• by Steve Smith , Staff Writer @popeyesm, September 5, 2008

"We don't have anywhere near the backbone bandwidth we need in this country," says veteran analyst Stuart Zipper, senior editor of Broadband Business Report. "In two to three years, we will see some kind of crunch."

First and foremost, a digital media age rests on reliable, abundant distribution mechanisms. Eerily paralleling our uncertainty about oil supplies, demand for wired and wireless bandwidth will grow exponentially in the coming decade as massive new segments of the world's digital ecosystem increase their appetite for gigabyte-guzzling features like video, data backup, applications, "telemedicine" and more. Are the pipes fat enough to fit it all?

As in the oil debates, the various stakeholders in the data-value pipeline have their own self-serving answers. Cisco estimates IP traffic will double every two years through 2012. ISPS just in the U.S. already struggle with the YouTube effect, massive bumps in video streaming that increase per-subscriber bandwidth use 1.5 times every year. "We are already seeing choking in many peering points on a daily basis, and this will only get worse," says Tom Leighton, cofounder and chief scientist of Akamai, which distributes content to the edges of the network for more efficient delivery. Who is going to pay for the necessary build-out? The heaviest users with tiered pricing, or the Googles and Hulus that peddle the new media crack?

Everyone agrees that the Internet is scheduled for a major protocol upgrade as the old IPv4 system runs out of available IP addresses by 2012 and requires a slow, costly shift to IPv6. Otherwise, "there is a risk of not being able to get online," Web pioneer Vin Cerf told the BBC last year as he stepped down from Internet oversight group ICANN. The increasing availability of fiber worldwide will help ease the pain, and press cable ISPS to adopt the DOCSIS 3 standard, which many expect to hike data rates substantially by 2011. Ultimately, probably via DOCSIS 3 cable or fiber, most of us will see household bandwidth hit the 50MB/s speeds more common in Japan and Korea. But with a Net Neutrality debate raging over who pays for it all, no one knows. "This is an important issue with long-term economic implications for all the parties involved," says Leighton. "It will not go away any time soon."

On the wireless side, ambitions are even greater for squeezing mobile TV, 4G services like WiMax broadband, and maybe even HDTV into an undeniably finite radio spectrum. According to the FCC, 70 percent of new broadband subscriptions in recent years were wireless, and John Walls, vice president of public affairs for the CTIA wireless association, recently said flat-out that "the need for more spectrum is real, plain, and clear." But where does it come from, thin air? Well, yes. Like wired bandwidth, wireless spectrum will spend much of the next decade mired in policy debate. As the FCC auctions off more spectrum no longer needed by analog devices like TV, how much should government continue to restrict spectrum uses? "The biggest challenge we are facing right now is a philosophical one," says Patrick S. Ryan, researcher at University of Colorado at Boulder, Interdisciplinary Telecommunications Program. "Do we believe that devices and companies that generate them are able to function in a world that has rules but has less oversight, or do we still have to rely on a system of spectrum allocations?"

The recent auctioning off of 700MHz spectrum requires owners to open the bands up to multiple vendors and uses, an unprecedented move instigated by Google. Digital devices will be able to filter noise and share space with one another more effectively than their analog grandfathers. Slowly but surely, most expect the FCC to release more spectrum with more open licenses and to find ways to repurpose the "white spaces" of underutilized spectrum. Letting the market and smarter devices determine its uses is the overall trend. "The principle of the future is going to be based on an open platform that anyone can adopt and can be integrated on the device level," says Ryan.

Clearwire is a deeply funded consortium of Sprint and multiple cable, tech and online entities, including Google. Some observers think that company's WiMax wireless broadband would affect the landline ISP model as well as mobile, but the technology has suffered multiple false starts.

But wireless speed is not the same thing as capacity. Even WiMax's promised high speeds (up to 100MB/s) start disappointing users in a spectrum where multiple people share and degrade performance. JupiterResearch analyst Ian Fogg warns, "It's good for snacking, but not for high-def, two-hour movies ... unless no one else is on the network." And at this late hour, WiMax must compete with a growing field of unratified wireless protocols: 802.20, or "MobileFi," and 4G Long Term Evolution (LTE) among them.

"Everything is in place for a battle," adds Zipper. The prospects for wired and wireless bandwidth in the coming decade seem filled with noise.