All Things Digital

An estimated 15 percent of Americans still use dial-up to connect to the Internet. And they might as well. Because according to a new study by the Communication Workers of America, the typical real-time Internet connection speed in the United States isn’t that much faster. CWA’s Speed Matters survey found the median download speed in the U.S. to be a mortifying 2.35 megabits per second.

Pathetic. In Japan the median download speed is 63.60Mbps. In South Korea it’s 49 mbps. For crying out loud, in Finland it’s 21.7Mbps.

How is it that the median download speed of the country that invented the Internet is this abysmal? No wonder it’s fallen to 15th place among industrialized nations in the percent of the population subscribing to broadband. No wonder broadband adoption slipped to a seven-year low in the second quarter of 2008.

The wheels have apparently come off of AOL’s advertising business–not that they were ever really on in the first place. On Wednesday, Time Warner reported a 26 percent decline in second-quarter income as the troubled Internet division continued to weigh on its performance. Revenue at AOL fell 16 percent in the quarter, while ad sales rose just two percent. In contrast, Google (GOOG), Yahoo (YHOO) and Microsoft (MSFT) all reported double-digit ad growth in the same period.

Numbers like that don’t speak well to AOL’s future as a successful advertising venture. And with losses in its access business mounting (604,000 subscribers in the second quarter alone), the division isn’t in the best of health. Little wonder, then, that Time Warner (TWX) announced today that it will indeed split AOL’s dial-up-access business from its advertising and content business. Beginning in 2009, the two divisions will be run independently. “We’ve now made key financial and strategic decisions that will enable us to operate the access and audience businesses separately,” Time Warner CEO Jeffrey Bewkes said on a conference call with investors this morning. “We have the necessary flexibility to do something strategic with either of these businesses today.”

Translation: Either could be sold or merged with another company. And, according to people familiar with the situation, Time Warner hopes to do one or the other with both divisions. The company has recently held informal discussions with Yahoo and Microsoft about AOL’s advertising and content business. And it’s said to be aggressively searching for a buyer for its access business.

Dial-up users don’t like broadband?

Obviously, that’s why they’re dial-up users.

An estimated 10 percent of Americans are surfing the net via dial-up connections, according to a report released Wednesday by the Pew Internet and American Life Project (PDF), most of them by choice. And 62 percent of dial-up users reported no interest whatsoever in upgrading to broadband.

Price was obviously an issue for some (about a third) and access an issue for others (24 percent), but 19 percent said that nothing can convince them to get broadband. Which means broadband growth in the states may be nearing a plateau. “… Solving the supply problem where there are availability gaps is only going to go so far,” said John Horrigan, the study’s author. “It’s going to have to be a process of getting people more engaged with information technology and demonstrating to people that it’s worth it for them to make the investment of time and money.”

And until then, the percentage of adult Americans with home broadband connections will continue to hover around 55 percent.

In 2010, it could take as long as two minutes to download an episode of “Chad Vader–Day Shift Manager” from YouTube, instead of the few seconds it takes today. This according to a new study from Nemertes Research Group, which claims that the Internet could be approaching its capacity. “Our findings indicate that core fiber and switching/routing resources will scale nicely to support virtually any conceivable user demand,” Nemertes explains in “The Internet Singularity, Delayed: Why Limits in Internet Capacity Will Stifle Innovation on the Web.” “But Internet access infrastructure, specifically in North America, will cease to be adequate for supporting demand within the next three to five years.”

And what does that mean in lay terms? “Users will experience a slow, subtle degradation, so it’s back to the bad old days of dial-up,” said Nemertes President Johna Till Johnson. “The cool stuff that you’ll want to do will be such a pain in the rear that you won’t do it.”

To avoid such a scenario, Nemertes says backbone providers need to invest up to $137 billion in Internet infrastructure capacity–more than double what they’d planned. If they fail to do so, we may see that slow degradation to which Johnson referred and a stifling of innovation. “It’s important to stress that failing to make that investment will not cause the Internet to collapse,” Nemertes explains in its paper. “Instead, the primary impact of the lack of investment will be to throttle innovation–both the technical innovation that leads to increasingly newer and better applications, and the business innovation that relies on those technical innovations and applications to generate value. The next Google, YouTube or Amazon might not arise, not because of a lack of demand, but due to an inability to fulfill that demand. Rather like osteoporosis, the underinvestment in infrastructure will painlessly and invisibly leach competitiveness out of the economy.”

Nemertes’s last point about underinvestment in infrastructure is one worth noting. Because in the run-up to the Telecommunications Act of 1996 the incumbent telecoms promised to provide fiber-optic connections to millions of households across the country. In exchange, they were given some $200 billion in tax cuts and higher service rates to pay for it. But the telecoms didn’t spend that money on fiber upgrades–they spent it on long distance, wireless and inferior DSL services. “By 2005, if the Bell companies had actually delivered on their broadband promises, approximately 86 million households would have had fiber-optic-based services,” Bruce Kushnick, executive director of New Networks Institute, explains in “The $200 Billion Broadband Scandal.” “These state commitments also would have rewired schools and libraries, hospitals and government offices. And in most states, the plan called for ALL customers to be rewired equally, whether they were in rural or urban areas, rich or poor. Universal broadband was to be accomplished state-by-state because customers were, in essence, de facto investors funding these network upgrades.”

Something to think about when the Nemertes’s study begins popping up in telecom arguments against Net neutrality, as it almost certainly will.