When a small Illinois IT research firm published a study on the future of the Internet last week, it didn't expect to create an international furor. "
"I had no idea it would get spun this way, twisted this way," report co-author Johna Till Johnson, president and a senior founding partner of Nemertes Research, said Wednesday.
"I've read all sorts of interesting stuff that bears little relation to the truth, but people seem to be basing it on the study."
All the study concluded, she says, is that a mismatch between demand and access capacity will be reached in three to five years that will have to be met by billions of dollars in spending by carriers. Otherwise, the next YouTube may be throttled because the Internet will be hard to access.
But the reaction was headlines like "Internet Facing Meltdown," and "Internet Blackouts Predicted by 2010."
Which baffles Johnson, who notes that the entire report was available for reporters to read and accurately quote.
"We explicitly are not saying the Internet's going to break," she says.
In hindsight, she adds, the firm should have foreseen the reaction from Internet lobby groups who she says put their own negative spin on the report. "They really failed to see that it's entirely straightforward to build their case [for supporting the Internet] around the findings, which were intentionally policy-neutral."
"It surprised me there was this bipolar response that had nothing to do with the findings."
Nemertes, which advises clients on the business value of IT technologies, began the research to find out "exactly what was going on with the Internet," in Johnson's words, because studies on the use of the Web had become out of date.
Rather than make a projection on the rate of growth of the use of applications such as e-mail, it created a model based on bandwidth consumption and compared it to the infrastructure spending academic and industry research organizations predicted are coming.
The work was supplemented by interviews with more than 70 enterprises, vendors, service providers and investment companies.
The conclusion is that "Internet access infrastructure, specifically in North America, will likely cease to be adequate for supporting demand within the next three to five years."
It estimates access providers will have to spend between US$42 billion and US$55 billion to close that gap, which could be 70 per cent more than they plan to invest.
"It's important to stress that failing to make that investment will not cause the Internet to collapse," the report's executive summary adds. Instead, difficulty accessing the Web will "throttle innovation" by companies trying to create the next YouTube or Google.
While confident in its conclusions, Nemertes also acknowledged an "overwhelming conviction" that the industry needs more and better data.
One reason why North America will see this crunch more than other parts of the world is the aging infrastructure here, Johnson said. Another is that other parts of the world are more willing to invest in broadband wireless access, which will help ease the problem. Some Canadian industry observers are skeptical. While refusing to comment on the work of a competitor, Alex Pares, network equipment research analyst with IDC Canada, said an access problem is "not going to happen in the short term."
"I believe the customers are demanding more of the network, but they are demanding more of the ISPs as well," he said. Meanwhile providers have tools such as quality of service and MPLS to guarantee sensitive traffic -- including voice and video -- will get through.
"Telus recognizes that it's not just investment in the access networks that's required to maintain the robustness and scale of the Internet," says Chris Langdon, the Vancouver-based utility's vice-president of network services. The core also needs to scale. "Otherwise," he said, "you have big pipes going into skinny funnels."
So last year Telus announced US$600 million in spending over three years to deploy next-generation fixed network services, and this year is testing Gigabit Passive Optical Network technology.
Supply and demand "always have a way of leveling and finding the happy medium," he said.
He couldn't speak for other providers but said "we're making the requisite investments to make sure there's no mismatch."