Is this finally the year of DSL?

It seems like years since we first heard about how various flavors of digital subscriber line technology were going to revolutionize Internet access. Actually, it has been years. But now, the recent deal between America Online and three major regional Bell operating companies indicates that DSL finally may be poised to come into its own.

Earlier this year, the Federal Communications Commission closed some of the loopholes the RBOCs had been using to delay competitive local exchange carrier deployment of DSL. By the end of the year, the FCC is expected to introduce some new DSL deployment rules for the RBOCs. The combination will eliminate most of the regulatory uncertainty that has suppressed DSL. And because cable modems, the archrivals of DSL, have been gaining acceptance, it's likely that all types of potential DSL providers will move quickly to recover lost ground.

The recently announced pact between AOL and RBOCs Ameritech, Bell Atlantic and SBC is surely a reaction to the recent regulatory trends. However, it also raises a question about DSL: the technology may be wonderful for consumers, but can DSL pay its own way in the marketplace?

Three years ago, AOL was the poor stepchild of the Internet, the company everybody said was going to be pushed aside by the new wave of open Web clients and the explosion of new ISPs. Now AOL is about the only ISP that can be called successful, and the closest thing the Internet has to a blue chip stock. Why? Because AOL can draw on advertising revenue and retail sales.

More ISPs have been the victims, rather than the beneficiaries, of the Internet's success. Their traffic grows daily as more people visit those neat e-commerce and auction sites, but their revenues don't grow much at all. AOL is the exception. That's why AOL can step forward to push the market toward realizing the goal of always-on, high-speed Internet access through DSL.

Most people want to believe DSL will eventually sell for about what premium Internet service goes for today, perhaps $30 per month. Logic says that if ISPs aren't reaping a profit selling 28.8K bit/sec services for $30, they probably won't find selling DSL at megabits per month for the same price very profitable, either. However, an always-on digital pipe to premium consumers ought to be exploitable as a marketing channel: a set of digital strings to make the consumer dance like a puppet to the merchandisers' tunes. To exploit this pipe, ISPs must be able to control their customers' graphical user interfaces to the point where they can be used to present retail offers and ads. AOL can do just that.

This isn't to say that the only road to DSL reality is paved with a zillion unwanted ads. What AOL is teaching us is that when a network service is offered at a fixed price per month, something must be added to it to generate additional revenue to offset increased usage. Traffic costs money to transport, no matter what the free-bandwidth advocates say.

DSL providers will be searching for DSL revenue sources, to be sure. Bell Atlantic, for example, has also launched a menu of DSL service offerings targeted at small businesses. DSL will likely be the focus of some of the application hosting offerings, given the fact that small businesses are somewhat more likely to be consumers of application hosting than their giant-corporation partners.

AOL's partners are speculating that total DSL penetration could exceed 30 million households within two years, and AOL says its partners could potentially cover about 55 million households. So will DSL explode, as providers seem to be promising? Prob-ably not in the near term. The press releases on the AOL deals often contain the qualification that they include "forward-looking statements." In other words, they're hype.

Nevertheless, it's clear that DSL is now gaining momentum, and that at least some Internet users can expect DSL-based access at 768K bit/sec for only about $20 more than standard modem access. What remains to be seen is how much of that 768K bit/sec will really be available. An admittedly small and unscientific poll I conducted of cable modem users in mature service areas shows an effective Internet rate of only 72K bit/sec.

Will DSL do better, and if not, will users value small performance gains (and the benefits of always-on Internet) enough to justify even the modest cost increase?

What's important now is that it looks like we'll finally get enough DSL experience to judge its real potential.