As network upgrades approach completion,
service providers aim to launch new services

Cable operators today must be prepared to handle a spate of technical and operational curveballs. The most recent "bender" to be thrown cable's way was Excite@Home Corp.'s decision to abruptly switch off the provisioning spigot. Fortunately, that was a short-term problem, but it still left MSOs pondering how to avoid such problems in the future.

At the same time, those bumps in the road haven't stopped cable operators from training their sights on VOD, SVOD, personal video recorders and other new interactive services and applications. While operators have taken steps to squeeze out what they can from existing set-tops, what essentials will populate their "next" box?

On top of all this, the industry hit a major milestone when CableLabs stamped the first set of cable modems and cable modem termination systems for the advanced DOCSIS 1.1 specification. As the underpinnings of the industry's PacketCable infrastructure, DOCSIS 1.1 provides QoS and other elements necessary to usher in IP telephony and other multimedia applications. Still, does that mean operators are going to make a quick switch to 1.1-compliant gear, and, if so, how soon? How does DOCSIS 2.0 enter the picture?

CED Editorial Director Roger Brown and CED Assistant Editor Jeff Baumgartner recently hosted a palaver with three of the industry's top technical and operational minds to cover these issues. Participating in the call were Comcast Corp. Vice President of Strategic Planning Mark Coblitz; Cox Communications Inc. Senior Vice President of Technology Development Chris Bowick; and Jim Chiddix, who recently was named president of AOL-Time Warner's new Interactive Personal Video Group. An edited transcript follows.

CED: We'll start off at 30,000 feet before going into some of the more meaty details. What are your top priorities right now?

Coblitz: First is finishing the rebuilds, (and) we're just about done. We're really down to the very end. The deployment of video-on-demand, and the beginning of the deployment of DOCSIS 1.1 are the main things going on. Obviously, [also factoring in are] the increased penetration of digital set-tops and our high-speed data service. That's where the investment and CapEx are going. The last one, which we'll get to later, has to do with our own provisioning of IP (Internet protocol) services.

Bowick: I think I'd have to echo that. I'd have to say that our primary focus this year is also completing the rebuilds. We'll be about 83 percent of the way there by the end of this year, and about 90 percent or so by the end of next year. If you look at our top 15 markets, which serve 70 percent of our subs, you can add about 10 points to that. We're well along the way, and pretty close to completion through the end of next year. But then the focus is to continue to drive RGUs (revenue generating units), and that's digital video, high-speed Internet, and also telephony, which we're doing very well at. And then (it's) video-on-demand. I think you'll see a similar level of spending in 2002 as to what you've seen in 2001.

Chiddix: I'm transitioning out of the CTO role at the cable company and off to something new. The cable company is mid-90 percent complete on upgrades and is working hard to take advantage of that platform. We have placed a lot of emphasis on modem deployments and also multiple ISPs [Internet service providers], which is really just beginning to take effect. In terms of digital deployment, we're 85 percent S-A [Scientific-Atlanta Inc.] and 15 percent Motorola [Broadband Communications Sector], and continuing to roll out boxes at a healthy rate. And then there's video-on-demand. We announced recently we're going to launch 10 more divisions in the next two quarters. Now that we're able to get a rich selection of product, we're really pushing down the gas pedal. It's a good new source of revenue, and a good counter to DBS (direct broadcast satellite), and we are very interested in IP telephony. Our trials have gone well, and we're formulating plans for next year, although that's not completed yet.

CED: Specifically related to CapEx, Chris, what are your forecasts for beyond 2002?

Bowick: I think what you're going to see pretty much throughout the industry, but I'll certainly speak for Cox, is a continued mix of the CapEx from the fixed capital, which is being used on the distribution side for rebuild upgrade, and then transitioning or continuing to transition to variable CapEx, or what some people call "success-based CapEx," which is all the CPE (customer premises equipment) for all of the new services. So you'll certainly see the mix change, and I think, overall, CapEx spending will begin to decline as most of the upgrade is behind us.

Mark mentioned provisioning on the HSD (high-speed data) side. We're working very, very feverishly to figure that one out with everything that's going on in the @Home front.

CED: What are your choices? You can do a homegrown thing or you can continue to do a third-party thing. What are your options, or which way are you leaning?

Bowick: First and foremost, our goal [is] to control our own destiny. What we don't want to do is get ourselves into a position that you've seen in the past with @Home, where we don't have control of our own destiny and our ability to provision our own subscribers and control our own metropolitan area networks. There are various ways of doing that. One is to build your own infrastructure from the ground up, and we've been looking at that feasibility or possibility for the last six to nine months. But there are other possibilities out there, too. You mentioned outsourcing various components, and that's something we continue to look at, and make the most appropriate business choice going forward, but we haven't publicly announced yet exactly which direction we're going.

Chiddix: This is not entirely through being clever, and there's an element of luck in there, but because we started Road Runner, and because with the break up of the Road Runner partnership with MediaOne following the AT&T acquisition, we simply own Road Runner, and it's our provisioning arm.

Bowick: Keep in mind we're about 70 percent @Home, and we have some chunk of Road Runner subs, and we have a large chunk of subscribers that are Cox high-speed Internet subscribers, roughly to the tune of about 30 percent.

CED: On the economic side, do you think you'll be able to get a better deal per subscriber than you're getting today with Excite@Home?

Bowick: I think the answer to that question in general would be yes. That would be true whether you did it in a home-grown environment or whether you did it with another third party. You could certainly do it cheaper than we are doing it today.

Coblitz: I would add [that] there was always an expectation as we came to the end of the original contract that if we were successful and got millions of subscribers as has been the case, that there would be a change in the underlying cost structure, a renegotiation with @Home. It's not unexpected that the amount of dollars that are now being generated with the success that we've had would change the mix of economics. In some respects saying we could do it less expensively is something that we've known for a long time, if in fact we were going to be successful, it was bound to happen in the @Home context.

Bowick: I think that's absolutely the case.

CED: Moving on to what's happening with the CLECs going down, and the fact that we've even seen an incumbent like SBC slowing down its DSL plans, it's pretty obvious that there's a demand for high-speed Internet, but how have some of these events impacted your own plans for next year? How aggressively will you be as we move into next year?

Coblitz: We were very aggressive this year, and we plan to continue irrespective of what the DSL guys have done. I think you'll see more types of offerings around the high-speed data offering itself–different tiering and things like that in order to become even more attractive to different sets of customers.

Bowick: I would certainly echo that. Our run the end of the third quarter was around 8,500 subscribers a week, and that's up significantly over the previous quarter, and I fully anticipate we will continue to market high-speed Internet very aggressively next year. I think we've got a great window of opportunity. I think DSL has seen a lot of fits and starts, they've had some difficulty on the technology side just understanding who they could and could not provision, and that's played directly into our hands. And I think it's been a great opportunity for us, and will continue to be next year.

CED: Another area we wanted to touch on is access to multiple ISPs. We've seen news from Time Warner Cable almost on a daily basis here, but we also like to get an update of what's going on over at Cox as well as Comcast with the trials that you have planned or have up and running right now.

Coblitz: I would say that we certainly are moving down the same road, but Jim had commented that because he had Road Runner, which was wholly owned, that meant that he could deal with all the sensitive issues of provisioning and other things pretty directly. For us, we're really working very hard on getting our provisioning up so that we can do that, and then go back and continue to push on open access. We're still doing some technical kind of work, but the real push for us right now is to get provisioning first.

Bowick: The only open access trial that we've got ongoing right now is in El Dorado, Ark., and it is really only a technical trial. There's been no business relationship established. In fact, we just brought Earthlink and AOL on board in that trial, but it's too early to give you any actual results. It's not a marketing trial; it's just a technical trial [with about 50 participating homes], so there wouldn't be much in the way of marketing kinds of results that we could share with you at this point anyway. It's just now kicking off. What I will say about that platform is that it will be the platform that we would be using in what we would call "self reliance," where we're doing our own provisioning platform for high-speed Internet.

Coblitz: I don't have all the specifics for those trials, [just] whatever's been announced by our online team.

CED: We weren't sure if that trial was already up and running.

Coblitz: No.

CED: Jim, because Time Warner Cable has so much experience with [multiple ISP rollouts], what have been some of the biggest technical or operational obstacles that maybe other MSOs might encounter when they dig into this further? For instance, we've heard that maybe setting up e-mail from several ISPs can prove to be kind of difficult?

Chiddix: Clearly it is an imperative for our company to roll out multiple ISPs as quickly as possible, and we are doing that. We are launching Earthlink and AOL in market after market, and we have obvious incentive to do that. The Federal Trade Commission put the structure in place, but clearly we will be one to offer AOL subscribers broadband service, and along with Earthlink, but the issues have not been technical, particularly. They've been operational. There are lots of issues ranging from franchise fees in thousands of different franchise areas to very practical issues about billing and communicating with the ISPs and so forth. It's blocking and tackling, and it just has to be worked through, and we're working it through.

CED: We also wanted to discuss a key milestone the cable industry hit in September, when CableLabs handed out passing grades for DOCSIS 1.1 to a couple of cable modems and two varieties of CMTS (cable modem termination system). Since we're at the point where we finally have that out of the way, what's the next step for you as PacketCable enters the fold and so on?

Chiddix: Our plans are just in the formative process, and beyond the trials we've not announced anything further, but we're clearly very interested in moving ahead. The next step is launching IP telephony and making sure that the cable modems that we've got in place can handle DOCSIS 1.1 with software upgrades.

Coblitz: Our goal next year is that we're going to install new, DOCSIS 1.1 CMTSs. Obviously, we will expand existing CMTSs with DOCSIS 1.0, and that's the beginning of getting DOCSIS 1.1 capability out into the field. Then I would agree with everything that Jim said.

Bowick: Over the last six months or so, we've been deep into the evaluation of all the various next-generation CMTS vendors. We have selected two. One was ADC [Telecommunications], or the old BAS (Broadband Access Systems) product. And the other one I don't think we've announced yet, so I won't say anything there, but we're looking forward to the time that they would actually be CableLabs qualified, and believe that they will be in very short order, and will be very quickly upgradeable to that via software. These are the devices that we will be deploying, or have been deploying for a while, and will continue to deploy through next year in anticipation of becoming fully 1.1-compliant. We'd like to push toward that, toward beginning to get 1.1 compliant through the end of next year.

CED: That's interesting, Chris, that you guys have done that deliberately. Should we assume from your earlier comments, Jim and Mark, that both of you guys really aren't that far down the road yet with DOCSIS 1.1?

Chiddix: We're doing CMTS evaluations, certainly, and we're obviously using sort of a "DOCSIS 1.05" or something at our trials, but yes, I think it's fair to say that we're not that far down the road, and most of our CMTSs are not 1.1-compliant.

Bowick: Everything that we've got out there is certainly 1.0 compliant, but not 1.1 compliant. The point I was trying to make is, for the last six months or so, we've been purchasing equipment which is upgradeable and guaranteed by the vendor to be upgradeable to 1.1. We're fully anticipating that as a possibility next year. We have qualified internally one other next generation CMTS vendor, which will also have that capability. The drive will be next year toward 1.1, but I don't think we'll have all, or nothing but, 1.1 compliant CMTSs out there. That certainly would not be the goal.

CED: Here's another question for Mark. Because Comcast has conducted IP telephony trials, what are you looking to do next year now that we're starting to see some 1.1 equipment get through CableLabs. Are you going to wait for that technology to solidify before taking any major steps?

Coblitz: We're in the same position really that Jim is. We're evaluating all of those, we're looking at the technologies, we're evaluating call agents, looking at business files, but we haven't said anything about what our next steps are going to be publicly, so that's where we have to leave it.

CED: A lot of the DOCSIS 1.1 discussion centers on IP telephony, but it also could open up opportunities for you to offer guaranteed tiering of services. Although you can still offer different levels of data speeds with DOCSIS 1.0, how do you think 1.1 will play into any plans moving forward?

Coblitz: First of all, there are different service levels, or tiering, which we could do with 1.0. You'll probably start to see different tiering coming from us where there are different levels–people can buy different speeds. DOCSIS 1.1 isn't a requirement there. It would be if we were to give a guaranteed floor, but that's not something we're looking at at the moment, although that could always be there as a business if it turned out to be one.

It's not just that DOCSIS 1.1 is for telephony. DOCSIS 1.1 is for anything which requires quality of service–any real-time multimedia product. It's one of the things, at the end of the day, which will help to differentiate what's delivered over our data networks. This could be streaming video, streaming audio, it could be certain things having to do with gaming where you want a guaranteed delivery, where you want to integrate voice with games, and do so in a quality way. As we move into other kinds of home network services that are delivered in the home, as the CableHome process is worked on, [we] would extend that QoS into the delivery of multimedia products inside the home.

Bowick: That's a very good point. I think therein lies somewhat of a dilemma for us, because we reach a decision point...where we have to make the choice between deployment of older generation CMTSs versus newer generation CMTSs from some point forward. I believe what Mark just said. If we do believe that we will be deploying next-generation technology in anticipation of their support for future services, we do have that decision point to make.

I think each of us is struggling with exactly when to make that happen. We're in a little different place from Comcast or AOL-Time Warner in that we do have a circuit-switched telephony play, and are continuing to do fairly well with that. We announced in the third quarter just under 400,000 subscribers, and just over 500,000 lines, with significant growth and, believe it or not, some actual contribution margin. It's growing quite nicely for us.

When we look at IP telephone, we certainly need a platform that provides quality of service between the CMTS and the home. The circuit switched telephony is serving us quite well now, and we're talking with vendors about strategies that allow us to move toward voice-over-IP without stranding capital. We like that model, where we can upgrade our circuit switched provide a platform which is voice-over-IP capable or circuit switched capable, and we can do that fairly cheaply.

You also asked a question relative to tiering. We're doing tiering in various markets today. Obviously, it's not guaranteed data rates, but it is tiering. I think you'll see us do more tiering in 2002. [DOCSIS] 1.1 would allow us to provide the guaranteed levels of service, but Mark is absolutely correct in that once you do that, you begin to gobble that bandwidth, and you can gobble it up fairly quickly.

Chiddix: Chris, in your tiering, I think it would be interesting to talk about whether you're tiering speed or actual usage.

Bowick: We are tiering speed as opposed to usage.

Chiddix: We certainly are [looking at that] as we deliver more and more value to customers through high-speed modems, regardless of the speed that the customer is using, and that's an easy thing to tier, both upstream and downstream speed caps. But the other aspect of value is how much stuff they are getting through it, and that's usage based tiering. I think it behooves us all to move to tiers that have defined levels of both speed and usage in both directions, so that we can charge the really heavy users more if they are using more resources and they are getting more value. Doing that is theoretically pretty straightforward. In practice, there's a lot of work to be done, and we have an initiative underway to put that in place.

CED: All of a sudden, we have DOCSIS 2.0. While reducing noise has some obvious advantages, have you seen a need for a more symmetrical data service, based perhaps on recent subscriber usage? Are customers leveraging the upstream these days for file sharing or even video conferencing?

Chiddix: There's no question that usage is getting heavier and more symmetrical. It still is asymmetrical by a fair measure, but less so than it was earlier. The usage that drives that, we aren't always aware of. Certainly things like Napster drove a lot of upstream usage. Telephony will drive a lot of upstream usage, and peer-to-peer or communications applications will drive it. Then there are people out there who have huge usage numbers, and I don't know if these are some sort of mass porno spammers or what they are, but...[laughter]

Well, it comes from somewhere, and some of them are cable modem customers. DOCSIS 2.0 is important in terms of getting more upstream bandwidth. When we first began talking about advanced upstream modulation, the theory was that it was a way to launch cable modems in really lousy cable systems with lots of leakage, and that never got a lot of traction. I think we've all found how to manage our cable plant properly and make existing cable modems work quite well. But more upstream bandwidth is certainly a good thing. Some of these advanced techniques look like they are going to make it available to us.

Bowick: Any technology that gives me more bits per hertz both upstream and downstream, I'm going to be very interested in.

Coblitz: Yes, exactly.

CED: We've had a lot of companies come through here, Jedai Broadband and Narad Networks come to mind, discussing ways for cable operators to expand plant to the business sector. And while we are already starting to see some products ready for trial, how important is this market to your company?

Bowick: The commercial segment of the market is huge for us, both on the telephony side and the high-speed Internet side. I have looked at these technologies and will continue to look at these technologies. I don't know that we've announced any specific relationship with either a Jedai or a Narad, but I do like the concept of improving the amount of bandwidth available to a specific market segment on a specific piece of cable plant, perhaps, that I can get incremental revenue from and only deploy some incremental capital that can be directly attributed to that incremental revenue.

Chiddix: I would second that. We've got large commercial services revenues in a number of markets just using DOCSIS, and there certainly are customers who would like more symmetrical [services] and just plain more bandwidth. We need to be careful as we start thinking about using up the untapped potential of our coax, for example. There are people out there with some clever ways to do that, but we may discover some day that we need that spectrum for other things. We could, for example, deliver a very high speed, very symmetrical service to all of our customers if we moved eventually to a high-end return. Before we use up the spectrum we just need to think that through. We've pushed fiber very deep into our communities, much deeper than anyone else. There are ways to leverage that, and whether the last kilometer is a fiber that we overlash, or a piece of coax that we retrofit in a particular way, I'm sure we're going to be pragmatic about going after that new revenue from business customers.

Coblitz: We have to be extraordinarily careful that we don't take up bandwidth that we otherwise need sometime later on, which doesn't mean that we won't deploy something; it means we're just going to have to be extremely careful and make sure that the business models really work and generate the value for having used that spectrum.

CED: Moving from the network into the home, we'd like to touch on another favorite subject: set-top boxes. There are a few orders out there for some of these so-called thick client set-top boxes, including some with PVR capabilities and other enhancements. It seems like the path right now is to continue to try to squeeze everything out of the boxes in the 2000-class variety. In your view, what will be essential in your next box? Especially for systems that are Motorola-based, we've been hearing that operator interest has gravitated to this new DCT-2500 box.

Coblitz: Let me at least start with the way that the question was phrased, that we sort of gravitated to the existing box. The practical reality is we have millions of digital set-tops on which the applications that we know about that can make a lot of money can be run. You don't spend capital unless you need to, and if you can find a way to run the electronic program guide, which drives the ability for people to get to all the digital channels that we have, and you can do video-on-demand, then that's where most of the money is that we know today how to get. If you are going to have a set-top [with] more capability, it has to earn its way by having applications that can't be done on the one that we already have millions of. That's the negating factor. We do have an order out for a more significant set-top that has higher processing capability and more memory in which we can do things, and we can begin to work with that set-top to deliver new applications, but there's an awful lot that we can do with the one we have.

Chiddix: I think we're all beginning to get up toward the first level of saturation of digital set-top boxes. Growth is still strong, but you can see it beginning to flatten a bit. And, until there are new drivers, there will be some limit to the uptake of digital boxes in our markets. So far, nobody has found compelling new revenue drivers from new, more powerful set-top boxes. There are lots of possibilities, and people will keep trying things. We've gotten an order in for some set-top boxes with hard drives (Scientific-Atlanta's Explorer 8000), and we will buy 100,000 of those and experiment with those as a business.

But even more intriguing is the possibility of doing new things on the existing set-top boxes, and video-on-demand is very much in line with that. The new organization that I'm building is focused on building on-demand services for the currently deployed DCT-2000 and Explorer 2000-class boxes, and looking at server-based PVR functionality and individualized advertising insertion and infomercials and a very large library of server-based content. We think that there's a lot of revenue there, and we think that there's a compelling service that will help us compete with DBS and help us drive more revenue using the very boxes we have today.

If it turns out that that's all we ever find, we will just push for lower-cost plain vanilla boxes. I'm sure the future's not so simple, though, and I think that there is a market for local storage. I think that there is an interesting market for home servers or home hubs–essentially big set-top boxes that have multiple tuners and use home networking to feed multiple sets. It will be interesting to see the innovative folks explore the market for those, but that's a long way from being ready for the mainstream.

Bowick: I think what you heard from both Mark and Jim is "show me the money." We feel the same way. We've got a mix of both Scientific-Atlanta and Motorola product out there. I agree with Mark's sentiments. You've got millions of these boxes out there. They are doing quite well in the environment. There are many applications that we can envision that those boxes will be able to support. People felt the DTC-2000 series would not support interactivity in any major way, and we've shown that that is probably not the case. Over time, you'll see some compelling interactive products reside on that box.

Others in the industry placed rather large orders for the DCT-5000 series box, and I don't know of any that are in any significant commercial deployment out there yet. It's just not very cost-effective to do so. We, too, have placed some PVR type boxes on order. We are very intrigued by PVR as a functionality within the set-top, and believe that there's a significant market out there for that kind of functionality. I have TiVo and Replay[TV] in my home today, and I love them both. But as Jim said, we have the opportunity for network-based PVR capability, and we talk to our VOD vendors a lot about that. It could be that you'll see at some point in the not too distant future a mix of both set-top based PVR and networked-based PVR being able to play in concert with each other.

Coblitz: PVR is extremely interesting from a technological perspective, as well as from a market perspective. If you look at it as just a competitive tool, meaning if satellite providers [offer] PVRs, do we need to provide PVRs? One can do so, as Jim commented. We're looking at server-based or network-based PVR capability. A PVR can be used to change the cost structure of the delivery of movies. For example, the hit movie is going to be made available at a certain time, if that movie is downloaded to the PVRs that are throughout the network. When the consumer decides to look at that movie, the movie is already there. Then we will have moved that movie in a much more cost-effective fashion. Obviously, the satellite guys can't look at it that way. They have to look at it as a way to get information into a set-top so someone can have anything on-demand.

CED: We'd also like to touch on the tricky issue of conditional access. We're bringing this up because of NCTA's recently announced plan for retail distribution of digital boxes with the embedded security while OCAP (OpenCable Application Platform) is being ironed out. But with that in mind, what do you, as operators, think you need to do in order for this plan to be implemented properly and to insure that a real market can emerge from that?

Chiddix: I guess the question is: is there a real market? There's been a lot of political pressure for retail availability of set-top boxes. In the long run, I think that's in the interest of the industry, but there are a lot of issues to be worked out. We've got PODs (point of deployment modules), so in terms of conditional access, we've got an answer, but there are a bunch of other issues about that: applications, and what happens to the guy when you take a Motorola box into an S-A system, and vice versa and so forth.

So, the way to test whether there's really demand for a retail market for set-top boxes is to go ahead and make them available and pledge as an industry that we will authorize it and offer to buy it back if the customer moves out of our area so that they aren't stuck with something that doesn't work. From a customer's standpoint, that makes set-top box ownership practical, and removes the risk.

Longer term, OpenCable–with POD conditional access and the OCAP software platform, which allows the local operator to download the local applications so that local services are available–provides a much more elegant answer. That would mean that you could actually move the hardware from one market to another and get the right services when you got there. So this is an interesting approach to see what customers really think of this.

Bowick: It's going to cause us some operational stress as we try to do this. How do we insure that the box is a legitimate box–it wasn't bought off of eBay? [We'll need] the ability to track them and make sure you've got the conditional access platform set up appropriately to be able to authorize it. These are transactions that we'll just have to muddle our way through, and I don't know that we know exactly all of the issues associated with that. But we do have a team in place at Cox to try and work through those types of issues to make sure that it's a smooth process.

CED: It's almost a chicken-or-egg proposition, right? We're not sure if there's a market for it yet, but one way to find out is to at least try to start one, and to see what happens?

Bowick: I think that's right. More and more of the DOCSIS product is being sold directly from retail, and you can't call that anything but an unqualified success. I think what we're striving for over the long haul is a model not too dissimilar from DOCSIS on the OpenCable and OCAP front.

CED: For this NCTA plan to work, until we have PODs and OCAP, vendors are going to have to obtain a conditional access license from Motorola or Scientific-Atlanta, but some of them have voiced some difficulty in obtaining those licenses, particularly Motorola's DigiCipher II. Is there anything that you, as cable operators, can do to make things any easier for other vendors that want to make it into the market now under this new NCTA plan?

Coblitz: We have PODs. If one wanted to, one could build a host, and there are some that were demonstrated at CES last year. You can build a host, we can attach it with a POD, and you don't need the conditional access license at all. So, it's not impossible to do. But, [what] I would say, probably more importantly, is that this is an interim step. OCAP is the solution. This is a way to get into the marketplace. It's not the end game of being in the marketplace.

CED: We've talked a little bit about video-on-demand, but we want to discuss subscription video-on-demand. Time Warner Cable has had a couple of trials. But I was wondering if we could go around the table and get a summary as to what you are doing in that area.

Chiddix: We've got trials up in a couple of systems, and expect to try more. Clearly, it's something that appeals to consumers. The VOD business to date [has been] a nice business. The hope is that SVOD is a larger business, because it provides a better proposition to customers. They get a large selection of titles available on an on-demand basis whenever they want it, and don't have to worry about running up a huge bill as they use the services.

CED: Do you have any other markets, or any other trials or tests planned?

Chiddix: Absolutely, but nothing that we've announced.

Bowick: We don't currently have anything going on with SVOD in an existing market, but you could expect us to launch SVOD in a few markets next year. I think our marketing department is very intrigued with the concept, and we noted with extreme interest the enthusiasm with which SVOD was grabbed in the AOL-Time Warner trial. It certainly drove traffic on the network. It drove the number of simultaneous sessions that the DNCS had to have up. So that was very interesting to see and understand so that we can understand some of the scalability issues associated with that.

I feel very comfortable from a network perspective on scalability with regard to pretty much anything we can throw at it. [One of] our greatest strengths as an industry is exactly that: the scalability of the platform. We have so many tools available to us to ensure that, in fact, the bandwidth that we have today is going to take us into the future. I think there were some bumps in the road earlier, but I think we are quickly moving past those bumps.

Coblitz: We're looking at it. We think it's a very good customer proposition. We think it's something that can be handled on the network. We'll do some trials, and I would expect that, although we certainly haven't announced anything, that it will become a part of the video-on-demand architecture that we have. Part of the products that we go with would be SVOD. I sure would want it.

CED: Regarding interactive television, what are you looking at for next year?

Chiddix: I think there's lots of work to be done in launching VOD widely. While it is an available product and it works well, and customers like it, launching it is a lot of work. There are system integration issues depending on the mixture of stuff you have in a given system. In terms of interactive television, video-on-demand is the one that makes money right now.

Bowick: I would agree with Jim. Our first focus in the interactive television arena would certainly be VOD. But, as you know, we've been working with Liberate as well on both the Motorola and Scientific-Atlanta platform, both in the Liberate "S" model and the "Compact" model. Next year, [we'd like to try] e-mail and what you might call a "walled garden" kind of application, with news, weather, sports, those kinds of things.

CED: Given today's landscape, what are your biggest concerns for the next year or so?

Chiddix: For me, it's software development, and working on new kinds of content deals and new kinds of advertising models that we'll need to build this new on-demand business.

Bowick: I'd have to go back around to the very first question, and address the priorities that we're focused on, because those are our primary concerns over the next 12 to 18 months. That includes upgrades and getting those done and over with, the continued drive in RGUs (revenue generating units), and, of course, the provisioning piece.

Coblitz: I would agree with Chris, and then I would add getting VOD scaled and out across our whole get VOD behind us, start building on that platform and continue to push into new products after that.