain-points in the industry.

[itvt]: What kinds of telcos or other companies will you be targeting your end-to-end platform at?

Dolvane: Our whole Evo IPTV Service Platform architecture is designed for tier-one telcos and other broadband providers, so that will continue. We will continue to provide modular client or server components to software and hardware OEM's that are trying to put together various pre-integrated end-to-end IPTV solutions for operators big and small. Finally, you will also see us building increased channel partnerships with companies that have the scale and size required to provide tier-one telcos with "single-source," end-to-end IPTV solutions as a true alternative to Microsoft.

[itvt]: Finally, what's your strategy for taking on Microsoft in the IPTV marketplace with your new solution--in that, obviously, Microsoft has much greater resources than Espial?

Dolvane: Microsoft's ambition is to absorb all the value in the IPTV value chain into the Microsoft platform, to the extent where their monolithic architecture makes it very difficult to expose any of the benefits provided by other software and hardware components--network gear, set-top boxes, etc.--in the network. This would be OK, if indeed they were able to provide a best-of-breed solution. However, Microsoft has hit several hurdles with delivering a quality, scalable solution to date. The reason for this is a whole 'nother topic, but simply speaking, the ambition to try and do everything in software and expect that hardware across the network and set-top boxes will catch up like it did in the PC world makes little sense, given the difference in markets, economics and the time-to-market pressures for telcos. Further, given the early stage of the market and the amount of innovation still required, together with the hyper-competitive service-provider environment, it's difficult for us to see how Microsoft's approach of build-everything-in-a-closed-environment is a winning formula.

Espial's approach is to provide an open, modular IPTV solution and, through our Future-Proof Framework, to enable various ecosystem partners, OEM's, integrators and channels to add in the value and benefits of the innovations in their products and services. Further, we understand the need for some telcos to have a single vendor deliver and support the solution. To address this, we will be working with large channel partners, in order to provide "single-source" options for an end-to-end IPTV solution that includes one or more components of our Evo IPTV Service Platform. Working with a strong partner ecosystem, I believe we can innovate, integrate and deliver a better alternative to Microsoft that will provide a long-term winning solution to telcos, and enable them to compete successfully in their geographic markets.

We are already in touch with many of the tier-one operators who have done the dance with Microsoft and are looking for an alternative. In addition, we are working with many of the large telecom equipment manufacturers and service integrators that also have the need to defend their value propositions and are building their own open, best-of-breed solutions using the Evo platform. 2007 is going to be a great year not just for us, but for the entire industry, because it will see titanic shifts in the current paradigms and partnerships for delivering a complete, carrier-grade IPTV solution. And I also predict it will be the year when tier-one service providers and telecom equipment manufacturers alike finally get the courage to turn their backs on Microsoft, before it does what it did to the PC industry.

URL: Espial

Originally Published: November 27, 2006  in [itvt] Issue 7.06

Click: http://www.itvt.com to subscribe to our free email newsletter, which contains all the news stories you see on this Web site, and additional breaking news and scoops, in-depth features, interviews, screenshots, videos, and other exclusive content you will not find anywhere else.

[itvt]'s

Early Bird Registration is now open:

http://www.thetvoftomorrowshow.com

March 13-14, 2007

San Francisco, CA