CED’s 2010 CTO Roundtable Discussion
Our CTOs’ plates aren’t full – they’re overflowing.
Communications service providers are all headed toward a future in which subscribers are given greater access to more content and richer applications, accessible from wherever they happen to be and in possession of just about any device.
But there are a lot of customers out there, representing an enormous array of service needs to be fulfilled. There’s a boggling amount of content that will never stop growing, a vast amount of ground that networks need to cover and a proliferating number of devices being attached to the network. In short, while CSPs may all be heading in roughly the same direction, there are many different paths for getting to the general vicinity of anything, anytime, anywhere.
CED: What are your technological priorities for the next 12 months?
Scott Hatfield, Cox: Number one is certainly our wireless project. We’re launching both MVNO and built network markets, and it’s exciting to add a fourth product to the bundle. Also, we have a lot of work in video – we’re in the process of launching our Cox Design Guide, doing that on a tru2way platform and including multi-room DVR functionality.
Pragash Pillai, Bresnan: Number one is we’re about to complete our 10 gigabit core ring, with MPLS at the end of [April]. The second is our DOCSIS infrastructure. We began an upgrade last year. Our target is to have 70 percent of our subscribers on D3 by year-end. The third is enhancing our OSS platform, initially for provisioning and mediation of high-speed data.
Michael Hawkey, EchoStar: 3-D is very high on our roadmap. TV Everywhere remains superhigh – place-shifting. The third would probably be the multi-room, multi-device support – the ability to watch in one room, record in another, start a recording in one room, finish in another – move HD around the house.
Marwan Fawaz, Charter: We continue to invest in SDV and DOCSIS 3.0 to expand network bandwidth efficiencies and capacity.
A significant part of our technology and engineering focus is on commercial business solutions, from small- to medium-size companies to large carrier deployments.
We are developing technical solutions for the commercial space in the areas of scalable network architecture, feature-based applications, hospitality video solutions, L2VPN over DOCSIS, and back office and billing enablement.
Other major areas for technology focus include video UI and product enhancements (including launching tru2way) and EBIF.
Brian Whitton, Verizon: The addition of 3-D programming to our service offering is an exciting development we’ve been anticipating for a couple of years.
Most of the early 3-D content is likely to be half-resolution 3-D, requiring the same bandwidth as a single HD channel. As more full-resolution 3-D programming becomes available, bandwidth requirements will increase since full-resolution 3-D requires double the bandwidth – two HD channels’ worth of bandwidth.
Hawkey: And that’s the sad thing – consumers will buy a 3-D TV and they’ll get a Blu-ray player or a set-top, and a year and a half from now there’ll be a new 3-D out that’s even better. Consumers are going to have a problem with this – that the 3-D you buy today isn’t the end game.
CED: What are CSPs going to have to do to support 3-D?
Hawkey: It’s just more bits, whether it’s coming down a satellite, cable or IP connection. There’s nothing special or unique you need in the set-top box for the picture. It’s the TV that’s processing the 3-D.
What I do have to do technically in a box is that I’m putting up 2-D graphics, and they look really wrong. Those graphics need to be changed to be 3-D.
You can take the TV out of 3-D mode when you go back to the main menu, as long as there’s no video. If there’s quarter-screen video, the video’s not in 3-D and the graphics look fine, or it’s still in 3-D, or you need to do all of the graphics in 3-D because the video’s in 3-D, because you can’t mix and match the two different systems.
The big technological issue with 3-D right now is the communication path between the set-top box or Blu-ray player and the TV. HD has been HDMI, and that’s pretty well standardized. But getting the HDMI 1.4 release and beyond to support the connection between the two devices, and getting the community to agree on how to do it the same way, is the challenge.
Fawaz: We believe adoption of 3-D initially to be slow as technology standards continue to be developed. However, we are well positioned from a technology standpoint to offer 3-D content both linear and on-demand, should we choose to do so. There is no specific network build-out or expansion needed to offer 3-D if, as we anticipate, 3-D is mostly offered through the SDV platform for linear channels.
Hawkey: We will have to make tradeoffs. Do we put up an HD movie or a 1080p movie? People aren’t going to be expecting everything in 3-D for a long time.
Personally, I don’t see the advantage of talking heads on CNBC talking about stocks in 3-D. The kids’ cartoon, it doesn’t need to be 3-D.
The bulk of Dish’s bandwidth is used on local broadcast, and when are local broadcasters going to broadcast everything in 3-D? This is speculation on my part, but by the time we get to everything 3-D, the broadcast technologies and compression ratios might get us to where it’s the same bandwidth we have today.
CED: The next big thing before 3-D was TV Everywhere. For some, it’s a counter-measure against over-the-top. For some, it’s a path for adding over-the-top. For some, it’s a synonym for place-shifting. Where are we with over-thetop and TV Everywhere?
Fawaz: Charter is working closely with our programming partners to continue the evolution of the TV Everywhere platform in our Charter.net portal.
Pillai: We are looking at the TV Everywhere approach. It makes sense to do it, we have the content relationship and we have the network to do it.
I’m intrigued about how we get content from the Web into the TV platform, either Ethernet directly into our set-top box to extend what we do or getting content to our VOD platform – for example, getting RSS feeds to our VOD platform. YouTube, or RSS, maybe, based on genre or user preference.
Whitton: Bringing the Internet together with our video broadcast television and video-on-demand applications on a network level, and at a device level, gives us lots of options for how to serve and provide access to over-the-top material to our customers.
We could create a broadband video widget or application where you select the source, select the content and material, and play it back to different devices. Or we could do something more elaborate, depending on how things go with content owners.
We already have a YouTube widget that brings over-the-top video to FiOS TV subscribers, and we are working on an open SDK that will enable third parties to develop additional widgets, showing there are ways to make over-the-top work.
Hawkey: There’s place-shifting, and that’s a great way to do it. We recognize over-the-top delivery is in the market, and it’s not going away. The technology is making that transparent to the customer. He wants to log in to his Time Warner Cable account and search for a piece of content, and he doesn’t care if it’s Slinging from his home set-top box, or from the HBO website, or from the Time Warner Cable server farm.
So how do we make that all work together – a great search engine, a great playback engine, an authenticated content engine – and add in what the subscriber’s already bought, and also give him a place that he doesn’t already have access to content?
I want to be able to offer him VOD and onetime purchases of things. I want him to be able to control his home DVR, because he might be doing this away from his house so he can watch later. TV Everywhere is about being able to watch my TV or control my TV.
Saying that is marketing, but making it work across multiple players and keeping everyone happy is a technical challenge. You’ve got Sling technology, streaming technologies, you’ve got content providers with different authentication modes, and you want to be able to search all across this content, and there is not one good source of metadata. There are several providers that sell metadata, and there isn’t one that has every single channel and every single website with video. Searching across all of that is a challenge.
Google has a great text search engine. They don’t even have a picture search engine yet, so imagine doing a video search. Anybody, not just Google. It’s a technical challenge.
Hatfield: We offer a very high-speed Internet service that lots of people can do lots of things with. We think that’s great.
We’re not unique with the any device, anytime, anywhere mantra. We’re committed to mobility. We’ve already taken some steps to think about getting to anywhere. We’ve made progress with anytime – you’ve seen our MyPrimetime service. We think there’s a lot of potential there.
You’re seeing us trying to think about how to open up architectures and migrate to IP – TCP/IP – to get to different classes of device than we’ve traditionally been able to address beyond the set-top box.
Our philosophy is that we’ve got to enable the customer to easily get to multiple devices in multiple locations at multiple times. If we’re doing that, if we make that easy, if we can make that work better because we understand their identity, their services, their devices and the network. The focus has to be on us doing what we need to do and less on what the over-the-top guys might do.CED: One of the first implementations of the multiple devices in multiple locations approach was multi-room DVR. Where are we with multi-room delivery?
Hawkey: From a user standpoint, he’s got two or three boxes in his house, and what we’re providing is virtual tuners and virtual hard drives. The subscriber doesn’t care where it’s stored or what tuners they are, or whether there are two tuners in one box and one in the other – they want to know why they can record on one but not on the other.
What we have to do is provide virtual storage and virtual hard drives. If one drive is full and one is empty, let them record “Lost” on the one that’s empty. Cable tuners, satellite tuners – that’s the same thing. There’s only so many tuners. If you have two boxes with two tuners each, then he should be able to see any of those tuners. Our technology today – the majority of our DVRs, they drive a main TV and a second SDTV. Either the user has the full experience or they’re sharing the two tuners in their box. We’ve been doing that for several years. U-verse is making a big deal of that. Dish has used it. Bell in Canada has used it.
Pillai: We’re going to be deploying multiroom DVR soon.
Whitton: From the inception, we actually designed our home network to encourage and enable connectivity between ’Netenabled devices. For instance, Verizon was the first U.S. service provider to offer multi-room DVR functionality using MoCA technology.
CED: So multi-room DVR is an element of a larger home network. How does home networking fit into your company’s plans?
Fawaz: Consumers are more reliant than ever on their home network to connect devices to the Internet and manage personal content and information reliably and securely.
Charter has been successful in providing a home network service to our customers, and we know that the key to a successful home networking solution is primarily about the service and care we offer our customers.
Having the right tools for self-care and easy network setup/management is essential. From a service provider network perspective, it is imperative that home networking CPEs keep pace with speed and the number of connected devices.
Pillai: We deployed the wireless network into the home late last year, and it’s gone well so far. We haven’t had any major issues.
But with more devices behind our modem, that will probably generate more service calls related to that network. You have to have the support staff to handle the customer issues. Besides just having the people, you have to have the software and tools.
With Wi-Fi, we provide a self-install for your network – it even walks you through setting up your network. We can even do some troubleshooting remotely.
Whitton: We install in all FiOS homes our Broadband Home Router (BHR) that integrates MoCA, Wi-Fi and Ethernet interfaces to provide our customers with a variety of connectivity options.
We are currently in the process of upgrading our in-home devices to support the next-gen MoCA technology, which provides throughput up to 175 Mbps, and next year we will be introducing MoCA 2.0-enabled devices, which provide up to 1 Gbps.
In addition, we have upgraded our BHR to support Wi-Fi 802.11n, and we have included a GigE interface in all of our GPON ONTs.
Our BHR has the necessary horsepower and was designed to provide flexibility for supporting a wide variety of services and additional attached devices. For example, any USB device with content on it can be attached to the BHR and can become a content source.
And we’ve deployed a Wi-Fi-based option that lets you download photos from an Android-based wireless handset with a flick of the screen.
We anticipate an explosion in ’Net-enabled devices in the home, especially as we launch connected home services for functions like home security or energy management, which will ride the home network and connect to the world via the Web. Our home network design covers both the capture and the connection functions of the next wave of devices pretty effectively.
Hatfield: We already enable consumers to do a lot in the home because of the open IP network that we’re giving them. Cox high-speed Internet, if you’re an Internet customer, you have an unfiltered, unbiased IP stream you can bring into the home and do anything you want with it. That’s pretty powerful.
Hawkey: For pictures and audio, which are for the most part DRM-free, there’s an application on the 922 today, if there’s another DLNA-enabled device, I can see those pictures and music and play them back from my device, and I can also serve them up. If there’s an audio recording on my set-top, the PC can play it back.
We believe in the DLNA and support that. Video, when it becomes DRM’d, when you get digitalized management on it, it becomes a different play, and moving it around is a different challenge, which is why we prefer to Sling it.
We also have a challenge in the home network itself. It’s a big bandwidth hog. With Wi-Fi, I may be able to handle the data, but start adding video, and you can start running into QoS problems, which is why a lot of us like the MoCA standards. MoCA gives us a big fat pipe to move stuff around the house, with QoS we’re going to know about, versus collision-based type networks.
Networks are going to get complex for people. People aren’t going to understand. “Hey, I just bought the latest Linksys wireless router, I’m watching TV and streaming my Netflix, and my kid just got home and started doing his homework, downloading a whole mess of stuff, and my movie started breaking up.” Wi-Fi does not have QoS associated with it.
The newest version of Wi-Fi is supposed to be addressing that, and we’ll see how that goes.
We offer MoCA, we offer Wi-Fi, we offer HomePlug, and that’s solid. There is no one panacea of technology to solve the home network and allow us to share everything.
CED: There are two focal points to home networks: One is the PC, where the broadband connection comes in, while anything that has to do with video goes through the set-top. There’s been increasing discussion about a third device that would be a single, central point for home networks: the gateway.
Pillai: Gateways in the home is just a matter of time. Technologies like DLNA will play an important role. We’re starting to see consumer electronics playing with what we’re delivering. It’s coming down the pipe.
Hatfield: The definition of a gateway is still pretty fluid. No one knows if that requires me or prohibits me from doing something that I might need or want to do.
It’s clear to us that customers want help managing their technology and managing their networks, and we’re best served if we help them do that.
We don’t know the final answer yet, but you are seeing some re-thinking about how network assets get reused. If you can get it all to an IP environment down the road, will that let you simplify? Today, we’re putting a lot in the home – we’re putting in cable modems, E-MTAs, STBs and femtocells. Can you converge all of that and go with network storage and save money for the consumer?
Hawkey: Does it have storage or not? Does it have security in it? How many tuners does it have; does it need tuners?
If it truly is the controlling point in the house, do you need two of them? I want redundancy. If I’m watching the Blackhawks and the Maple Leafs playing for the Stanley Cup – damn, I don’t want the gateway to go dead.
On the other hand, having a gateway that’s managed by the same guy I’m paying a monthly fee to to guarantee it’s always up, and maybe doing all of the configuration for my home network, that would be cool. That might be the answer to virtual tuners and virtual storage; the gateway could be the point that makes sure you can see the same thing on every device.
The cable guys are pushing harder and faster than the satellite guys about a gateway and what its contents are. We’re trying to help them resolve what a gateway is. Will it replace every single set-top box? If yes, that just means all of the money moved from the set-top to the gateway, but there’ll still need to be a bit of software on the TV.
Whitton: There is a clear shift in the industry from the currently distributed set-top box architecture to a centralized IP-based client/server architecture. Technologies like DLNA and RVU are rapidly enabling this new architecture.
We believe the client/server architecture will be beneficial to both our consumers and Verizon. The new architecture will allow our subscribers to use a variety of widely available IP devices for in-home content consumption and sharing, therefore enabling our vision of a truly connected home, while at the same time allowing us to reduce the number and variety of costly proprietary STBs.
Eventually, we envision client STB functionality being integrated into different CE devices, or our customers can acquire the client STBs at different retail outlets, while Verizon supplies only the gateway/server unit.
Similar to our current BHR design, these Verizon managed gateways will continue to serve as the hub of home networks and the enablers of services. For instance, we could push home security software to the device and run the service from there. There’s no question that powerful, dynamic gateways are the future of home services, and we’re glad we went high-end with ours.
Hatfield: This is a cross-industry exercise we’re about to embark on, and even when you have effective industry groups that will help you set standards, that begs the question of how will you do that across industries that perhaps have been competitive in the past. That doesn’t feel like a simple, quick or easy process.
Look at gateways as an example of the promulgation of cross-industry standards. If you’re a programmer and you’re trying to write an EBIF application that runs across both companies’ networks and you want to do that once, we’re going to have to think about ways of having standards at the company level, standards at the cable level, and how would we do standards across industries? It’s an emerging need that we probably don’t have a great answer to.
CED: Where are we with EBIF and tru-2way?
Hatfield: We’re very excited about tru-2way. It’s created a layer we can develop services to. Lots of people have given consideration to what it means for retail, but what I really love is that it’s created a development platform for us.
We can think about device independence, vendor independence, and we’ve made use of that. Our guide and multiroom capabilities are built on tru2way. We’ve proved we can write on top of it.
We’ve proved it’s a good platform. That’s the point I’d like to emphasize – it has value to us in terms of speed, quality and independence that’s going to be very valuable as we move forward.
Fawaz: In 2010, we are focused on getting EBIF scaled throughout our footprint to enable advanced advertising and interactive video applications.
We are on track supporting tru2way capabilities for set-tops later in 2010. We are also evaluating gateway solutions with IP video capabilities. The approach for migrating to an IP gateway solution is contingent not only on standards work, but also regulatory requirements that the National Broadband Plan task force mentioned in its March report.
CED: Success is increasingly contingent on having a bundle of services. How is your bundle portfolio progressing?
Hatfield: We’re very excited about the fourth product. We think mobility is strategically very important. That’s true of all of our products. Video needs to find a way to become mobile, data needs a mobile component, and if you think about voice and wireless substitution, that’s a place that requires mobility.
It’s not really a fourth product; it’s adding mobility to what we already have. It’s not just one more widget to go out and sell; it’s a strategic evolution of communications services. We’re very committed to having it be an integrated service offering. It’s not a sidecar standalone RGU.
Pillai: We bought [700 MHz spectrum]; strategically, we need to have it. We haven’t done anything with it. We continue to have discussions with vendors. If you look at 700 MHz technology, it’s not readily available for deployment. Those deploying now plan to migrate to 700 MHz later. We’re watching to see as it becomes viable. Maybe 2011, 2012. My understanding is that it’s lacking CPE.
Hawkey: The quad play is still a big deal. Satellite for the most part is a one-way distribution of linear content (although there are two-way systems). The way to address it is to go to a hybrid satellite/IP system, but I still end up having to pay two providers – the satellite provider and the data pipe provider. Bundling and how that will get together, I expect my grandkids will pay someone $100 a month for a gazigabit that will go across the house and the thing they carry in their pocket and into their head, and it will all come from one pipe.
CED: The promise of an expanded bundle with converged services is the potential for cross-platform features and services. That implies a back office capable of enabling and supporting such features. OSS/BSS seems to be a common priority.
Pillai: There might be more intricacy and complication in the OSS than in the hardware piece. We continue to invest money in software because it’s the way you package your services, it’s how you reduce the silos within products, how you integrate packages – you get visibility into what products customers subscribe to.
What we’re doing is taking what we have on our DOCSIS provisioning and mediation platform and taking it to the next step. It brings in IMS and all of that stuff. Advertising, everything – service integration, convergence. This is the first step that’s required.
Right now we have one vendor for both platforms. We’re going to two vendors, each providing best of breed in one approach.
We’re taking our high-speed data back office platform and migrating it completely to the new platform. Right now we’ve got silos – voice, video and data. This will let us converge them all onto one platform. We’re working with the vendors, and our goal is to have this migration done soon.
Hatfield: Cox has always prided itself at having a pretty strong back office. We’ve had a consistent set of applications for three products. You’re trying to provision cleanly, share identity, converge marketing messages, converge billing, converge care.
There’s no way to do that and add mobility without having a well-architected converged back office. It’s an enormous amount of work, but we think Cox – and the cable industry – is well positioned to do that.
Nobody takes that lightly. We talk to guys all the time. There are vendors that can drop into some country and deliver a standalone prepaid wireless package and bing, bang, boom, they’re done. That’s hard, but it’s not that hard. We’re trying to do something different; we want to converge the services and get to a place where it’s four products with different network delivery systems all hanging together and being architecturally clean, and that’s a much harder challenge.
We looked at a couple of different platforms, we did an RFP process, we had a couple of good options, and in the end we selected the Oracle suite of products. We’ve stood up essentially a parallel environment for wireless using pretty much a complete set of Oracle products from the care/ordering interface through orchestration through billing.
We believed wireless was the right place to try a different set of applications. It’s going very well. It’s a very complex thing, but we’re seeing lots of benefit. We’re seeing the value of commercial product and things like configurable product catalogs for speed and frequency of product evolution.
CED: What’s the status with interactive advertising?
Pillai: We’re doing a trial with dynamic VOD ads with Arris and BlackArrow. The technology works, we’re just proving out the business model.
At the end of the day, advertising revenue has to drive it. We haven’t seen a fully baked business model around interactive advertising where we can say, “OK, it makes sense for us to do it.” So we’re doing the trial to make the business model.
Hatfield: We’ve been focused on the guide as our first priority. We’re following the other guys. We’re aligned, but we’re following right on their trail.
CED: What about interactive applications?
Hawkey: The answer is: It’s got to be Webbased – that’s my personal opinion. It’s just the logical choice. The world knows how to write applications for the Web. Our 922 box is a Java virtual machine engine. Java and HTML are here to stay.
Internet TVs are here. They’re great. But every Internet TV I’ve used, I have two different remotes giving me two different experiences. The TV wants to control eyeballs, and the STB wants to control eyeballs. They have different interfaces, and there isn’t a good clean answer.
They all have their own remotes, they all have their own look and feel, and they don’t like to share data among each other. I look up my menu on my Internet-enabled TV, and he’s not paying for the channel information, so I have to go somewhere else to find out what’s on HBO, so I have to go to my set-top box or the guide from my gateway … it isn’t clean.
Look at OpenCable CableCard TVs out there. The OCAP app takes over the TV, and the CE manufacturer doesn’t get to put any of his neat, whizzy stuff up there, and hell, they integrated it. You get either the TV experience or the cable experience, even though it’s all integrated in one box; they just can’t agree. They will. Technically, there’s not a problem. Everyone just wants to own the eyeballs and the real estate.
Whitton: If our vision of the TV as the communal device in the home is to be fully realized, interactivity is fundamental. And, once again, having the Internet and the entertainment content meet at the device is critical to future developments.
Clicking a TV remote’s cursor on Fonzie’s jacket to order a vinyl reproduction from Aéropostale should be easy enough to do, or if not that, pausing the TV and opening a “Buy This” page related to what’s onscreen might be an option. Technology and enterprise tend to meet and make amazing things happen, so, it’s coming.
CED: What about some of your other priorities for the next year?
Pillai: We started upgrading to DOCSIS 3.0 last year when we started upgrading our major market. Instead of adding new capacity to our 2.0 platform, we started updating to the 3.0 platform and re-deploying the 2.0 platform to our smaller market footprint. Our target is by the end of the year, 70 percent of our HSD customers will be on the 3.0 platform. We’ll be able to deliver 3.0 if that’s required. It will depend on the market – we’re already the market leader for speed in the markets we serve. We’ll continue to improve that for our customers. We can get to 3.0 speeds pretty quickly if that’s what the market demands.
The platform for D3 has to be a carrier-class platform with full redundancy. You don’t have to have manual intervention if there’s a failure. Even in the 2.0 mode, we’ve deployed multiple downstream, four downstream and two downstream. So upgrading to 3.0 won’t be that much work.
It’s a must for voice. With video, people are not going to live with tiling or anything. We tend to take video for granted, but with our experience, it’s helping us through with video over IP. You’ll see that with things like the iPad. Building this infrastructure, it’s positioning us for whatever comes down in the future.
Hawkey: If it’s a Sling-loaded box, something we just launched with the 922, I can Sling within the house or outside the house. I’ll do you one better than whole-home, I’ll give you whole-world TV. I can Sling to the laptop, to another TV, to a mobile phone, to an iPad, and that gives me whole-home.
We can Sling to our set-top or anyone else’s set-top. We’re talking with other box manufacturers and TV manufacturers about Sling-loading their products so we can stream content between those other devices.
Whitton: The next five years are when we’ll wow customers with everything from multi-screen TV viewing, where what you are watching follows you from device to device, to Ultra HD video, all the way through next-gen PON systems and mashups of previously incompatible content and services on previously incompatible devices. That’s where the magic comes next, and it’s why companies that have access to land-based and wireless networks and to home networks and smart systems will have the advantage in attracting and retaining customers.