Fast enough for you?

Current calculations of bandwidth needs often focus on the long pole in the tent: high-definition television (HDTV) service, multiple simultaneous streams of which can raise total capacity requirements well above 50 Mb/s. Vermont's call for 3 Mb/s and 20 Mb/s reflects the state's uncertainty as to what extent video service should dictate the speeds of its broadband initiative. If the U.S. isn't lagging behind the Koreans in terms of TV, why should that be included in our broadband goals? “Today's video market is generally met through cable and satellite,” Murray said. “Tomorrow's video market is [video-on-demand].”

The aforementioned proposals share one notable common trait: a call for symmetric bandwidth, a controversial notion because consumers generally use far more downstream bandwidth than upstream. “What are you going to use the upstream path for?” asked John Celantano, president of Skyline Marketing. “You don't need 100% symmetry today or in the near future.”

According to projections made last year by Corning Cable Systems, video and gaming applications might require nearly 25 Mb/s of upstream bandwidth by 2012. Heavy users might require as much as 65 Mb/s upstream by then, but average users would need only 20 Mb/s, Corning predicted. What would users do with 100 Mb/s or 1 Gb/s upstream? Some say distance learning. Some say telemedicine or videoconferencing. But the catchall is the “X” app: The one that hasn't been invented yet, which users will nonetheless become addicted to once they have a Gigabit going both ways. That argument was easier to dismiss before YouTube-in the year following its launch, the Web site generated more traffic than the entire Internet did in 2000.

Part of the reason a number of disparate goals have surfaced, by many accounts, is the vacuum created by a lack of national broadband strategy. President George W. Bush's 2004 call for “universal, affordable” broadband was roundly criticized for being too vague, which may have encouraged others to avoid the same mistake.

Still, calls for blanket bandwidth, ubiquitous and symmetrical, may clash with the cultivated wisdom of the telecom industry. In an interview with Telephony last year, Carl Russo, CEO of broadband access vendor Calix, decried any quest for an abstract ideal broadband speed because real-world speeds are determined by a range of factors that vary by market-everything from a carrier's network plant quality and its local competition to the soil type in a given area.

“You hear all these arbitrary numbers,” Russo said. “27 Mb/s. No, 42 Mb/s. It's always going to be more over time. But the specific answer for any given market at any given time has many variables in it.”

To the extent that the bandwidth debate hinges on which specific speed to deploy everywhere, Russo said then, “It's the silliest debate I've ever heard.”