Gameplanning for Broadband
CED recently conducted a conference call with some of the top technical minds in the industry to check the pulse of MSO priorities and to discuss the status of technical hotspots such as content mobility, IMS and mobile voice services, downloadable conditional access, bandwidth management and the network-based digital video recorder.
Joining the call were Chris Bowick, SVP of engineering and CTO of Cox Communications; Wayne Davis, EVP of engineering and CTO of Charter Communications; David Fellows, EVP and chief technology officer of Comcast Corp.; and Mike LaJoie, EVP and CTO of Time Warner Cable. An edited transcript follows.
CED: To get things started, please explain what your top priorities are for the balance of 2006.
LaJoie: For Time Warner, I would say we're looking to [rapidly increase] our phone [customers] this year. We've had a lot of success with the product, so we're really focused on continuing that rollout. We're also very focused on DCAS, and working very closely with Cox and Comcast on that. We're rolling out our Digital Navigator this year [for legacy set-tops] as well as the OCAP version.
Bowick: I'll break it up into voice first, then look at HSI (high-speed Internet) and video. With voice, it's really voice ubiquity, taking voice to all of our markets. We're taking VoIP into new markets but also introducing VoIP into our circuit-switched markets, and, along with that, some focus around what we call "phone tools," which are really the integration of features and functionality with our CHSI (Cox High Speed Internet) platform.
On the CHSI front, we're focused on PacketCable Multimedia [and] rolling that out in a few markets and looking at speed previews and gaming and turbo button capability there, and also doing some e-mail enhancements with a new Web-enabled user interface.
From a video standpoint, we're certainly focused on on-demand ubiquity; we're not quite there yet, but should be very shortly. [We're also planning on] digital simulcast ubiquity and doing a lot of work on OCAP and preparing for trials later in the year and rollouts next year. We're also going to do some work with switched broadcast later in the year.
Davis: A big focus for us is rolling out residential telephony. We're going after six million to eight million homes passed through the rest of this year, which, from a zero ramp from the middle of last year, to eight million homes passed is a huge effort. Residential telephony—there's a lot of heavy lifting there [with] getting the HFC plant ready. In order to be successful, we think having a singular focus, or maybe one or two top priorities, is most important in order to do that right. With voice, you have one opportunity to do it right. We're going to do it right first.
With OCAP, that's a big focus for us right now. We're also focusing on the technical operations, which is improving the customer experience so that all of this product we put out there works, and works well, [under] a hardened architecture. The big focus we have on the company is managing our networks well through putting a mentality on the folks who touch our plant—whether it's our own technicians or vendors or contractors—[and ensuring they] have a carrier-class mentality. [That] is very important because as we rollout these services, it's not only the service itself but it's how well the service works and what the customer experience is.
CED: Dave, can you give us a general description of the RNG project and what the plan is for that this year?
Fellows: RNG stands for Residential Network Gateway. It's a set of set-top boxes that uses DSG (DOCSIS Set-top Gateway) for signaling, has an advanced video codec inside—MPEG-4 Part 10 or VC-1—is capable of running OCAP, so it has the microprocessor and horsepower and memory footprint to run OCAP, and comes in flavors from low-end boxes, through PVR (personal video recording) and up into PVR in the basement or whole-house PVR boxes. [It's] a family of boxes where the common things are DSG, OCAP, DCAS for security, and advanced video codecs.
CED: Will any of those be rolled out this year?
Fellows: Standing in the way of DCAS is an embedded security ban for the middle of next year. I think the first versions of these or one version of these boxes may be host boxes. They'll have multi-stream CableCARD slots. Certainly, this is the box DCAS will be introduced to. And, by the way, DCAS and OCAP go hand-in-hand. I don't know whether we'll get any [RNG boxes] out this year that have that. If it was just advanced video codecs and DSG, the answer would be yes, but the security side may not make sense, to introduce a new box for a couple of months and then have that box become illegal. That's something that will be played out in Washington [D.C.].
CED: Are others on the call doing anything similar—working on set-top box specs internally—or are you going to be taking the lead from the vendors?
LaJoie: We're definitely working on boxes with similar profiles. We're certainly looking at incorporating DSG into our future set-top designs. All future set-tops that we buy will be OCAP-capable. All the boxes that we're buying now are OCAP-capable. We'll incorporate DCAS, and we're also looking at advanced codecs for hi-def. I don't think we'll formalize it quite as much [via] a program as Dave has, but we'll have boxes that I think will have the same requirements. We buy a lot of Moto[rola] boxes, so I think we'll benefit from the work that they're doing.
Fellows: If you look at it from the other direction, from Cisco's or Motorola's perspective, they don't want to produce a box for Time Warner and then a different one for Comcast and a different one for Cox. They are going to try to herd us cats together. In the meantime, we all—all the MSOs—face the same challenges. Even if we think we're innovating in a specific area, we also learn and benefit from each other. Whatever any of us is doing tends to wind up benefiting all of us.
Davis: We all benefit from all of these things. Comcast and Time Warner are doing the heavy-lifting, and we all benefit from that and appreciate that. We're all dealing with the same issues, the same challenges and the same opportunities. To the extent that Comcast may spec a box and come out with these great features, when that becomes a product, it becomes available to the rest of us to be able to buy.
CED: At the consumer electronics level, there's a lot of activity with portable video and music devices. EchoStar and DirecTV have also moved ahead with portability plans. Why do you think cable has not yet capitalized on this trend and come out with its own portability products and services?
LaJoie: At Time Warner, we've looked at it. Initially, we have [a project] underway to spec a personal media device working with some undisclosed CE manufacturers. I think you have to recognize that, with the personal media devices that are out there, basically we do support them all, and so does everybody that offers high-speed data. If you have an iPod and you're a cable customer, you're downloading all of that stuff and enjoying the benefits of your high-speed data service. The question kind of is, are we going to develop interfaces that would support the connection of these devices to your cable set-top box? I think the answer is unclear right now. Certainly we can do that, but the idea of us coming out with a branded media player, like the PocketDISH from EchoStar, that's really not a technology question. It's not hard to do.
CED: Many of these digital cable boxes have a USB port, so, outside of the digital rights management element, is most of the groundwork there to give customers the ability to record stuff on their DVR and transfer it to a portable media center?
Bowick: From Cox's perspective [content mobility] is definitely on our roadmap. As we look through 2009 and 2010, we believe that portability and mobility of product and allowing our customers to have their content wherever and whenever they please will be very important.
CED: Staying on the mobility angle, what's the status of the Sprint joint venture? What are some of the key engineering elements that will be required to make this work?
Fellows: Remember that, in the near-term, the Sprint joint venture is a resell of Sprint service, and in the long-term, it's a unified, fixed mobile convergence play—mobile phones go into the home and services and attributes and such carry with that service in the home and outside of the home. What we're busy doing work on right now are the things that you would need to be able to offer someone else's service—changes in the billing system, interconnecting provisioning systems. One of the things of interest [is] that the MSO does tier-1 support—so getting the tools in front of our tier-1 staff, our customer account executives, so that when a customer calls in for service or with a complaint or problem or some question, they've got the tools in their hands to be able to answer that and offer wireless service.
In the slightly longer-term are the things that our bosses played up with all of you—programming your PVR from your cell phone, watching streaming video that you may have stored away. That's a longer-term development but there are teams working together there and that will lead us later to a conversation on IMS (IP Multimedia Subsystem).
LaJoie: Our challenges are more marketing and operational and that's where the focus is. But there is activity going on now to figure out what the engineering challenges will be to accomplish fixed mobile convergence. Those conversations are underway.
CED: As far as the shorter-term portion of this, the reselling component, is there a timeframe that you can talk about yet?
Bowick: I think that we've publicly said that we'd probably go into a couple of markets this year, but those are unspecified at this point and it will be later in the year. I'll just add that one thing that we didn't want to do is just go out with a 'me-too' wireless product right off the bat, so there has to be some, albeit minimal, integration of the product. For example, we'd like to have a common voice mail integration between wireline and wireless. Integration of our e-mail portal, for example, with the portal for our telephone and wireless [products]. We wanted to ensure that we've got free calling, if you will, between our subscribers' home wired phone and the wireless phone and vice versa. Those kinds of things require a heckuva lot of billing integration, as Dave pointed out earlier, but we didn't want to come out with a 'me-too' thing right off the bat. It has to have some integration.
CED: This brings us to you, Wayne. What can you tell us about Charter's mobile voice plans, since Charter's the only one on the call not officially part of the Sprint joint venture?
Davis: It really does go back to a comment I made earlier—a singular focus on being successful with residential telephony. It's my understanding that there is the option that we have to become part of the J.V. at some point, and the business executives look at all that and the timing associated with it. I'm sure in time we'll hear something about that. But from a technology perspective, there really isn't much to say in that area.
CED: This discussion dovetails with our next subject, IMS. It seems like we are getting a lot of information everyday from the vendor community on this subject and how they're incorporating it into their different products. From a high level, what do you see as some of the primary benefits of IMS?
Fellows: Let me start with a little background. IMS is based on SIP (session initiation protocol) as the communications mechanism between the client and the devices wanting service and the network and the devices that are providing the service. Within Comcast, we have three places that we're using SIP. One is for things like video instant messaging. These are for clients that talk SIP. We have a SIP infrastructure in production to accommodate that service. The second place we find SIP being useful is connecting PacketCable Class 5 offices together. Already, our PacketCable softswitches talk to voice mail platforms using SIP and it would make sense that they would talk to each other using SIP. The third place that SIP is useful is as we approach fixed mobile convergence. As you interconnect your service delivery platform with another company's, say Sprint's, IMS forms a very useful set of border gateway protocols where their service delivery system can talk to an MSO's.
When you see Comcast deploy a branded IMS as opposed to maybe a purely SIP infrastructure, I think it will be in response to fixed mobile convergence or the joint venture with Sprint.
LaJoie: The only thing I would expand on is that IMS really came out of the 3GPP effort, whose focus really was on fixed mobile convergence of wireline phone and wireless. But the promise of IMS is much broader than that. It can converge services across networks from any source. The promise of being able to converge services of high-speed data, voice and video—whether you're fixed or mobile—is what's really exciting and what gives us a big step up. IMS has got a ways to go. There's a lot of progress going on, but one concern I have about it is that it's complicated. But this ain't your daddy's cable service anymore.
Bowick: I think it's still evolving, so I think that's part of the problem we're facing here. If you're wondering what it gives us, one thing I think about when I think of IMS is rapid service enablement. This is going to be kind of a stretch here, but it's almost a middleware for the network that allows for you to deploy services rapidly. Keep in mind, too, that there are elements of IMS within PacketCable Multimedia, 1.0 and 1.5, and we're trying to align PacketCable 2.0 with IMS.
LaJoie: The complexity and nascent nature of IMS notwithstanding, we have a working trial up and running in our lab and we're probably launching a very small market trial sometime this year—probably more of a technical trial than a marketing trial—on a production network. This stuff works. The value of it is very apparent, and the power of being able to rapidly turn on and deliver different kinds of services like video and gaming and chat and voice and mobility...I think the promise is real. We're very excited about it.
Davis: We're staying very close to it from a very intellectual discussion at this point. As you deploy wireless, it becomes one of the things that removes all of the platform silos that you have. But another capability that it has is that it also incorporates presence. That is a feature that, as you move across platforms—knowing who you are and where you are and what you're doing—is important. From an operational perspective, that's a nice feature.
CED: Perhaps this is a loaded question, but one that provides a gauge of where you think IMS is today. On a scale of 1 to 10, with 1 being a pipe dream and 10 being ready for deployment, where is the technology today?
Fellows: Five or six, maybe headed to seven. The framework, I think, is proven. There are pieces within the framework that are deployable and working well. Can you believe everything that's on an IMS vendor's viewgraph? The answer is definitely not. But I think we've all gained enough confidence in it that this is a good road to go on.
LaJoie: The full promise of it is probably a couple of years away, but there are components of it that are deployable now, especially on how we start to interface to other networks. As you see the wireless carriers and the fixed phone providers more and more putting this signaling core in their networks, I think it's pretty important for us to be able to peer with those networks and to be able to signal with them. I think this is something that's going to be deployed; there's not much doubt about that.
Fellows: Comcast has something called "Bedrock," which is sort of our first stab at a service delivery middleware. We have something called the GDS, the global data store, which is a lot like the HSS (Home Subscriber Server—an element of the IMS architecture). We have identity and presence that are enabled in this with tools and boxes that carry into the IMS framework. Some of the pre-IMS pieces are there. Mike is taking it even further—there are IMS pieces that are deployable and are in his lab in trials. I've got them in my lab. I don't have a timeframe for a trial yet. Again, we're confident that this IMS framework will serve our needs.
CED: Moving to DCAS, there has been some scrutiny of the project in some recent filings with the FCC by computer companies and other consumer electronics interests. How has this impacted the timeline for the project?
Fellows: It may be dangerous to admit this, but I have not read any of the filings, so the answer is that it has not impacted the project at all. The goals of the project, no matter what the filings say, are to have a retail and, hence, by that definition, an open, security system. But by its very nature, it is the core of the business that we are in. The CE community will be given everything they need in order to manufacture devices. The chip people, including those for the secure micro, will be given everything they need to produce and compete in that space. And for the conditional access companies—they will get what they need to create downloadable versions of their product line. This will be as available at retail as it is in the leased boxes. Those are all the goals. I don't think anyone disagrees with the goals; they may want to know more about the system and know more about the interfaces, but we are charging ahead.
CED: Looking at the most recent filing, there was mention of a separate facility related to DCAS. What will its function be? The distribution and creation of the secure micros?
Fellows: There is a facility that inserts the secret keys into the secure micros and keeps track of what is known as the "trusted authority" that keeps track of the database of those keys. That way, as boxes wander from Charter territory to Cox to Time Warner to anywhere, our headend consults this trusted authority to get the information, in order to download, in a secure fashion, the CAS (conditional access system). Then the local conditional access system takes over—Mediacipher, PowerKEY, NDS, or what have you. So there's a facility and a national database associated with the DCAS project.
CED: Because this project is linked to an FCC extension, is there anything readers should be getting ready for just in case you don't get the FCC extension?
LaJoie: I really think the reason why we're doing DCAS is because it just makes sense. It's a better way to do it. While we have presented this methodology to the FCC, and it provides a direct benefit to us and the consumers and to [the] CE [industry], and it sure would be nice if they gave us an extension on separable security—that isn't really the reason why we're doing this. It's just a better mousetrap.
Davis: I agree. For many years we've tried to get control of conditional access to not only manage that better and drive down costs associated with that, but also, more importantly, to bring some level of changeability into that paradigm. If I have conditional access A in a market and I, for whatever reason, decide to make a change with that, that's a very expensive proposition. Very laborious and complex. Today, it's something that you don't do. In the future, I'd like to have the capability to do that.
Fellows: In terms of your readers, the current embedded security ban is July of 2007, and the DCAS project has been clear that by July 2007 we will not have a nationally deployed downloadable conditional access system. We have not said to the FCC what our asks are; that is not something that we've done. So there is a disconnect, and I think it is at least possible that multi-stream CableCARDs in leased devices, which is the current regulation of the land, is in our future.
CED: So, does that mean you'll go with that first?
Fellows: No, I didn't say that. I just said that's the current situation. What we're going to do about it is probably not even up to any of the engineers on this call.
Bowick: I think that's true, but as the engineers on the call, I think we all have to be prepared for whatever comes down. If the question is what do we do to get ready now, just in case, we've been working very hard with our vendors to make the multi-stream CableCARDs and set-tops available in a reasonable timeframe. We've been working very hard with our vendors on the OCAP front. So all of those things are what's going on in the background to prep just in case.
CED: Moving to digital simulcast and bandwidth management, what's the status of your rollout? When do you expect to have it rolled out across the board?
LaJoie: On simulcast, we're about half-way done. By sometime relatively early next year, we'll have simulcast up everywhere. We're also doing switched video in a few locations now, and it's going really well. We're seeing really excellent bandwidth savings there. We have plans to roll that out further this year and into several more sites.
Fellows: We ended last year at about 70 or 75 percent simulcast, and at the end of this year, we're just going to 80 percent simulcast. We're already taking shipment of many—hundreds of thousands—of set-top boxes that are all-digital—they don't have analog tuners in them. We've also deployed a service where if you just sign up for enhanced basic, you get a digital set-top box. So, in addition to the roughly 80 channels of video, you get access to a fair amount of free video-on-demand programming, so we're deploying video-on-demand even to basic subscribers in some markets.
Bowick: We have an internal goal of trying to be ubiquitous with digital simulcast in the first half of the year. I can tell you that it will be difficult to do in a few of our markets, but we're pushing very hard to make that happen. I certainly believe that by the end of the year we should be there in most of our markets.
CED: Chris, what are some of the gating factors that are dictating the difficulty in introducing simulcast in some of the Cox markets?
Bowick: Technologically, there aren't that many issues to deal with. I think those are being solved as we go. It's really a question of timing and when the bandwidth will be available to do simulcast across some of our markets, and the timing of being able to use switched video in those markets as a tool and add incremental channels.
Davis: We ended the year with 20 percent of our homes passed with digital simulcast. I don't know where we'll end up by the end of 2006, but we're going after it as much as we can manage with all of the other priorities that we've got going on. It's been a real boon for us in terms of our set-top boxes, because we're effectively "mining" our simulcast markets to only buy all-digital boxes. That allows us to move boxes, analog boxes and boxes with analog componentry from those markets into our other markets, and leverage those a bit. The savings has been pretty profound across our footprint because we've got enough of the large markets deployed already that it's feeding set-top boxes into the rest of the markets.
CED: One of the long-term goals of simulcast is to reclaim analog bandwidth. What are the plans in this area? Five channels a year? Ten channels a year? Is it really that straightforward?
Davis: At Charter, I'll call it just-in-time deployment of bandwidth. If you look at the size of the pipe, 750 MHz or 860 MHz, it's a huge pipe. To the extent that 60 or 70 percent of our bandwidth is used in analog, that's nothing but pure opportunity for us to either move those [customers] to digital or [upgrade] to 256 QAM or higher modulation. When the time comes, market by market, as we need the bandwidth, we'll go after it. We have some pressure in some markets where it's 750 MHz and we've got 20 or 30 [off-airs]. In the Los Angeles market, for example, we have off-airs that use up a lot of analog spectrum. As we take a look for the need for bandwidth and HD comes into the picture...we'll use every tool that we have in the toolbox in order to manage that. One of the questions that we'll get to is whether we're looking at upgrading to 1 GHz. That is a long way off. We're not even considering it because we've got so many options to mine the bandwidth that we already have.
CED: Maybe I'll throw that into the mix—bandwidth expansion. In your 750 MHz or 860 MHz systems, is that going to be enough or are you also looking at 1 GHz or more?
Fellows: At Comcast, we're not looking at 1 GHz. I guess if I were a little bit slower in the rebuilds I might have finished off the last systems at 1 GHz, because that is the amplifier that's out there now. But we're not at 1 GHz. At 860 MHz there's plenty of bandwidth. At 750 MHz, it is tight. We're going to need tools like switched broadcast in order to launch the number of channels, especially high definition channels that we're going to want to have on our systems. I think switched broadcast is in our future, and any kind of plan to go to 1 GHz is not.
LaJoie: From Time Warner's perspective, we're looking at it slightly differently. We believe that sooner or later, no matter what, you're going to run everything through a switching fabric, because it's a more efficient use of the spectrum. We really don't intend to upgrade [but] there might be one or two systems in isolated areas where we might go to 860 MHz or 1-Gig, but generally speaking, we are going to switch and we're going to move channels from analog to digital. It does require some effort to manage it. We think if you spend money upgrading, you're just delaying the problem, because if you're broadcasting—even after you upgrade—and you start throwing in a whole bunch of hi-def channels and a lot more bandwidth for high-speed data and whatever other new products that come along, sooner or later you're going to be in the same space. You're going to want to switch. We are pushing switching a lot more aggressively than either Cox or Comcast are, but we figure if we push hard on that stuff now we can manage and have a much better product earlier.
CED: Chris, where are you right now on switched broadcast and your bandwidth efficiency plans?
Bowick: We've mapped out our strategy for the next several years. As Wayne said, we've got a bunch of tools in our tool chest that will allow us to not have to reduce our analog channels by five or ten channels per year. We consider the AO, or additional outlet, to be a strategic competitive advantage for us, and we'd like to maintain analog channels at least in the five-year window to the tune of 70 or so, which is approximately where we have them today. But we, too, will be using tools, as well. Switched digital broadcast, I think absolutely it's a tool in our tool chest. We'll be using that throughout in order to make this plan work and be able to add upwards of 30 to 50 HD channels in the future and expand our CHSI speeds well beyond where we are today. We'll also be using node splits and advanced modulation techniques like 256 QAM and service group reductions and, of course, adding digital simulcast to the mix. I don't see us doing 1 GHz in the near future at all.
CED: What about systems that aren't built out to 750 MHz or 860 MHz, like those few that might still be 550 MHz? Have you considered doing a hard cut-over to digital and skipping the simulcast step?
Fellows: Between now and 2009 I think most of us will keep at least the must-carry stations in analog. Even on an 80-channel system that you want to expand and go "all-digital," you'll still keep something like 20 of those channels in analog and convert 60 [analog channels] into something like 600 standard-definition [digital] channels. At Comcast, we've looked at this on and off. At one point, we're going to do an area, but I don't have an update on that.
Davis: [In areas where Charter has done a cut-over], we've learned that there are advantages to doing it. One of the challenges you've got is having a set-top box on every single outlet in every single home and that's an onerous thing to think about managing. To the extent that you can continue to keep some of the analog spectrum out there, certainly the off-airs, then you're not driving a digital set-top on every single outlet. Listen, 60 percent of our customers still live in the analog world and do not have capability for VOD and things like that. [It would be beneficial] if we could drive some type of digital device in those homes to [offer] a menu or a navigator and a customer could see value with additional services like VOD and [we could get] additional revenue out of those homes. Those things are all really interesting to us and I think we've had some impact in the markets where we've had a digital simulcast out there for some period of time. It's had an impact to some degree on the level of churn as well as the level of awareness of digital in the marketplace, driving penetration in areas we haven't seen before.
CED: At the ET show (SCTE Conference on Emerging Technologies) there was more discussion about the residential node (or residential gateway), a device that hangs on the side of the house or in a closet or basement that takes digital and puts out analog for every cable outlet in the home. Is this still a long-term approach to the digital migration question?
LaJoie: We look at that stuff like every three or four years. And every time we look at it, it costs too much money.
LaJoie: In a sense, it's what the fiber-to-the-home guys are doing. They have to do it because they don't have any electrical signaling coming to the home. All they have is optical signaling coming to the home. They've got to convert it at the home into some kind of appropriate signaling for the devices you're going to plug into it like a computer or a television or a telephone. Those devices don't know anything about optical signaling; they're all looking for electrical signals.
Davis: But that's something else in the tool box that's there to help manage bandwidth. You have a big digital pipe coming into the home and then you have a residential gateway at the home that you can then convert to analog, [with] a headend-on-a-chip. As that product becomes more and more viable, in terms of its technology, and it's really early with its technology, but I think we're all staying close [and] watching it and learning about it and understanding what the economics are associated with it. Again, it's another one of the tools we can deploy and use as we have need.
CED: Onto the network DVR. There are different options out there such as allowing consumers to rent space on the server and take advantage of the old Sony Betamax ruling. What's the status of your nDVR plans in terms of the technology and timelines being considered?
Fellows: My plans are to wish Cablevision [Systems Corp.] luck and to watch them closely.
Bowick: And Time Warner.
LaJoie: We are doing PVR in the network with Start Over, but we've actually secured the rights to do that with a number of channels. It's really a very popular service. (Editor's note: TWC reports that 70 percent of customers with access to Start Over use it about eight times per month, on average.) We intend to roll this out to a couple of other places (Editor's note: TWC first launched Start Over in Irmo, S.C.., but has since put the service in front of about 65,000 customers in the Columbia area.). I think it's pretty much an eventuality and as content distribution starts becoming broader and more distribution models occur, the notion of effective, secure [content] distributed around the country is something that I think content providers will need to have. I think, logically, the cable companies have been the most secure partners [for content providers] over the decades, and we're going to play a role there.
Bowick: For Cox, we're paying very close attention to Start Over with Time Warner, and what Cablevision is doing, but our attention right now, in the shorter term, is on VOD and multi-room DVR capabilities for our subscribers. I agree, though, that network-based DVR has got to be on the roadmap over the next several years.