DSL fever.

At Supercomm each year, the best and brightest vendors and telecom providers unveil their finest products. However, this year's bumper crop of the best access products looked and sounded strangely familiar from booth to booth.

"We're getting into digital subscriber line," several proclaimed. "It cuts down on traffic on the public network," others echoed. "It's much simpler to provision than ISDN and cheaper than T-1 lines," even more repeated. "Our solution is a unique one," vendors said again and again.

Analysts who attended the show estimated that anywhere from 40 to 80 companies showed, demonstrated, talked up or promised DSL products. And they're not fly-by-night, over-caffeinated engineers working out of a garage. These are some of the big boys - Lucent Technologies, Northern Telecom, Fujitsu, Siemens and Alcatel, as well as start-ups by alumni from these and other telecom companies.

And at this point, those 40 to 80 players are still about 40 to 80 more companies than this nonexistent market can support.

It looks like DSL fever has infected the telecom industry. There are many strains - asymmetrical (ADSL), one of the most common; high-bit-rate and symmetrical, two related strains; rate adaptive, which some confuse with symptoms of ADSL; very high-bit-rate (VDSL), a brand new bug; an ISDN bit rate version; and even a new medium rate, for those who haven't been stricken with enough of the others. The varieties mainly differ in bit rates and copper loop distances.

Nowhere yet this year was the DSL condition so evident as on the show floor at Supercomm '97. The buzz three weeks earlier at Networld+Interop in Las Vegas turned out to be only a prelude. The competitors seem to be driven mostly by the apparent interest in DSL on the part of the Bell regional holding companies and GTE over the past year.

Of course, Internet service providers also are getting into the game, but except for a few large ones, most have the money to deploy only a small number of lines. Competitive local exchange carriers are among the early adopters and will probably grow a nice DSL data business for themselves if and when the market takes off. And of course, the big interexchange carriers want to offer DSL, but they're still waiting for widespread access to the local loop and probably will be tied up in court for some time.

So for now, the ones with the most money to spend appear to be the RHCs that are running DSL trials or have declared their commitment to DSL, and even some that have promised millions of lines to businesses and homes in the next five years.

Aiming to please Suitors are lining up to please the deep-pocketed and pro-DSL RHCs. RHCs love asynchronous transfer mode, so the vendors say, "Let's give them ATM over ADSL." That sentiment was doubly reinforced by the Joint Procurement Consortium of Ameritech, BellSouth, Pacific Telesis and SBC Communications, which specifically requested an ATM over DSL solution.

Still, others are determined to run the Internet protocol (IP) or point-to-point protocol (PPP), which encapsulates IP. So at Supercomm, the buzz among those companies was whether to run IP or PPP over ATM over ADSL. The situation is similar to the way the crowd splits up for the debate of ATM switching vs. high-end IP routing - except this time most of the Netheads and Bellheads have agreed to use ATM, whether network traffic managers or end users can see it or not.

Although many within the industry have become suspicious of this kind of maneuvering and hype, that does not automatically mean the market will fizzle out or become insignificant. While the initial excitement and eventual disappointment of technologies like switched multimegabit data service and interactive video should not be forgotten, DSL does deserve a chance to prove itself.

But that will be a product of time, not a product of hype and artificial excitement. "We're going to see a dramatic shakeout in the DSL industry in the next year." said Kieran Taylor, broadband consultant with TeleChoice. "Some will go out of business, some will partner with larger companies and some won't make it at all.

"Let's say there are 60 companies out there now. Less than half will be left by the close of 1998 - either from acquisition, merger or bankruptcy," he said.

Bobbi Murphy, analyst with Dataquest in San Jose, gave a DSL round-up presentation at Supercomm in which she answered the question, "Who is ready for DSL right now?" Her answer did not include the typical suspects of ISPs, RHCs and consumers usually mentioned in vendors' marketing strategies.

Most ISPs don't have the money, don't want to be in the infrastructure business and don't feel comfortable with collocation, she said. RHCs are worried about cannibalizing their own T-1 markets, the high expense and another ISDN fiasco, so they're playing tests and deployments conservatively. Even most consumers aren't ready for DSL, Murphy maintained. They don't want to pay more than $19.95 a month for unlimited Internet access and don't yet see the value in super-high megabit speeds.

So who is ready? Murphy said it's a specific niche of certain CLECs and ISPs going after the T-1 and ISDN market with low-speed services from 128 kb/s up to 2 Mb/s, IP data services, and low- to medium-density DSL access multiplexers with other services mixed in.

Who will succeed? If DSL begins to take off as predicted in 1998, that brings us to the question of which particular vendors will survive. First, remember that the DSL equipment business has several different components. All DSL systems require a piece of customer premises equipment, which could be a modem or network interface card, and a piece of equipment for the central office or point of presence (POP).

Many of the CO devices are referred to generically as DSL access multiplexers. However, companies like Pulsecom and DSC Communications have designed or are working on equipment that fits into a digital loop carrier POP or cabinet using existing channel banks there.

That said, there are viable players in both areas. According to TeleChoice's Taylor, traditional telco equipment vendors tend to pick up the CO equipment side, while traditional modem vendors gravitate to the customer premises side. Remote access companies fit in either with CO or CPE or both.

In the telco space, Nortel, Lucent and AG Communications Systems are ones to watch, Taylor said. In the remote access server companies, he chose Ascend Communications, Netspeed, Copper Mountain, Tut Systems and Paradyne. In the customer premises end, Taylor named U.S. Robotics, Hayes, Westell Technologies and Amati.

Murphy also named some DSL companies she believes are ones to watch, and not surprisingly, many overlap with Taylor's. She chose AG Communications for its strong telephony experience and broad product offerings; Ascend Communications for its excellent channels and dial and frame access market domination; Copper Mountain for offering the first low-cost single-pair or symmetrical high-bit-rate DSL system and having good connections; Paradyne for its strong access product line and broad DSL offerings; and Orckit for its broad offerings, VDSL experience and good European positioning.

Of course, these choices don't mean that the other DSL vendors - and new ones no one has heard of yet - are closed out of the game. "Sure, there may be 60 companies right now, and there's probably another 30 we haven't even heard about yet," Murphy said. Only time and the market will determine which succeed.

"Just because you're just starting to put out product now doesn't mean you're DOA," Taylor said, "It's the companies with no strategic partnerships or who aren't doing anything to differentiate themselves that are the ones most likely to run into trouble.

What happens next? Until the DSL market takes off - assuming it will - the next months could be largely a waiting game during which partnerships will probably continue to be formed, new products will be announced and maybe even the first DSL company closures or buyouts will occur.

"Rate of deployment is definitely a mitigating factor. Who's going to be able to last this winter when contracts are few and far between?" Taylor said.

It's worth mentioning what several of the less hype-oriented vendors and some analysts like to remind those prone to being swept away: DSL is only an enabler. It is simply a transmission technology, and it can't solve all network problems. Even if the last mile to the home or business is revved up beyond our current imaginings, the home or business still has to be billed, the system still has to be managed within the service provider's network, and it has to be tested, maintained and easily rolled out at a low enough cost to be palpable.

As one traditional telco vendor executive put it, "Getting a DSL product out is not a big deal; almost anyone can do it. The big deal is in supporting the product afterwards when it's being used."