Spanning the Globe: CED’s 2011 CTO Roundtable
Big and small, here and abroad, cable’s top execs are pushing the frontiers of technology.
With increased video competition, a range of new partners, the deployment of new technologies and services, and subscribers pulling vast amounts of video content into their homes, there are a lot of engineering balls that cable engineers need to keep in the air these days. The real trick is serving all of the subscribers’ current whims while also planning for future network needs. Not to worry though: CED’s roundtable of chief technology officers has the answers.
CED: What are the top three things on your engineering to-do list this year and over the next five years?
Balan Nair, CTO of Liberty Global:
Deliver new products, new user interface, wireless MVNO and online streaming. Increase capital allocation toward more product innovation now that network upgrades are coming to an end. Focus on network reliability.
Dennis Steiger, group vice president of engineering at Shaw Communications:
Well certainly this year our big one is launching our next-generation set-tops, although we’re kind of moving away from that set-top terminology with what we’re doing. We are launching some new tru2way set-tops – that’s a pretty big focus for us – and we’re also launching a hybrid gateway-style home entertainment system. Those are probably the two big ones this year.
After that we’re into building out our mobile network in the AWS spectrum. We’re working very diligently on that and finally increasing our Internet speeds. I would say those are the top things we’re working on this year.
Tony Werner, CTO and executive vice president at Comcast:
I’d say for this year’s top three, its getting Xfinity onto multiple connected devices, not just iPads. Also, continuing to expand our footprint for upstream channel bonding. Third is our overall transitioning of more services onto our IP plant – which spans about a dozen other things.
Frank Miller, CTO of BendBroadband:
We continue to focus on reliability and performance for our business and residential customers with automation, monitoring, continuity, and proactive maintenance and capacity engineering. In residential, we are setting the platform investment for gradual movement to a rich, all-IP platform while bringing key Internet experiences to our customers in the interim.
We see key opportunities in business services in the middle- and last-mile connectivity; telephony from MTA; SIP trunks and hosted PBX; and, finally, collocated, managed, hosted and infrastructure/application as a service. We have opened a state-of-the-art Green Data Center, known as “The Vault,” as our foundation on our regional transport and product leadership.
CED: Can you give specific examples of how your company is moving toward IP? Any thoughts on CMAP?
Nair: I believe that we will have a unified IP delivery of video in the year 2014 or so. Today we deliver video over IP and DVB-C to our customers using separate platforms. There is no immediate need for us to make the conversion right now due to our bandwidth availability, our next-generation gateway in the home and our ability to contain costs. However, when the CMAP architecture matures and it drives better cost efficiencies, then it may make sense to consider a move.
Our program has three steps: We are starting with an all-IP solution in the home gateway terminating DVB-C and streaming video to our IP clients in the home for second or third TV sets, as well as other IP end points. Second, this gateway has a DOCSIS 3.0 EMTA built in that will enable IP streaming to it when we migrate DVB-C to IP multicast. Third, we will then also migrate the apps to be Web services-based and not as heavy of clients on the box.
Werner: Sort of from left to right, our CDN is based on IP as a way to more efficiently handle how we move video around our backbone, from source to destination. In the headend, it’s about adaptive coding in cases where we need to transmit to devices with MPEG-4 decoders, and figuring out how to make our CMTS gear capable of converged service delivery.
That brings in the CMAP work. In the home, our cable modem strategy will involve CPE with increasing numbers of tuners to coincide with our wideband channel bonding efforts. And, in the back office and everywhere else, we’re transitioning “old to new” with Web service interfaces and “cloud” services, for lack of a better term.
Miller: We are reviewing upgrades across all components of the ecosystem, from deployment of a hybrid IP/digital television gateway to scaling for transport capacity and spectrum in our plant. A key first step was our move to all-digital in 2008. In terms of tactical progress, we have introduced a whole-home solution that provides a bridge to all-IP.
Steiger: We’ve been progressively migrating to IP for probably 10 years. The only remaining piece that isn’t IP now is the MPEG QAM streams in the digital cable service. Of course, we still have analog, but that’s the last remaining piece. Even with that, we’ll be evolving over that in the next couple of years onto a pure IPTV type of platform.
We’ll be testing CMAP in the fall and will probably standardize something in the spring of 2012. We do IP video on broadband today, like a lot of MSOs, and we’re working very hard on our fiber-based networks, which will be purely IP delivery. We’re doing RFoG in the interim, but IP is the strategy we’re going to have in the long term.
We’re going to have to wait and see a little bit on whether [CMAP] will be integrated or modular. Right now it’s looking more integrated because that fits into our architecture well, but certainly that could change.
CED: How far off are cloud-based video navigation and delivery for your company? What are the steps to getting there if that’s on your roadmap?
Miller: We see distinct benefits in the creation of a cloud-based video ecosystem to allow for innovation, personalization and a single experience no matter what the device. This is a complex path to navigate, especially for smaller operators. We are reviewing key partnerships with peers, CableLabs and innovative companies to help bring this solution to reality.
Werner: You could argue that our Xfinity iPad app is a form of cloud-based navigation, but that’s really just a starting point. Certainly all of us are going to need ways to maintain a consistent look and feel of our brands as we move our services into connected devices that consumers bring into the home.
Whether you hang the TV on the wall to watch video or look at it on a tablet, or on an even smaller screen, the Comcast Xfinity navigational environment needs to work consistently. That kind of activity works best when it’s up in the network – the device essentially reports its state in terms of resolution and bandwidth availability. Then we send down the right navigational environment for it so that it looks and works as good as possible.
Nair: This is definitely on our roadmap. There are a number of approaches here, and we are considering all of them. You can do it today, with current MPEG-TS, using technologies from companies like ActiveVideo, or you can use other emerging standards solutions over IP, and we are exploring all of them.
Steiger: There are certainly a lot of advantages in the cloud. We use a cloud type of architecture for our current VOD platform. We have a very thin client, and the navigation is actually provided at our headend facilities.
We certainly will be applying that to everything we do with our next-generation boxes, both our tru2way and our hybrid gateways. It’s something that we believe has a lot of advantages, including the ability to develop quickly, rollout changes quicker to the audience and much less testing than we’ve had to do historically. So it's something that is definitely on the roadmap for us.
CED: Is home networking evolving to one super gateway in the home, and if so, what is the timeframe? Do you think Internet-connected TVs will resonate with subs?
Miller: In the end, it is integration with simple UIs and the portability of content across fixed and mobile devices. We see this as a mix, as not every consumer television will come with IP and a client ecosystem capable of running the application environment like HTML5, Flash and others. The deployment of a hybrid DTV/IP-based gateway and players ready for the IP video ecosystem allow for setting the foundation for innovation. A key value for cable is the continued aggregation of content experience, this time for the Internet, linear and, finally, on-demand experience.
Steiger: The super gateway is a nice idea, but in practice, it isn’t something we can implement everywhere, and we shouldn’t implement everywhere. We’re going to be looking at different kinds of access mediums, both cable and fiber. We’re going to be looking at all kinds of different customers that have different needs.
While it would be nice to say we’re headed toward a super gateway, I would say that’s a few years down the road. Even our hybrid gateway box we’re working on now is intentionally not used for Internet access through its own mechanism. The Internet is provided through our existing Internet, whether they are fiber- or coaxbased DOCSIS modems. We do that for a reason so we can unlink the Internet services from the video services if need be.
So I think we’re a ways away from having a box that can do it all. Certainly it’s a nice idea; once the cost model is right for it, we’ll probably look at it more seriously.
Werner: Although gateway is a great word, and they appear to fix everything, it’s harder than it looks. We’re still not resolved on a single gateway strategy. It’s because there are so many variations of gateway devices in terms of what shows up on the bill of materials.
Whether consumers will “connect” their “Internet-connected” TVs is more of a function of available content than technology. As a company, Comcast is working hard to give customers both the experience and content they want on whatever device they choose to view it on.
Nair: I don’t know about super gateways, but there definitely needs to be a device in the home that can terminate your broadband connectivity, allow for digital rights management, host a browser or some form of application front end, and manage a number of LAN PHYs. This is not super complicated and can be built for a two-digit price point. We have something close in the form of an IP client box that we are developing with Samsung.
Internet-connected anything will resonate with customers. However, the successful ones will need an interface that allows the user to access what they want with more ease and relevance.
CED: Cable has a new set of partners when it comes to delivering Internet video to different devices in a home. What is being done to find common platforms and interfaces?
Steiger: Like most service providers, we’re struggling with that. We believe strongly in things like the entertainment ID registry, where we can work together with partners to stay coordinated and save our own development time and work on things with the MSO community instead of trying to do everything on our own. But we’re a ways away from that.
Certainly there’s no clear winner or de facto standard in a lot of areas. For instance, you’ve got Adobe with Flash, which we thought at one time would be a universal standard, but certainly it’s not turning out that way. So there’s a lot of work that needs to be done. We certainly believe that there’s incredible value in working with partners, other MSOs and third parties, too, to create that synergy, but it’s going to be a few years in the making.
Nair: We definitely have new partners, but our traditional vendors have also stepped up their game to adapt to our needs. CableLabs is doing an excellent job of merging all of our needs. However, there are many standard components out there, and it is no longer necessary for all of the cable operators to be deploying the same solution, yet we will all be standards-compliant.
CED: We hear talk of using more Internet-based applications and software by cable operators. Can you give some examples of being more nimble in these areas?
Werner: In a Web environment, we can develop and put out complete releases much faster. A good example of this is the Xfinity iPad app, which today is already on its fourth version in terms of features added.
Miller: As a bridge technology, applications are seeing adoption in the cable ecosystems that play to the foundation work for cloud video infrastructure investments.
This allows for the ability to bring the value proposition of cable content to subscribers in 2011, while the ecosystem is being developed for current and future fixed screens (TVs).
Steiger: Well, the Internet does bring us, with its ecosystem, a lot of benefits, and one of them is the ability to develop things quickly because a lot of the de facto universal standards are already there. We would love to leverage that. We’re probably not going to be launching any new products that don’t have some kind of a Web interface for control or find, or watch, for example.
Most of our content is going to be available to be managed through Internet appbased applications, so there’s lots going on there. If there’s one thing I would never bet against, it’s the Internet and all of the applications it brings, which is something we want to endorse fully. I’m an Internet engineer, so I’m always pushing the guys toward this type of environment. So I’m expecting we’ll see a lot more of it in the future.
Nair: All the cable operators are working on a similar objective, which is to provide our content to users, regardless of device. It used to be that a TV was the device, but no longer as smartphones, tablets, netbooks, etcetera, proliferate. Our broadband capacity and speeds have enabled these devices to be as good a video consumption device as your TV.
We are building a Flash-based UI that is very rich and would have taken a lot longer using our traditional approach. We are also building HTML-based apps for our gateway, plus other IP devices. Our developments are now measured in months, not years.
CED: So is this the year EBIF really breaks out, and what is the status of tru2way?
Werner: In terms of adding new features to the fielded base of legacy boxes, yes. We support 19 million set-tops in terms of EBIF right now, and we’re enthusiastic about what’s coming out from Canoe this year in terms of RFI and other interactive and national ad products. Tru2way is one of those terms – depends on what shade of it you’re talking about.
The majority of our network is tru2wayready, and we’re continuing to develop tru-2way for our devices. I think this year will be a big year for tru2way deployment.
Steiger: We’re not doing anything with EBIF. It’s not currently part of our plan, and part of that is due to the regulatory environment we have with advertising here in Canada that is somewhat different from the U.S. Advertising is one of the key drivers of EBIF, but we’re only doing that kind of work in our next-generation set-tops, our tru2way boxes.
We think that tru2way is an incredibly efficient bridge while we spool up IPTVtype technology over the next few years. It brings us a next-generation set-top box with a very easy environment to develop in. The VOD client that we spent years building in digital cable was built in the tru2way environment in less than three months. So there’s a lot of value in tru2way for us right now on the set-top side of the business.
We are going to be pushing ahead quite aggressively with our home gateway, but there will be lots of applications where a next-generation set-top box will be required for the customer, and tru2way will really fill that niche for us.
Nair: Given that we are an international cable operator, EBIF and tru2way are not really on our roadmap. We are building our interactive devices with a browser and a Flash run time. All applications will run on these two front ends, so it will be a rich experience. We are using CDI and Media Highway from NDS to abstract the chip drivers from the front ends.
CED: 3-D TV – a game-changer or a buzzword?
Werner: I think we got the answer to that question last Christmas. It’s hard to know if it will become mainstream or more like picture-in-picture was, a feature positioned as a major selling point, which some people used, but most didn’t see it as a reason to buy a new television.
3-D is a feature that will become standard in most HDTVs. Some people will use it. Some will wait until there’s an eyewear-free version.
Steiger: We’re supporting a fair amount of 3-D TV content in our VOD system already. Occasionally we have it in broadcast for special events. We see 3-D TV as being important as a niche application. It’s a niche application, but to that niche it is important.
There’s a subset of customers that find great value in it, and we’ll be supplying it for them. For the majority, it’s probably something that they’re going to pass on because it doesn’t add a lot to their entertainment or viewing experience. Primarily that’s because the technology for 3-D isn’t really there yet. It’s much better than it was in the 1950s, but the resolution isn’t there – having to wear glasses is a major problem. The kind of environment you need to effectively enjoy 3-D is too limiting.
We think it’s going to be a while in developing, but for that subset of our customers who like it, they like it a lot.
Nair: Neither. It will take off once there is well-produced content, and lots of it.
CED: Each of you faces a different competitive environment to some degree. How important is a mobile or wireless service in your respective markets?
Nair: We certainly feel that wireless access for data is important and have approached it on a situational basis. In Chile, we are building out our network with our own core plus radio towers, using spectrum we acquired.
In Australia, we sold our spectrum to a government entity to have them build out the network, which we will then use from a wholesale perspective.
In Europe, we are building out our core network with call control, HLR and rating engines, but we’re renting an MNO radio network. So we are active in this space but have different solutions based on local conditions.
Miller: We consider wireless to be a foundational investment, both a Wi-Fi perspective and 4G futures. With 4G, it’s an all-IP ecosystem with key integration for convergence of the cable experience.
BendBroadband has spectrum in both AWS and 700 MHz. We launched a fixed wireless over AWS in 2009 and are reviewing different strategies toward movement into a mobile play. The majority of customers expect a blended data/cellular experience.
Steiger: We are very active in the wireless area. We’ve acquired AWS spectrum in the majority of western Canada, which is good coverage with our cable operations, so there’s a natural synergy there.
More important than in the past is Wi-Fi to supplement that licensed spectrum. Just because of the volume of data that we perceive is going to happen, we’re certainly looking very strongly at supplementing with Wi-Fi using Cablevision’s model, which seems to have been very successful for them, with parts of the licensed spectrum model to build something we’re going to call the mobile Internet.
CED: Any plans for upstream channel bonding this year?
Steiger: Not yet. We are actively working on taking our services well beyond 100 Mbps on DOCSIS. On the horizon right away for us is 250, and beyond that probably will be 500 Mbps on DOCSIS. After that, we’ll probably be heading more toward a fiber-based infrastructure.
At 100 Mbps, there’s not a real driver for us to do upstream channel bonding. Once we go beyond that to 250 and 500, we’ll have to look at it seriously to create that upstream bandwidth. We’re going to be targeting primarily commercial customers with those kinds of upstream speeds.
Werner: We are in trial in several systems.
Nair: We are starting the process and have deployed it in Hungary and have plans for elsewhere, but I am more focused on the downstream than the upstream.
Miller: We will complete an upgrade to QAM64, and opportunities for channel bonding, specifically for business services, will be in 2011.
CED: What’s on the commercial services roadmap this year?
Nair: Cloud-based services for our customers, remote apps from Microsoft and others, virtual storage and application management. However, it will still be secondary to connectivity in the B2B space.
Miller: The ability to deliver a rich suite of transport, telephony and hosted platforms from the smallest business to the largest.
Steiger: We’re actually thinking very carefully in this area of taking commercial services from somewhat of a fringe product for Shaw Cable into a core product for Shaw Cable. I think one of our key focuses is on the Ethernet-everywhere style of networking, where we can provide private networks to customers, essentially giving them a data pipe linking them with their cloud service providers.
Services as simple as hosted PBX out to folks that will host their whole IT system for themselves. Giving them a simple way to do that with private Ethernet networks will probably be the core of our efforts.
CED: Let’s go around the horn for your final thoughts on any projects that I didn’t think to ask you about, or other areas of importance.
Steiger: Well, certainly one of the lessons learned is that there is tremendous value in pursuing global standards for all of the North American operators. Sometimes we tend to look too much at what we do within our own shores, and we need to look more around the world and see what is going on there. We need to be a bigger part of what is going on globally.
The other thing that I think we’ve all learned is that there are two things to never bet against. One is IP and the other is Ethernet. Those will be around for a long time and will be the center of the things we do for many years.
Miller: I think we have to foster a win-win with the content providers to both ensure innovation and develop a sustainable business model for both parties – as partners – as we move forward into this decade.
Nair: Our networks are becoming more complex and more advanced. Network reliability and operational excellence cannot be ignored. This is a high focus for me. What are the relevant metrics, how do we design our networks and how do we allocate capital are the non-sexy parts of our jobs that cannot be forgotten.
Also, our development cycle is longer than some of our competitors, therefore we either fix that or we need to be a lot better in spotting trends.
Werner: Only that we are much more than just cable engineers nowadays. Today some of the world’s greatest technologists, engineers and minds from across industries are working in the cable industry. It’s an amazing group of minds, and I’m thrilled to be a part of it.