Net neutrality for wireless? Don't count on it.August 11, 2010: 10:46 AM ET
Unrestricted access rules for wireless networks would hurt users more than help them. They just don't realize it.
Earlier this week, Google and Verizon brokered a compromise on the definition -- or at least, their definition -- of net neutrality, a set of rules that ideally, would ensure that no company could place data-access restrictions on Web content, sites, platforms, and associated equipment. The deal itself sparked controversy over whose interests Google (GOOG) and Verizon (VZ) really had at heart and whether the deal would -- or more importantly, should -- be used as a model by the Federal Communications Commission.
In the discussions, there was one glaring carve-out: wireless traffic, which was not covered by the agreement. Detractors have argued that not including wireless in the companies' net neutrality proposal is a sign of corporate greed at work, a clear move by the carriers to prevent one pipe from turning "dumb."
But the reality has much less to do with profit-chasing than with physical plant. While consumers are increasingly agnostic about how they get their Internet -- whether over wires or cellular, they just want their Facebook -- the folks supplying the Internet see two entirely different worlds, one where adding capacity is a breeze, another where it's anything but. Net neutrality would be a serious problem for wireless networks, who all-but-have to prioritize certain types of data-hungry types like say, point-to-point streaming media, over others due to simultaneous usage and current bandwidth limitations. That makes the idea of net neutrality over the air unlikely to see the light of day for at least the next five years.
Here's how the current system works. Wireless networks are built assuming people will get on and get off quickly. Like with cable networks, anyone accessing a cell tower is using spectrum that someone else in the area can't. But unlike cable, which is built with high-capacity fiber, a wireless network is built on the theory that just 10% of subscribers will be using the spectrum at the same time.
That was fine in the voice days; not so much when high-pixel phone screens are sirens crying out for streaming Netflix (NFLX). If wireless carriers using today's technology wanted to mimic the capacity of DSL or cable, they'd have to triple the number of towers for each subscriber. Upgrading the network is neither simple nor quick. While it only takes a minute for one anime-over-Android addict in a neighborhood to bring down everyone's cell service, it takes three years in some cities for a new cell tower to get approved. The carriers want to be able to control their traffic; customers should, too.
"Someone gave a very good example," says Bernstein Research analyst Craig Moffett. "'My mother just got a pacemaker that will wirelessly contact the hospital if she suffers from cardiac arrhythmia. Are you telling me it would be illegal to prioritize that traffic over a video of a squirrel on waterskis?'"
Eventually, the carriers will have to make improvements -- if they have any chance of staying competitive. But it won't be cheap and they'll fight any data-increasing net neutrality mandate with every lobbying dollar they have.
It's easy to see why. Recently, in a report it called "The Wireless Data Exaflood," Bernstein noted that wireless data traffic has grown about 120% a year over the last two years while revenue per subscriber has stayed flat. That means less cash to pay for the overloaded towers and clogged backhaul brought by all of that smartphone traffic. Bernstein estimates that carriers will have to hike their spending on their networks by 19% this year, another 5% in 2011 and then just slightly less per year after that.
Attempting to slow this flood is the very reason AT&T CEO Randall Stephenson and Verizon Wireless Senior Vice President and CTO Anthony Melone said unlimited data plans must go and why AT&T (T) transitioned to plans with 2 GB or 5 GB caps. Consumers will take it as a serious blow, but in the end, the switch might very well be in their best interest. Given that, data prioritization seems not only desired, but necessary -- for now -- to keep service quality at even its currently lousy state.