Search for a scapegoat

DSL has been hanging around with the wrong crowd, if several of its proponents who attended a recent Global DSL Summit held in the Czech Republic can be believed. They say the technology has a great future — if only it can separate itself from its sluggish telco handlers, several highly publicized financial failures and the practice of building out coverage rather than developing services.

Shaking loose from that crowd isn't likely to be easy. Perhaps that's why the April 20-21 event was held in Karlovy Vary in the Czech Republic, far from the prying eyes of DSL naysayers. Sponsored by Virtual Access Networks and Paradyne, the event featured panelists from those companies as well as BenchMark Technical Services, TeleChoice and the DSL Forum. About 60 people attended the event, which was held to analyze problems with DSL's initial rollout and discuss the technology's potential.

Attendees didn't deny DSL's troubled history but rather attempted to identify reasons why DSL-centric competitive carriers in the U.S. in particular have struggled so mightily.

‘It is not the DSL business that died. It was killed by the way they ran the business.’
— Leonard Conn, DSL consultant

Leonard Conn certainly has reason to question what caused DSL's fate. Conn is former chief technology officer of ConnectSouth of Oklahoma City, which was a data CLEC until last month when its financing was cut off. But Conn has no quarrel with DSL technology. Rather, he points to larger financial developments as evidence of why data CLECs have struggled. “The [data CLEC] industry has fallen victim to concerns like financial companies not letting them run the way they should,” Conn said. Profit expectations should be based at least on a five-year timeframe, he said. Data CLECs also need scrappy legal groups to deal with the impediments placed in its way by incumbents, he added.

Gavin Young, chairman of the DSL Technical Forum and vice president of engineering for U.K.-based service provider AdEvia, said somewhere along the development of DSL networks in the U.S., the rules changed.

“The U.S. rewarded big footprints,” Young said. “Now they want profit.” He predicted that the market will see less emphasis on footprint expansion and more on value-added services such as security, software and home-worker connectivity.

Phil Kyees, director of business development for Paradyne, figures that in the U.S., 70% of all loops can be served with DSL from the central office. In Britain, that figure is 98%. But all of the summit's participants agreed that valuing a company on the number of DSL access multiplexers deployed or COs covered — not the revenues generated from customer services — is a mistake.

“We've learned a lot in the past five years about [operation support systems], unbundling and line sharing,” said Adam Guglielmo, DSL analyst with TeleChoice. He noted that Korea has the highest penetration of broadband, and its telephone service is an oligopoly that doesn't unbundle services. “To get the U.S. in line, the FCC is going to have to be more active,” Guglielmo said.

One lesson most of the summit's participants seemed to have learned is that DSL is simply a transport mechanism. Unless some value is added, it is a money-losing proposition. In the U.K., for example, companies pay network service providers for video-on-demand applications.

Much discussion ensued over the benefits of providing such services from the edge or from the core. Some support provisioning from the core, while others are pushing for distributed intelligence with central management. Either way, all agreed that the customer's provisioning experience is critical to DSL's success.

Randall Barkowitz, director of operations for Benchmark, said a typical truck roll for customer service costs from $170 to $300 — but problems typically involve users' computers, not DSL connections. “DSL is not even close to half the problems we see,” he said.

Michael Clegg, chief technology officer for Virtual Access, agreed. “Eighty percent of the service calls are at the IP layer or above,” he said. “Ability to probe up five layers is critical.”

More worrisome is trying to get space in a CO, which several summit participants said is easier in other countries than it is in the U.S.

“The prospects for European [data CLECs] are better with their ability to be more flexible,” said Tim Johnson, principal analyst with U.K.-based Ovum Consultants. Since the European data CLEC is a service provider, it can put a cabinet on public ground and save the cost of putting a cage in the CO and the hassle of dealing with the incumbent.

Part of the reason U.S. incumbent carriers have been slow to cooperate appears to be that the incumbents have not found a way to profit from DSL — in fact, it cannibalizes their T-1 revenues. Rob Faw, chairman of Virtual Access, said it is easier for cable companies to subsidize high-speed Internet access with their existing video packages than it is for a telco to find cross-sells. The Telecommunications Industry Association says that cable modem usage will continue to grow faster than DSL (see figure).

TeleChoice's Guglielmo said cable firms have a huge advantage in that CableLabs assured them that all DOCSIS 1.1 compatible gear interacts. Not so with DSL, which is still grappling with standards issues.

“It's a bit like herding cats,” Young said. “It took an awful long time to get the right people in the same room.”

On top of that, there was agreement that ANSI was always ahead of ETSI. That traces back to the strength and credibility of the old Bell Labs. Now, the time-consuming standards process replaces Bell Labs' work. Or does it actually speed things up since telcos are often slow to innovate?

Ron Stein, vice president and general manager of Paradyne's technology business group, said there is not a need for too many flavors of DSL, but there is a need for three or four. Asymmetrical DSL is established and doing well. High bit-rate DSL and very-high bit rate DSL look like they have futures.

Here, again, the Europeans had an advantage.

“Europe had the opportunity to get cost-effectiveness sorted out right away,” Young said. He noted the Europeans have the advantage of learning from America's plentiful mistakes.

With the benefit of 20/20 hindsight, Conn, now working as a DSL consultant, said, “The LEC will always [win] by default unless the CLEC has a better plan, implements it better and executes it better. If you mess up, once you've learned your lesson, it's too late.”

Conn's final words of advice: “Watch your credit, manage your debt. It is not the DSL business that died. It was killed by the way they ran the business.”

Faw agrees: “The DSL market is the key to the success of the telecom market and eventually this whole economy.”

With a little guidance, a technology that seems down-and-out today is poised to be the fair-haired broadband access method of choice tomorrow.

---

Curt Harler is a freelance writer based in Strongsville, Ohio. His e-mail address is curt@curtharler.com.