Outsmarting the Tragedy of the Commons
By Mark Frauenfelder, Mon Aug 05 00:00:00 GMT 2002

"The spectrum is finite, but as a resource it is not limited." So, what seems to be the problem?

In 1968, an ethics professor named Garrett Hardin wrote a paper called “The Tragedy of the Commons,” which was published in the journal Science. In his now-famous essay, Hardin defined “commons” as any kind of limited resource that people share, such as water, timber, land, or apples growing on trees. Hardin went on to argue that the establishment of an unregulated commons ultimately results in tragedy.

If there is a large tree full of apples, it is in a person’s best interest to pick as many of the apples as he or she can, and as quickly as possible, because other people are thinking exactly the same thing. If there are just a few people who have access to the tree, there’s a good chance that new apples will grow quickly enough to replace the ones that get harvested. But if a lot of people live next to the tree, it will soon be stripped bare.

This kind of logic was responsible for the downfall of countless villages in Europe hundreds of years ago. In those days, villages often had an adjoining pasture that was shared by all the residents. The villagers would let their cattle graze in the common pasture, and whether they realized it or not, they employed a kind of logic that led to the tragedy of the commons.

They thought: “If I let my cattle eat one square yard of grass, then the value of my estate will increase by the value of that unit of grass. However, all the other villagers share the cost of depleting the pasture of one square yard of grass. So each villager is losing just a fraction of a yard of grass. This is a great deal for me and not a bad deal for everyone else. Besides, if I try to conserve the grass, the other villagers will just give it to their cattle.”

With this kind of greedy thinking, it didn’t take long for the bucolic, verdant pastures to turn into worthless patches of scarred earth.

Today, resources once considered “commons” are no longer treated as a free-for-all. Either they are converted into privately held assets to be managed by the invisible hand of the market, or they are turned over to a governing body that allocates the resources as it sees fit.

The wireless spectrum is a good example of a resource that has gone through all three types of management.

When radio waves were first put to use, they were treated like a commons. There were no laws regulating the use of frequencies. In 1912, the federal government stepped in to oversee the allocation of the radio spectrum in the US. Even so, many radio stations competed with each other by operating on the same frequency. It wasn’t until 1927 that the government responded to what it described as a “breakdown of the law,” and passed the Radio Act, which granted the government the power to issue broadcast licenses. The FCC and other regulatory bodies around the world were established with the mission of divvying up slices of the spectrum in a manner that ostensibly yielded the greatest benefits for the public.

But in the 1990s the government took a different approach and began to auction off portions of the spectrum to the highest bidder. Of course, this meant that only the wealthiest companies could afford to buy the spectrum needed for radio and other wireless applications.

A Fourth Way

It’s an unfortunate situation, everyone agrees, because auctions tend to keep new companies without deep pockets from delivering innovative services. But what can be done about it? Plenty, says David P. Reed, a systems design consultant and a former vice president and chief scientist at Lotus Development. Reed also holds a PhD from MIT (where he also served as a professor) and participated in the development of the original Internet protocols. For the past decade, Reed has worked on mobile and personal wireless data communications systems. He believes we should stop thinking of radio waves as a limited resource. “The spectrum is finite,” he says, “but as a resource it is not limited because our technology can use it more effectively over time.”

Reed says the radio spectrum doesn’t necessarily have to be considered a limited resource, like a pasture or apple tree. In fact, he says that neither the physics nor the architecture of wireless communications lends itself to the property model. His argument is bolstered by results from a couple of recent research projects that were independently conducted at MIT and at University of Illinois. The findings from these projects led to a bizarrely counterintuitive conclusion – it is possible to create a network that actually increases in capacity as more people use it.

The way to achieve this seemingly magical state of unlimited spectrum, say the researchers, is by creating a new kind of network architecture. Today’s radio and mobile device networks send and receive radio signals from a centralized source, such as a mobile tower. In Reed’s world, a “cooperative wireless network” would take the control from the central tower and give it to the devices themselves. “Your cell phone would start acting like a small cell tower for all the other phones,” he says. “As more and more people use the network, the capacity actually increases.”

According to the research conducted by Tim Shepard at MIT, and Piyush Gupta and P. R. Kumar at University of Illinois Urbana-Champaign, the capacity of a cooperative wireless network increases when you giving each mobile phone or wireless device in the network the ability to cooperate with one another, forward traffic to other devices, and adjust its level of transmitting power. The results from their studies indicate that it is possible to increase capacity in proportion to the square root of the number of additional users. In other words, if the number of users goes from 100 to 200, the capacity goes up by 40%. That’s impressive, but even so, this increase in capacity isn’t sufficient to handle the new users, which in the example above, increased by 100%.

This increase, however, is just the beginning. Reed believes that cooperative wireless networks can be architected in such a way to make the increase in capacity directly proportional to the increase in the number of users. If that’s the case, then more devices can use the network without the need to allocate more spectrum to it.

One way to think about cooperative networks is to imagine a pasture where the cows eliminate edible grass instead of manure. You could keep adding cows without any of them going hungry. In the same way, each networked device you add to a cooperative network produces as much capacity as it consumes.

Reed compares cooperative networks to a highway system while he likens traditional wireless networks to a railway system. With a railway system, he explains, you need companies to build and operate tracks, and you need companies to decide what kind of vehicles run on them. The trains operate according to a schedule and don’t deviate from a predefined route. It’s all very centralized and hierarchal. Once a highway system is built, however, anyone can drive on them using any kind of vehicle they want. And they can take any route they want to get from point A to point B. Like a cooperative wireless network, a highway system is decentralized and distributed, with each drive making his or her own choices.

Roadblocks to Rollout

With so much going for them, so what is stopping the world from switching over to cooperative networks? Three major roadblocks, labeled “architecture,” “economy,” and “politics.”

The architectural and economic barriers are closely related. Radio and television networks have based their operations around a technological infrastructure that hasn’t changed much since the early 20th century, and it would be impossible to switch over to cooperative networks without upsetting telecommunications business models. They have billions invested in traditional networking architectures. To make use of wireless cooperative networks, everything has to be scrapped – Broadcasting towers, satellites, and all the other heavy-duty equipment involved. End users wouldn’t be able to use their existing mobile phones or car stereos, either. Because these mature network architectures would be incredibly costly to replace, you’ll probably see the first use of wireless cooperative networks in emerging areas where infrastructures haven’t been built out yet, such as networked devices.

Then there are the political issues to deal with. The unstated goal of any government bureaucracy is self-preservation. The FCC is a large, powerful bureaucracy, and even though its mission is to try to achieve a balance between its citizens’ right to free speech and the management of the wireless spectrum, it’s undoubtedly apprehensive about encouraging the deployment of a new technology that would end up putting it out of business.

In a way, the devices operating in a wireless cooperative network serve the role of a governing body. The devices organize themselves into a network that automatically allocates resources in the most efficient way possible. Who needs a centralized government agency to manage the spectrum when the devices can do it faster and more efficiently?

But the FCC is in a tight spot. Wireless spectrum is in increasingly short supply, and it is exploring how it can create a new spectrum policy approach. As I write this, Reed is in Washington, DC speaking at the FCC Spectrum Policy review about cooperative wireless networks. The fact that they are willing to let him speak to him is encouraging. Let’s hope they listen to what he has to say.

Mark Frauenfelder is a writer and illustrator from Los Angeles.