DSL: Although digital subscriber line technology is still in the trial phase with most carriers, at least two have deployment plans for 1997 >BY BETH SNYDER, Switching & Transmission Editor

The good news about the asymmetrical digital subscriber line trials in the U.S.-designed to deliver bandwidths of 1.5 Mb/s or higher over twisted pair wiring-is that every Bell regional holding company as well as GTE is running at least one. The bad news is that only two carriers have committed to a time frame for commercial deployment.

In less than a year, the major local exchange carriers have gone from having no ADSL to hooking up users, testing systems and sorting through all kinds of issues. The ups and downs have been numerous, as evidenced by some of the news from the trials.

Good news: Carriers are convinced that the technology works and are now beginning to work out the details of offering the service. Bad news: Many details need to be worked out, including billing, packaging, marketing and network management.

Good news: ADSL is easier to install than ISDN. Bad news: It still takes inside wiring work and copper line cleaning.

Good news: RHCs are pushing hard for widespread, commercial deployment. Bad news: Customers have to be within 18,000 feet of a central office or even closer, knocking out an estimated 20% to 30% of potential buyers.

Good news: Prices are dropping on ADSL modems. Bad news: Prices are hovering at about $1500 per modem and won't drop much until mass ordering begins, which won't happen until mass deployment begins-the "chicken or the egg" dilemma.

Good news: Internet service providers in some areas are taking the initiative and offering ADSL before local carriers. Bad news: Remember those inside wiring issues? ISPs don't have nearly the level of expertise of the phone companies. ISPs also must rent scarce and expensive space in a telco CO to offer ADSL service-and they may even have to install their own lines to the customer premises.

Good news: New ADSL entrants are jumping into the market every day, giving carriers lots of choices. Bad news: New ADSL entrants are jumping into the market every day, giving carriers lots of choices.

What a difference a year makes Just a year ago, the telecom industry already had pronounced ADSL dead. None of the major carriers was looking very hard at ADSL, originally conceived for video-on-demand applications that proved to be tremendously expensive to deliver. Bell Atlantic in northern Virginia was the only carrier even running a video-on-demand trial.

Then along came the Internet. Consumers were very interested in surfing the Web-but downloading pictures and files became an exercise in patience with 14.4 and 28.8 kb/s modems.

At the same time, the number of telecommuters began to blossom. Even with a slow modem, workers could access information, e-mail and even corporate local area networks with home computers. But they, too, wanted faster transmission.

The need for speed caught the attention of both cable and telephone companies. Cable companies began hyping cable modems, offering speeds up to 10 Mb/s over cable television infrastructure, while telcos turned to the recently discarded ADSL. Between February, when GTE began the first trial, and October, when Ameritech became the last RHC to announce service, test cities were chosen, customers targeted and data collected (Table 1).

As the one year anniversary of the first trial approaches, at least two RHCs have announced definitive plans for ADSL service. U S West will begin offering single-pair high-bit-rate DSL followed by ADSL in 10 cities, including Denver, Salt Lake City, Seattle and Portland, Ore., by mid-year. And Pacific Bell plans to offer ADSL service in 10 COs beginning in September.

Other carriers are sure to follow close behind. Driven by the real or perceived threat of cable modems, customer demand and the need to unburden voice switches tied up by long Internet calls, the RHCs are bullish on ADSL.

Already in the game The RHCs already have been beaten by some ISPs that have begun to offer ADSL service. CADVision in Calgary, Alberta, is offering service at about $70 a month to customers. In the U.S., InterAccess in Chicago was the first to offer service.

InterAccess has hooked up five COs in the suburban Chicago area for about a dozen customers, according to President Tom Simonds. Although encouraged by the initially favorable results and response from customers, he admits it hasn't been a cakewalk.

"It's been wonderful once it's in," says Simonds. "But the set-up is atrocious. It's really a hard service to install because it's complex. There are complex inside wiring issues on both sides.

Telcos probably have an advantage over ISPs when it comes to wiring, he concedes, because they have been doing it for years. But at least one analyst says the difficulty of installation will probably become an issue for everyone this year.

"Reality is going to hit this year, and we're going to see that it's not easy to install," says Lisa Pelgrim, an analyst with Dataquest. "Phone lines are messy. It's a case of theory vs. reality." InterAccess isn't giving up on ADSL, though. "Feedback has been very positive. People are definitely interested," says Simonds.

At a cost of about $10,000 for each new CO location, Simonds says his company can add more users and plans to do so. The end users have to buy the modem-Westell Technologies' FlexCAP in this case-at a price of $1500. If several customers in a CO serving area demand ADSL service, InterAccess will consider adding the corresponding CO.

InterAccess is not seeing the full throughput of 1.5 Mb/s for its ADSL connections, but rates are as high as 800 kb/s with some applications, says Simonds. Bottlenecks in InterAccess' own infrastructure are responsible for the decreased throughput, he says, adding, "It's still a magnitude of performance over ISDN.

What about the RHCs? ADSL technology works. That's the standard answer from the major LECs when asked what they've learned from their ADSL trials.

The trials have been running long enough that they agree it is indeed a workable solution. So what's next? Carriers now are beginning to focus on working out the details of offering ADSL as a service. Call it lessons they've learned from ISDN, but telcos are not about to roll out another service that requires customers to have extensive technical knowledge. The key to offering any of the coming DSL services, they say, is to make it customer-friendly. They also plan to better prepare their own staffs in answering questions and taking orders.

"We want to take advantage of the high capacity without the pain of the technology," says Lorene Steffes, vice president of technology development at Ameritech. "The ISDN experience is one of the things that we're looking at as a guide.

Many carrier trials have been small technology trials, managing and servicing between 10 and 100 users. However, Bell Atlantic is already running an ADSL market trial. Customers pay around $59 a month, which includes Internet access. To support the trial, the carrier is using the same service model it used for creating an ISDN turnkey solution. In fact, it is even using the same customer support system, InfoSpeed.

"We applied the same model to ADSL and found out it worked even better-ADSL technology is somewhat easier to install and manage," says Frank Nelson, ADSL director at Bell Atlantic. "My goal is to have it appeal to those people who don't know a lot about the technology.

He says the contract service technicians the company is using for installation-from Bell Atlantic Communications and Construction Services-are installing systems within several hours. Unlike ISDN, ADSL service doesn't go through the CO switch, so it doesn't have to be provisioned or configured through the switch fabric.

Telcos believe ISDN will continue to thrive alongside ADSL, however. Pacific Telesis predicts that ISDN will continue to have double-digit growth through the next few years, says Mark Gallegos, director of DSL product management. Users who need voice and fax, switched connections or only slightly higher speed will be drawn to ISDN.

"We've learned so much from ISDN. It really was the first midband access technology on the market," Gallegos says.

If the Internet is the killer application for ADSL, then who or what is the killer market? Consumers want high bandwidth, but are they willing to pay for it? Businesses need high bandwidth for data dumps and large files, but are they willing to shell out a lot of money for burgeoning technology? Some of the RHCs plan to go right after the business market, while others say they will concentrate initially on the consumer market.

BellSouth plans to target consumers first, then small office/home office users, says Brian Ford, manager of residential broadband data services. Businesses will be targeted later, as the carrier learns to differentiate classes of service and quality of service guarantees.

"In the last year, ADSL has moved from an interesting technology to essentially what we hope is a consumer market," says Ford. "The highest volume will be with consumer use. High volume drives prices and the market.

"We get requests for ADSL from 9- and 10-year-olds on our home page," says Flynn Nogueira, ADSL program manager at GTE. "This is their medium-and the need for speed just pervades our society.

The consumer market may not be ready yet, says Emily Green, an analyst with Forrester Research. "My current thinking is it's not really going to be a significant consumer product for two to three years.

The other DSLs It's a nightmare of acronyms-ADSL, RADSL, HDSL, SDSL, VDSL and now even IDSL. They all stand for different variations on the same concept of high data rates over copper.

U S West will begin by offering single-pair HDSL, but the other Bell companies are firmly focused on ADSL as the starting point for high-speed copper offerings.

"We see a wonderful opportunity with single-pair HDSL," says Joe Glynn, U S West director of Internetworking products. "The others are very focused on a technology solution of ADSL and [asynchronous transfer mode], whereas we're focused on just offering high-speed service.

Other carriers would probably disagree with the general sentiment, but they would agree at least in part with Glynn: They are focused on ADSL. Many of them, though, are moving toward rate-adaptive DSL, which is "smart" enough to negotiate for the best bandwidth.

When asked what is coming next, most carriers say very high rate DSL for certain applications. VDSL can give up to 20 or 30 Mb/s speeds over very short copper loops and is probably best suited for large office buildings with a fiber drop already in the basement. But that conversation doesn't last long.

ADSL is what carriers are pushing. "Five years from now, some people will still be using dial-up modems, some will be using ISDN and even more will be using ADSL [Figure 1]," notes Jeff Waldhuter, executive director of research and development for Nynex Science and Technology.

Still the skeptics wonder. "The phone companies have demonstrated in the past that they aren't fast about putting in new technology," Green says. "Sure they've been eager to test ADSL, but how fast will they actually offer service?" That's what some might call the million-dollar question.