Solving a world of issues
In addition to rolling out advanced services, some of the industry’s
top technical minds are figuring out how to tie them all together
With rebuilds completed or mostly completed, most MSOs have shifted their attention to the advanced broadband applications and services that will run on the platforms they've spent a good portion of their lives building. But in addition to the usual service suspects (video-on-demand, digital video recording and Voice-over-IP), some of the industry's top technical minds are attempting to break down service silos and operationalize the business of technology–all in an effort to create a seamless service platform that better serves the bottom line.
CED assembled its annual virtual roundtable to explore how these issues are being handled today and what's on the drawing board for 2004. Joining CED Editorial Director Roger Brown and CED Editor Jeff Baumgartner on the call were Charter Communications Senior Vice President of Engineering Wayne Davis; Comcast Cable CTO David Fellows; Cox Communications CTO and Senior Vice President, Engineering Chris Bowick; and Mediacom Communications Senior Vice President of Technology Joe Van Loan. CED held a separate call with Time Warner Cable Executive Vice President, Advanced Technologies Mike LaJoie, whose comments were inserted into the following edited transcript:CED: Going around the horn, what are your top three or top three or four priorities heading into next year?
Fellows: Our first priority for next year is to finish off the upgrades. We reported in [the third quarter] that we expect to do 53,000 miles this year, which means that we end the year in the low-90s in terms of being upgraded. Next year we will just finish off the remaining five or six percent of the plant that needs to be upgraded.
Number two is to continue the roll out of video-on-demand, high-definition TV and advanced video services and interactive video services, including PVRs (personal video recorders) and high-def set-top boxes.
Number three is keeping up with the high-speed data demand, and four is beginning the trials and testing on a wider scale of Voice-over-IP.
‘A lot of people believe
that tiering is really more
of a defensive strategy
against what the RBOCs
are doing, and I don’t
necessarily look at i
that way at all.’
On the data side of the equation, we're in the process of doing a lot of work around transitioning the platform to (DOCSIS) 1.1 in anticipation of Voice-over-IP.
On the voice side, we have begun the process of trialing Voice-over-IP this year in Roanoke, Virginia, and we're anticipating a couple of market launches next year.
LaJoie: I don't know if this is in the order of rank, but it's more in the order that we have three main products: video, data and voice. Voice is the newest, but we're focused on continuing to advance new video services, continue video quality, adding more on-demand content on the video side of the house, along with lots of HD.
We'd like to continue to grow HD subscribers. I think we've already, in some markets, increased the [high-speed] bandwidth that we offer. We'll probably continue with more of that. We also want to aggressively launch Voice-over-IP next year.
‘We’re convinced that,
like others, it’s time to
move beyond the technical trials into
–Van Loan, Mediacom
We're completing the deployments of HD. We still have some 550 (MHz) systems left. We went through those this year, and where they were conventional trunk-and-feeder, we converted them to HFC, and we're putting in multiplexing to get some capacity back for the addition of new VOD and HD services.
Lastly, we've had a small, quiet telephone trial running, and we're convinced that, like others, it's time to move beyond the technical trials into commercial deployment of Voice-over-IP.
Davis: We have pretty much finished our rebuilds. At the end of this year, we will have wrapped up the rebuilds for the most part in the last properties that we acquired from AT&T (Broadband), and St. Louis is the biggest of those. Sitting right now we're at 94 percent of our plant at 550 (MHz) or better. Of that, about 87 percent is 750 (MHz) or better of the total number of miles of plant that we have.
‘We’re very excited about
the DVR. We’re taking a look at where we can
sell it in, and we’re
getting behind it with capital. We think it’s a
big part of our strategy.’
–Davis, Charter Communications
Expanding the VOD footprint is also important for us. We're in 22 markets now. We've [also] got some pretty aggressive goals for DVR across the company.
We have a commercial deployment (of) Voice-over-IP in Wisconsin, with about 2,000 customers on that. We're beyond trial, we're in commercial rollout, and we'll turn the heat up in that market through '04. We have the CBR (constant bit rate) deployment in St. Louis that we acquired from AT&T, and we'll have two additional markets that we'll roll IP telephony out in '04 as well.
One of the other things that we're really targeting…is shifting our workforce from a heavy reliance on contractors to more of an in-house effort. If you have people that have ownership in their cable system, their local market, that means an awful lot in terms of staying with you and the quality of the work that they do.CED: A clear trend appears to be on the operations side of the house, making sure everything is fitting in and working as efficiently as possible. What else are you doing in that area?
Bowick: We need to maintain a very, very strong operations focus for next year. We've had great success this year improving service calls to the tune of about 20 percent, and that's saving the company a lot of money, and I would anticipate that we'll focus a lot on continued reduction in trouble and service calls next year, as well.
‘I will be extremely
happy in ’04 if we have
a common vision, a common definition for
–Fellows, Comcast Cable
As you bring in new businesses, it's important to layer those onto your existing platform, your existing back office, work processes, software programs, OSS...and not to build a separate silo. As tempting as it is to make some quick money in a new business by rolling in a silo of back office support, one of the things I'm spending '04 doing is sort of breaking down those. I've spent '03 doing it. I'll spend '04 doing it.
2,400 hours of
Time Warner Cable
LaJoie: We should be getting to around 2,400 hours minimum in each [VOD] system. Some systems will have more storage than that. In 2004, we'll get to that at least at a minimum. Some systems do have as much as 6,000. It has more to do with how it was configured than anything else, but we think 2,400 hours of standard-definition storage is a minimum.
We're going to launch HD-on-demand. I think Austin has some HD-on-demand. There aren't a lot of HD titles available for on-demand, but that's increasing. As both the title availability and the availability of the systems [grow] so that they can accommodate moving the larger files around, we can see it going everywhere once we can get that under our belt.
Fellows: At the end of this calendar year, 50 percent of our homes passed will have our video-on-demand service, which is a combination of movies-on-demand, subscription video-on-demand, and a set of content which is available to anyone who is a digital cable subscriber for no extra charge.
In terms of the video-on-demand architecture, the plumbing that I'm rolling out right now is 8,000 or 9,000 hours of capability. We have programmed about 1,500 hours of that. But that programming tends to go up in 500-hour chunks as you assign a new pay service, which comes along with a subscription video-on-demand component.
At the end of this year we're going to trial something that scares me...and that's high definition on-demand. With no extra hours of content, that's roughly five times the capacity, so we'll very cautiously test the value of that proposition to our customers.
Bowick: For VOD, we're in four markets right now, which is about 25 percent of our homes passed, and we're looking at four additional markets by the end of the first half of '04, which would put us in roughly 50 percent of our homes passed.
On the DVR side, we're launched in two of our markets now, and we're very pleased with the results that we've seen in those markets. We're anticipating another four markets through the end of this year, which would put us in front of about 30 percent of homes passed.
Davis: At present, we're 22 markets (with VOD), which is a large part of our subscriber base because they're the larger markets. As far as storage hours, today we're at about 1,000 hours. In '04 it'll be 1,500 hours. That's what we're standardizing on.
For the DVR, we have a trial going on right now with a Digeo-Motorola box, the BMC9012. We're [conducting] a limited trial in the fourth quarter, and then roll out in first and second quarter of '04. We're very excited about the DVR. We think that this has competitive need on the DVR/PVR side. We're taking a look at where we can sell it in, and we're getting behind it with capital. We think it's a big part of our strategy.
Van Loan: We currently have three VOD markets. We're going to be busy. By the end of our year, 50 percent of our subs will have access to VOD. Regarding the number of hours, we currently run between 1,200 and 1,500 hours. For the deployments we're doing this fall, because a lot of these are small markets, we're turning toward the low side with the capability of adding hours as demand comes along.CED: On the high-speed data front, could everyone chime in with their progress with DOCSIS 1.1 or even the migration to 2.0? Are you doing just one or a combination?
Fellows: We will do 1.1 in a system this year, but only one isolated area as a test. By the middle of next year, we will upgrade 1.1 everywhere. We've also looked at the budget implications of going back and retrofitting 2.0 into our 1.1 world, but I'm not sure whether I'll do that or not.
Bowick: We've begun the process of transitioning to 1.1. I'd say that about 84 or 85 percent of our CMTSs are there. About 28 percent of our cable modems are already there. We really haven't begun worrying too much about 2.0 at this point, although any modem that we purchase and provide to our subscribers is today 2.0-capable. We're not buying anything that is not 2.0-capable.
Van Loan: We're taking 1.1 out there. We expect at the end of the year to have about 80 percent of our headends converted to 1.1. With respect to 2.0, we see that on the horizon for us.
Davis: Right now, we have one market running 1.1, and that's our IP telephony market. By the end of '04, we'll be there to the point that a very small portion of our base won't be capable of supporting 1.1. At the end of '04, we're projecting with additional launches and then some conversions about 20 percent of our base will be at 2.0. Again, these are just our largest markets. At the end of '04 we're projecting about 45 percent of our modem base out there to be DOCSIS 2.0 modems.CED: What are each of you doing with data tiering as DSL continues to try to put pressure on you with pricing?
Bowick: A lot of people believe that tiering is really more of a defensive strategy against what the RBOCs are doing, and I don't necessarily look at it that way at all. We continue to win seven out of ten high-speed Internet customers in our markets compared to DSL, so it hasn't really mattered that they've continued to drop their pricing. HSI tiering is more of an offensive strategy.
What we've seen in the markets where we've had tiered services before, we really haven't seen any...of the cannibalization of the premium flagship tier. What happens is that there seems to be an equal number of individuals who want the high-speed tier as do individuals who want the low-speed tier, and from our standpoint, it seems to be a pretty good wash. So we're looking forward to rolling this out more as an offensive strategy this year, and next year.
Fellows: We've not seen a pressure to lower our price or create a low price tier, a low value tier in response to competition. In the third quarter, we signed up almost 473,000 subscribers and are seeing continued demand across all areas. At some point we believe that the penetrations will hit numbers where to continue growth, we will want to go after those dial-up subscribers. We're not ruling out introducing a tier at some point, but not now, and not in response to anything to do with DSL.
Davis: Charter has always been tiered, and I think there's a lot to be learned from the Charter experience here. We had 256 (kbps), 768 (kbps) and 1.5 (Mbps) were the original tiers. We're actually moving away from that lower end tier and going to a two tier. Having that lower end tier has affected our ARPU (average revenue per unit). I don't know that that's so much a result of perhaps how you sell it in. If you sell it in that low end, of course that's what you're going to get.CED: The VoIP discussion has changed over the last 12 months. Last year there was lots of talk about trials, but now it's about deployments. Does that mean the technical and operational issues of VoIP have been solved?
Bowick: I think what has changed in the last year is that all of us have been really, really focused on it. I'm talking about for the last year to a year-and-a-half (focusing on it) in our labs, working the bugs, getting the bug fixes and working that level of detail with our vendors, doing the systems integration amongst all of the various servers and the PacketCable architecture and just working out the kinks in those interfaces, that has really been the lion's share of the effort in the last year.
I think we've progressed in leaps and bounds in the last year working those issues. And now the only real issue or question as we proceed is how scalable is the architecture? That's because we've just never done it before. That's a little bit of a concern going forward.
Van Loan: I guess part of our philosophy is never be first or last, and certainly with telephony that's probably a good one. We had a technical trial that we started about a year ago, and it was a good learning experience. Basically we used our CMTSs and integrated MTAs to provide connectivity to a Class 5 switch with a GR-303 interface and a bridge. We learned, one, we can do it and make it work, and, two, for us to take a brick and mortar approach by starting fresh probably doesn't make sense.
We currently have an RFP out in which we're seeking a partner that could provide all the switching and back office capability. We think that by next year we will be deploying commercial Voice- over-IP telephony in a couple of markets at least.
Fellows: We're more like Cox. We've got a circuit-switched business out there. Again, we reiterated [in the third quarter conference call] that it's a business that is making money. One of the reasons that it's making money is that a year ago, in merger time, we said, "Let's just cool it for a while on telephony, and go back and fix our video business, fix our high-speed data business, and then get back to telephony."
We've done enough work now in understanding the circuit-switch business that we own, and exploring the Voice-over-IP options that we are now convinced that this is a good market where we can make money. Going from this year, where we've been in one market, we're going to three markets next year, with each market (having slightly) different characteristics. Then we'll look to scale the business by the end of 2005.
LaJoie: We plan to be in three to four markets by the end of this year, and most of our other markets throughout 2004. We're in one division currently: Portland, Maine.
Davis: We'll expand two additional [VoIP] markets in '04. We have an RFP that will hit the street [in November] for the switch in those additional markets. It's not a market-specific RFP, but more of an architecture RFP.CED: Earlier this year there was a lot of discussion about this migration to an all-digital environment. A lot of the benefits of doing that have been discussed, but moving forward, are any of you testing this now or have plans to test it next year? If you are, what approach are you going to take?
Van Loan: We have been modeling a little bit, and we're probably going to try something in one of our systems that's been bandwidth-restricted. But we've got lots of concerns.
There is currently a sub-$100 box out there, but it's strictly a digital box. It means that all of the channels on the cable system have to be digital, which creates some issues. For one thing, broadcast basic will have to have dual carriage, because under the (1992 Cable Act), broadcast basic customers have to be able to use their cable-ready features– picture-in-picture, recording, and so forth. In that environment, every television set in the home will have to have the box in order to watch any television beyond broadcast basic.
It means a fairly substantial headend investment, which can be a problem for small markets. The alternative that I've wondered about is a combo box that has an A-B switch, a bypass switch like the low-end satellite box.
To some degree, it's not an "if" but "when" discussion as we continue the transition to digital in this country. So we are looking to do something probably in a smaller market on a trial basis next year.
LaJoie: We don't have any firm plans, at least not from a commercial perspective. We're starting to test the notion of going all-digital. We're looking at digital-only devices. I think that right now where I'd fall out is that there's going to be a migration to all-digital. There won't be a hard cut. It'll happen over a number of years. That doesn't mean that relatively soon we wouldn't see an all-digital device.
Bowick: The first thing I'd like to do is give some kudos to my friend Dave Fellows on this issue, because he has been very out front as Comcast relative to trying to define for us this infamous $30 box solution. It's amazing when Dave speaks, how quickly vendors react to that.
I think 2004…is really going to be more of a strategy year relative to the all-digital solution. I might do a trial of sorts next year, but my guess is probably not.
Fellows: I will be extremely happy in '04 if we have a common vision, a common definition for [an all-digital migration plan]. This winds up being not just about the set-top box, but about your whole architecture, and I don't mean hybrid fiber/coax. I mean the signaling mechanism, the modulation mechanism, what functionality is in the $35 set-top box, and what resides in the headend. It's an entire approach to providing services to the home. I'll be happy if we reach agreement on how to proceed, because I think the chip guys and the product guys are all lining up. What they want now is direction.
Davis: We put a big toe in the water already in this. We're converting one market. In fact, it'll be turned up before Thanksgiving to an all-digital market. But let me define what all-digital means. We are taking all 80 channels of analog and simulcasting those in 7 QAMs and coding the entire lineup. It's an 870 system, so we have the room to do that.
I think we'd be happy over time to be able to share some of the results there, but we're going to try to learn from that market, and I think when you take a holistic view of an ROI on this it may be more compelling than we think.
CED: As for that trial market, can you tell us where that will be?
Davis: Not yet. Part of it is that there obviously are some programming issues relative to this. The programmers are very anxious to do this. They know that the world is going to go this way, and I think that they're looking for some solutions, as well. But we're not 100 percent there.
We'd love to be able to get this out in front of our customers as quickly as we can because if you look at an all-digital picture through an HFC plant, you know that's pretty robust picture quality that you're putting in front of your customers to be able to compete against the all-digital package that DBS is.