Cablevision’s Set-Top Moves
Keep Open-System Dream Alive

HARDWARE
By Matt Sump and Jeff Baumgartner
From The September 16, 2002 Edition Of Broadband Week

WILT HILDENBRAND Cablevision Systems

BW: We wanted to give you a chance to clarify and react to some of the coverage out there about Cablevision’s decision to go with Scientific-Atlanta set-tops and headend equipment for its New York-area system.

Wilt Hildenbrand: What I’ve seen written up about this is how hard everybody is making it sound. That’s the part that fascinates me a little bit. I keep reading about a Needham [& Co.} report that set the stage with this parallel network silliness and everything else, and you guys seem to be following it and, quite candidly, I don’t get it and I’m sitting here in the middle of it.

Yes, the big step is working with Scientific-Atlanta to do [a conditional-access system from] NDS [Group plc]. Some of that is a little bit of religion and some of that is we liked the direction we were and are headed in with the Sony box. We don’t want to give that up. It makes it easier to mix and match, should we come to a situation where we want to do that.

The conditional-access system is one of the bigger components of any of these digital systems. You have the encoding process to digitize the signals. Other than a price and quality perspective, you can pretty much interchange encoders, not encryptors, but encoders.

So let’s say we’re going to continue to use the same encoded content — some being direct turnaround from the bird, some of it self-encoded that we use with Sony. At the moment at least, and for the foreseeable future, we’re going to continue to work with SeaChange [International Inc.], so we get to use the same VOD system. We’ve been using BarcoNet for the transport system, to carry signals around to the various places where we launched Sony, so that’s not a change or a parallel network.

Other than, yes, the S-A system has a separate controller from what we’re using to authorize the Sony boxes, I don’t know that we’re running a parallel system here at all, guys.

BW: Is it correct to assume the Sony headend and Sony boxes are in only X percent of the New York DMA and that vast portions of that area don’t have any type of digital headend?

Hildenbrand: It doesn’t make a lot of difference to me where the Sony boxes are or aren’t because most of the work — VOD, transport, encoding and, as we do with NDS, even encryption — is the same for both systems. Sure, the Sony boxes, for all intents and purposes, are 100 percent server-based. Sure, the S-A implementations up to this point and probably going forward will have some imbedded clients in them as opposed to server-based clients.

But architecturally, certainly at a high level, that isn’t two different networks. That’s just two different executions.

You know that I know that we didn’t get the Sony boxes populated throughout our entire area. We certainly didn’t put any in the Bronx and Brooklyn yet. We certainly didn’t put any in Connecticut, but there are some on Long Island, there are some in New Jersey.

I would have gone for the same type of system whether it was a Sony platform or not because we haven’t really decided to step back and go with just a channel-expander platform. We’re trying to follow the same logic we were following pretty successfully with Sony, which is trying to stay as close to an open standards based platform as possible.

BW: What were the reasons behind the decision to go with S-A boxes and headends?

Hildenbrand: I could take you through some of it, I don’t know if I could take you through all of it. We started to do something with Sony. We were actually relatively successful in getting there, short of a couple things. One, we were late, which I don’t think any of the vendors we’re talking to now didn’t have their period of lateness going.

Two, we designed a very high-end system that was focused on certain interactive-TV capabilities and certain other capabilities.
What happened to both Sony and us, our lateness coincided with an implosion in some of the interactive space. It became less important to focus on that and more important to focus on some other aspects of digital television, like HDTV [high-definition television]. We ended up agreeing it just didn’t fit the times anymore.

Technologically, I’d do it all over again. You got to see where we were able to take it both hardware- and software-wise. It’s pretty amazing, especially as you look at things you can run along side of it.

BW: Have you encouraged Sony to get a PowerKey license from S-A and build boxes based on that conditional access system?

Hildenbrand: We encourage Sony to do what they think is best for Sony, including, as they stay in the set-top box business, look at a way that we can be not the only ones they are selling boxes to. I wouldn’t ever steer them in one direction or the other.

On the other side of the street, you have S-A, which we happened to choose here, who went with PowerTV and PowerKey, although we’re going with NDS here. You have Motorola, who took a little more open look at the world and said: ‘Bring whatever middleware you want.’

We’re relatively transport-agnostic, VOD vendor-agnostic and conditional access-specific, but agnostic from a proprietary point of view. It’s there but for a couple of pieces of business that Motorola [Inc.] is in the system. And I wouldn’t say there isn’t a chance for them to get into the system.

[Cablevision New York president] Tom Rutledge has said it publicly: ‘We’re not going to ride just one vendor going forward.’ For a little while, we need to stay focused.

I don’t need to spin six plates at the same time, but we’ll probably in a very short order get to introducing other vendors.

BW: The system will continue to run the NDS. It doesn’t sound like PowerKey will play a part here.

Hildenbrand: We may start with it for expediency purposes, then switch over later in some areas. And in some areas, we’ll probably start with it out of the gate. I don’t think all those things are defined right now. I think the quote you got from NDS, saying the effort is about a 90-day effort, is about what we all figure.

We’re still working our way through it, but I’m looking forward to it. It will be interesting to see it go. It’s sort of an underlining of everything we and Sony did, if we can bring another vendor in here and have them match up.

BW: The dream of having an open-access system is alive and well in your future vision. Am I correct in concluding that?

Hildenbrand: I think you would [be].

BW: I also think some people might conclude you will just throw out NDS in a year or two and go with PowerKey, but I gather you would not necessarily want that to happen — that you could have them sit side by side and prove this open-access system can work, because you haven’t seen the new services develop, yet, for such a platform.

Hildenbrand: It’s like the great American novel. Unfortunately, there’s a little bit of fiction contained in there. You have got to be careful with stuff, because the turning points are so small that you don’t always see them as you are running past them. You’ve got the idea, with a very large ‘except.’

We don’t intend not to have PowerKey capability in the boxes, we just don’t intend to use it at some point. Although we are looking at a simulcrypt-type format, we’re not looking at running in a simulcrypt function, that is, where I’m literally running two conditional-access systems at the same time. This is the ‘except.’

I’m looking at focusing on the NDS conditional-access system, keeping PowerKey, since it seems to be willing to work that way at this stage in the S-A platform as sort of a bench bolt, if you will, where I’m not necessarily running it but I could fall back to it if we chose to at some point in time. As far as leveraging all the other products and services, we’re kind of taking it one step at a time here.

Cablevision's Wilt Hildenbrand said new developments convinced the MSO to add S-A gear to its digital mix. Pictured: The iO: Interactive Optimum menu screen.

BW: Have you considered the Explorer 8000 for customers who want an integrated DVR?

Hildenbrand: That’s certainly the difference between a Sony/Motorola and an S-A. These guys have been at it so long, they have high-definition and DVRs and PVRs. At Motorola they have this stereo receiver and gateway device. We’re looking across the board at those things.

The interesting live-by-the-sword, die-by-the-sword thing here is you look at a high-end box like the 8000 and it’s [Digital Audio/Video Interoperability Council]-based. It seems to me to be a contradiction in terms to go out with a high-end box that’s DAVIC-based versus [Data Over Cable Service Interface Specification]-based. So we’re looking to stay DOCSIS-based, for the most part, for our core digital deployment. So we’ll look at the 8000, but we’ll probably look at it when we get around to putting DOCSIS in it versus DAVIC.

That’s not a condemnation of DAVIC. It’s just if you’re going to go that high-end of a box, and spend that kind of money, you ought to build in the bidirectional capability you get in DOCSIS versus DAVIC. And it’s not that DAVIC isn’t bidirectional — it’s a scalability point.

BW: Is there anything in the 4200 that you really liked and that sold you on it?

Hildenbrand: It’s similar to what we had been working on all along with Sony, so it was an easy move. We had looked at cost-reducing some of the things in the Sony platform, because we really shot high. I think that’s why the GI [General Instrument Corp., now a unit of Motorola] guys were telling you we looked at the 5100. We’re looking at that family of products.

BW: I assume Cablevision will be able to do all this within the new capital-expenditure numbers you’ve announced to Wall Street?

Hildenbrand: You guys heard the announcements at the analyst conference. The core basis of the plan they put going forward, on the migration to this, the buildout of the system and the buildout of the plant and the adoption of these boxes, is part of the of the rolling plan. The good news is we are able to reuse a lot of what we did before. We didn’t have to throw out everything.

Let me be clear. I’m not using the same SeaChange server for this whole deployment. I’m having to deploy other servers. You have to put a mux per headend. You have to put a SeaChange server per a certain number of subscribers. It’s exactly the same architecture I would have built out, short of some of the very S-A-specific stuff. It’s the same architecture I would have had to build out to put Sony everywhere.

Everything that we had started to plan out to get a head start in going digital in Connecticut, New Jersey and everywhere else is still viable. It isn’t like we have to run the Sony system and quickly crank up an S-A system to run alongside of it. That’s not really true.

There are very specific S-A translation pieces we have to put in, but a lot of the existing stuff is either the same as we did with Sony, or it’s exactly the same pieces we did with Sony. And the difference there is whether or not and when we get around to running S-A side by side with Sony. Yes, we’re working on making that possible, but I can’t tell you from a product perspective if we’re going to do it. I can tell from an engineering perspective it’s going to be possible.

BW: On the conditional-access piece, on the STB side, will PowerKey be present in a removable card, or is it still embedded in the box?

Hildenbrand: PowerKey is an embedded system, so it doesn’t require a card. It turns out that the S-A box has a smart-card slot, and we’re just going to leverage the existence of that. A system that is running NDS will have a smart card in it at its simplest level.

You make decisions between two vendors that have been in the business a long time, the decisions aren’t always a sound bite. It’s a summary of a whole lot of little things and in this particular case, the whole lot of little things got us to S-A first. But I tell you, Motorola did well enough. They are a viable player. You could end up seeing some of their stuff here too.

BW: Can you walk me through operationally what happens with an installer with an S-A box? Will he activate the conditional access with an NDS card? And who keeps the card?

Hildenbrand: Right now, the card and the box get mated in the customer’s house by the technician when he goes to the house. Assuming the S-A scenario is going to follow the same thing, the installer will sign out X amount of boxes, X amount of cards, mate them up when they get to the house. The box is then provisioned from the headend and the system comes up. That’s pretty much how it works.

Cards are not necessarily assigned to a box at birth. They are assigned as part of the provisioning system, and it stays with that box until it either breaks or the box gets turned back in.

BW: If you wanted to switch to PowerKey in any of those boxes 18 months from now, would that mean a truck roll, or could you provision it from the headend?

Hildenbrand: The goal will be no truck rolls. Provisioning would be a software-controllable event. Having done a few software-controllable events here, it does work.