What's the Next Killer App on Faster Networks?

One of the recurring issues, when a new network launches, be that a next-generation mobile network or a 1-Gbps fiber-to-home network, is what new applications might develop that take unique advantage of the new platform. When 3G mobile networks first were launched, there was much talk about “new applications” the network would enable. As it turned out, there was precious little innovation.

About a decade ago, when the first commercial 3G networks were introduced, there was much talk about innovation and new applications the networks would enable, and the list looked remarkably similar to what people claim will happen with 4G. 3G history

E-commerce apps, for example, were thought to be an important 3G innovation. That is claimed for 4G as well, with more conviction, perhaps. “The availability of 3G services is going to have a profound effect on electronic commerce,” it was said. That also is said of 4G.

It was said that “3G works better” than 2G, and that was true. It also is said of 4G, and also is true. 3G wireless was sometimes characterized as a wireless version of the Internet, encompassing Web browsing, e-mail and media downloads. That sounds like 4G as well.

One might argue that few “really new” apps developed because of 3G. After a relatively long gestation, it turned out that mobile Web access finally became the “killer app” for 3G. But people also say Web apps will be the “killer” uses for 4G as well.

The point is that one might theorize that 4G evolution will resemble 3G evolution to some extent. 3G precedent To be sure, “more bandwidth” is better than “less bandwidth.” Whether those quantitative developments represent “innovation” or not is the question.

What we can say is that early on, important new technology tends to take the path of least resistance. In other words, a new technology tends to displace and cannibalize some important existing application, rather than immediately leaping to new applications which did not exist before.

The same sorts of issues surround fiber to the home and other faster networks as well. The hope and expectation is that, over time, radically new and important apps will develop. But in the near term, the apps that surface tend to be those which cannibalize some existing business, and simply shift revenue and activity from an older mode to a newer mode. 

You might argue that this is precisely what will happen as Google launches its 1-Gbps networks in Kansas City, Kan. and Kansas City, Mo. Instead of dramatically-new apps, the Google network will deliver entertainment video, voice and Internet access.

Looking at 4G, the value proposition mostly is “faster access.” That’s important, don’t get me wrong. But early on, the value will be provided by the same sorts of apps people can use on their faster at-home broadband networks, but in a mobile context.

In most cases, that will mean “video,” since the key issue is that video consumes about two orders of magnitude more bandwidth than any other application and has long holding times. In other words, where people talk for a couple of minutes per voice session, a video session can last 30 minutes to two hours. Video always is the issue

That has extraordinary implications for wireless networks.

To keep up with demand, U.S. wireless networks have traditionally doubled their capacity every 30 months, but this trend may not keep up with future demand, argues Michael Kleeman of the Global Information Industry Center at the University of California San Diego.

“We have reached a point of disconnect between the capacity of wireless networks and the emerging needs of today’s customers,” says Kleeman. Spectrum deficit

The volume of data traffic on U.S. networks is expected to increase by 1,800 percent over the next four years, the study estimates. By the end of 2011, video content will jump to 60 percent of network data volume.

According to another estimate, mobile video will more than double every year between 2010 and 2015 and account for two-thirds of the world’s mobile data traffic by 2015, the study suggests.

According to estimates, mobile data traffic in the U.S. was approximately 6 petabytes per month in 2008, 40 petabytes per month in 2010, and it is expected to reach 451 petabytes per month by 2013.

To understand the impact that even minor shifts in consumer behavior, and especially shifts in our consumption of video, could have for the U.S. wireless network, consider the contrast between U.S. video consumption and the capacity of our nation’s mobile networks.

U.S. viewers average nearly five hours of TV viewing per day (107,705 minutes per year) and as a nation we consume 1,266 exabytes of TV per year (1,266,000 petabytes).

Compare this to the output of U.S. mobile data networks, which transmitted approximately .48 exabytes (480 petabytes) in 2010 over the course of the entire year.

That means the U.S. wireless data network’s entire 2010 throughput was only sufficient to handle less than a day’s worth of the nation’s video consumption.

No matter what people hope will happen on 4G networks, the immediate implication will be much-greater consumption of video.