VIDEO OVER DSL: LOUD BUT NOT CLEAR

Early on in the lifecycle of telco video, most vendors and carriers figured out it takes more than just buying content from an aggregator and pushing some IP set-top boxes into the market to make the service actually work. A lot more.

And while early deployments also proved video isn't going to be immediately profitable, most agree that the strategic imperative of entering the business outweighs the initial numbers. Additionally, the payoff may come a little later with future services that most telcos can provide to differentiate themselves from cable and satellite competitors.

“We had a lot of learning to do,” said Frank Weiner, vice president of field marketing at Calix, which has nine telco customers running video over its access equipment. Among the biggest lessons? Regardless of how illogical it sounds, streams of data containing video cannot be treated the same way as streams of data containing traditional data traffic.

“Video might run over IP, but it doesn't behave the same way as data,” Weiner said. “It's not only a much higher frequency, but there are a lot of things that you have to pay attention to — like, a six-decibel noise margin isn't sufficient.”

Just as important is the customer premises equipment (CPE) and how it responds to data and video streams. In the pure data DSL environment, for example, it's not uncommon for modems to shut down occasionally to resynchronize with the network. Add video into the mix, though, and the idea of shutting down a stream for two minutes while a gateway resynchronizes is simply unacceptable.

“Some CPE loses synch every day while others never do,” said Weiner. “For video, you can't have the modem go down for two minutes so it can synch up.”

Another problem arises when carriers are able to deliver video streams to the home but have trouble delivering it to the multiple TVs inside the home. The culprit often is the inside wiring. In many homes, particularly any built before the 1990s, the copper pairs inside the walls can not sufficiently support the higher bit rates video requires.

“One of our biggest challenges is in the home itself,” said Bill DeMuth, chief technology officer of SureWest Communications, which claims more than 10,000 triple-play customers in and around Sacramento, Calif. “We spend a lot of time rewiring.”

Further upstream in the network, the problems continue. While access vendors have been busy signing marketing agreements with various piece-part manufacturers to create end-to-end triple-play packages, integrating all the elements takes more than just inking a contract, said Jeff Houle, director of telco sales for Skystream, which has partnerships with multiple access players.

“We took a proactive approach before [telcos] became our customers and either invited their access vendor to our lab or brought a product to them,” he said. “That proved to us while some configurations are different, we didn't have to go back to the drawing board. When we went on-site to do the installs and the turn up of our systems, there were really no surprises.”

Perhaps the biggest lesson, though, for video-over-DSL providers is that video isn't a big money-maker for telcos. Though still early, most telcos deploying video service have done it for defensive reasons (for example, responding to a cable competitor investigating the rollout of voice-over-IP service) or as part of a network expansion into an adjacent territory under the umbrella of a CLEC. The one segment that appears to be making significant money off the telco video market is the vendor segment. Tut Systems' VideoTele.com unit was completely responsible for the company's reporting of a 274% jump in year-over-year revenue in the fourth quarter. The company also is projecting a 30% rise in revenue this year, based largely on the number of head-end systems it anticipates selling to telcos.

“We're seeing more RFP activity in part because the business case is well proven now,” said Craig Bender, vice president of marketing for Tut.

That success doesn't always translate to telcos. However, the business case also is starting to improve as telcos appraise new services that cable and satellite operators can't or won't offer. While the exact technology most carriers will use to rollout video is still uncertain, numerous basic elements will give them a big advantage, including the ability to use multicast technology from the start.

“The telephony guys are in a relatively good position because they start with a clean slate,” said Jonathan Symonds, vice president of business development for ICTV, an interactive TV vendor that traditionally sold to cable operators but has been in numerous discussions with telcos. “We have some cable customers that are looking to move to an architecture that is more IP oriented.”

In fact, the IP TV model, which telcos use when they deploy video over DSL, should lend itself to a huge number of applications that will push the boundaries of traditional video services but won't be outrageous enough to discourage consumer usage. Under the multicast architecture, video streams are delivered to the home only if users are watching TV or requesting something from the network, such as a channel change. The effect is to preserve bandwidth in the access network, which is already limited compared with cable operators, while allowing for unique services. Key among those services will be video-on-demand (VOD). Indeed, the limited bandwidth over copper is leading many to believe that VOD is the key to any telco video play.

“By the very nature of what they do, [telcos] have been given a lemon of a network, and their only choice is to make lemonade,” said David Carnevale, vice president of marketing for Path1, which provides IP TV gear to broadcasters. Carnevale believes the limitations of the copper access plant will force telcos to approach the video market from a very different perspective than cable operators. In the Path 1 vision, telcos would enter the video market by offering virtually every channel on demand. Vendors and carriers have talked about the idea for some time, but broadcasters have hesitated to allow on-demand viewing, which would allow users to skip over commercials and break the traditional revenue model of advertising-supported broadcasts.

That mindset is changing, though, Carnevale said. The BBC, for instance, is currently testing a service that stores seven days of content on servers in their facilities. Users pay a premium for the service but can view any program that aired in the past seven days on an on-demand basis.

“If you could convince CBS, for instance, to let you do what the BBC is doing, you have a significant advantage,” Carnevale said. “[Telcos] actually have a chance to leapfrog the cable and satellite guys if you buy that you can only deliver things on demand.”

Similarly, the concept of network-based personal video recording (NPVR) would let users view programs on demand, but wouldn't be limited to specific channels. Instead, users would have all the capabilities of a traditional PVR such as TiVo but without the cost of the customer premises hardware.

NPRV was a hot topic early last year as the TiVo phenomena hit the mainstream but dropped out of favor when engineering types began balking at the amount of bandwidth required for such a service. In the next year, though, NPVR will make a return, said Weiner. And telcos are in the perfect position to offer it. Business models that incorporate both a revenue stream for the video operator and the broadcaster are a little difficult to detail though.

“I could see a time-delay service where content will be trickled down while you're watching something else,” said Weiner.

Not everyone is sold on VOD or NPVR as the differentiator, though, in part because cable operators have been quickly rolling out their versions of VOD over the past couple of quarters. And satellite providers have quickly incorporated PVRs into their set-top boxes.

“From a VOD perspective, if you're getting it from your [telco], are you getting something compelling enough to sign up for the service?” asked David Messina, vice president of marketing and product management for CoSine Communications. “In the U.S., there's a lot of talk about VOD first. What we found [in Asia] is starting with the consumer, there's not as much of an uptake in part because there are other models available to get that service. The bigger market opportunity is really on the broadcast side of the house.”

CoSine is pushing a model that blends traditional broadcast service along with access to Net-based broadcasting. Other vendors, including Tut, say the fastest and easiest way for a telco to differentiate itself is through hyper-local broadcasting, such as showing local high school events.

“A nationwide franchise isn't going to go down to the local school board and get the local content,” said Bender.

Many independent telcos also have been following that game plan by injecting local off-air channels into content from an aggregator or larger consortium, said Geoff Burke, director of global video marketing for Zhone Technologies.

“A lot of telcos are operating with a micro head-end,” he said. “It allows you to have things that are really going to be of local interest. “In a lot of small towns, people are looking for these things.”

Playing off their local presence, some carriers are providing simple services that are particularly popular in specific markets, he said. In the Southeast, for example, some telco video providers have had success simply by putting Speedvision, the auto-focused cable network, in the basic tier of channels. In other markets, telco video operators are refusing to carry adult video channels, a stand that often plays well with the local community.

“Not only do they not want the content, they don't want their local provider to even carry that content,” Burke said. “A lot of times we get stuck in the technology of what that new service is going to be when a lot of it is just basic marketing blocking and tackling.”

Whether it's the local school play or Net broadcasts of cricket matches from India, telcos over the next year must begin to start think in terms of differentiation, said Carnevale.

“At the end of the day, these guys are all competing against a very good video product from satellite.”