ARTS & FARCES internet

Creativity requires free resources—including a commons—in order to flourish. The question is not whether government or the market should control these resources, but rather whether the resources should be controlled at all. So argues Lawrence Lessig in The Future of Ideas: The Fate of the Commons in a Connected World.

(Lessig’s publisher, Random House, certainly doesn’t have a clue: the downloadable electronic edition of The Future of Ideas actually costs more than the hardcover edition, even though the cost of production for the electronic edition approaches zero while the cost of production for the hardcover edition is not insignificant. But that’s a topic for another essay.)

Clearly, some resources must be controlled if they are to be sustained, but Lessig’s point is that the default condition should be free, with control imposed when necessary for sustainability.

Free resources must be kept widely available especially in the case of resources that are nonrivalrous. Lessig uses the example of Einstein’s theory of relativity to illustrate a nonrivalrous resource. I can use it as much as I like, and there’s still just as much of it as there ever was available for your use. My consumption does not rival yours in this case. Public streets, on the other hand, are a rivalrous resource. If everyone decided to use a road at the same time, then your use rivals mine and we have a mess of a traffic jam.

Lessig bases most of his argument on the three layers within a communications system proposed by Yochai Benkler and based on the layers of network architecture:

• The physical layer is the bottom layer of computers and wires that link nodes on the network.
• The code layer is the middle layer of protocols and software that makes the physical layer computers run.
• The content layer is the top layer composed of the messages and data that actually gets transmitted across the network.

Each of these layers, according to Lessig, can either be free or controlled and he uses the example of Speakers’ Corner in London’s Hyde Park as a communications system that is free at all three layers. Conversely, cable television is an example of a communications system that is controlled at all three layers.

The Internet is a special case. While the physical and content layers of the net are controlled, the code layer has traditionally been free. Unfortunately, Lessig says, large companies like Microsoft are trying to migrate the code layer of the Internet to their control.

Comparing companies like Microsoft to the former telecommunications monopolist AT&T, Lessig indicates that this fight for control of the code layer of the Internet should come as no surprise:

“The natural desire of any company is to find ways to protect its market. And the chosen desire of a competitive market is to limit the ways in which a company can protect its market—but for most of the century, this chosen desire was not telecommunications policy. For most of this century, in this context and others that we will consider later on, the chosen desire of policy makers was to back up the desire of companies to architect and support systems that protected them against competition in the market. Competition was a bother; the vision of a telecommunications system was limited; and our telecommunications architecture—including, as we will see, broadcasting and radio—was architected to maximize the power and control of the few.”

The architecture of the Internet, and its free code layer, resulted in a network that was free, the opposite of systems that came before. Instead of being designed to consolidate power and control in a few hands, the Internet was designed as an end-to-end network where the intelligence is at the ends, leaving the network itself to be simple. Lessig points out that this design has three important consequences for innovation:

1. Because applications run at the edges of the network, new applications can be run simply by connecting them to the network.
2. Because the architecture is not optimized for any existing application, the network remains open to innovation we can’t yet imagine.
3. Because the network owners can’t discriminate against different kinds of packets, the network can’t discriminate against any new application or innovation.

Of course, these benefits are dependent upon the way the Internet was originally designed. There’s nothing that says that the Internet of the future will be designed in the same way. Lessig uses Quality of Service (QoS) solutions as an example of how some companies are attempting to redesign the code layer of the Internet so as to put it under their control. QoS would enable the network to differentiate—and therefore discriminate against—certain “classes” of information. Lessig, surprisingly for a lawyer, points out the obvious: instead of differentiating “classes” of information, it would be much more appropriate to simply increase the capacity of the network. “The best response to scarcity may not be a system of control,” Lessig writes. “The best response may simply be to remove the scarcity.”

Lessig goes on to criticize the American system of intellectual property control. Copyright was intended to give an author exclusive control over his work so as to create an incentive to produce but not to weaken the public domain. That’s why the rights granted in copyright were for a “limited term,” why ideas can’t be copyrighted, why the work must “promote the progress of Science,” and why copyright can’t be used to deny fair use.

Copyright, as originally conceived, gave authors an initial term of 14 years that could be renewed for an additional 14 years if the author was still alive. Current copyright law grants the author a term of life plus 70 years. To put this in perspective, Lessig notes that in the first 100 years of the American Experiment, Congress extended the copyright term once. In the next 50 years, Congress extended the copyright term once more. “But in the last forty years, Congress has extended the term of copyright retrospectively eleven times.” The watermark for copyright term extension appears to be whenever the copyright term on Mickey Mouse is about to expire.

Lessig then goes off the tracks in blind support of the open source software movement. Some of his support is well-reasoned: “[The users of an open code project] are not hostages to bad code—the right to tinker is assured. And they are not hostages to strategic code—open code can’t behave strategically. These two features together constitute the innovation commons that the Internet creates.”

Some of Lessig’s support for open source misses the mark: “Just as end-to-end’s openness assures a neutral network that runs your code and doesn’t turn against your innovation, open code’s openness assures that the foundation of the computing environment is neutral and can’t turn against the innovator. An open code platform keeps a platform honest. And honest, neutral platforms build trust in developers.” In this case, Lessig misses more appropriate alternatives. Open protocols and file formats would easily accomplish the same goal while simultaneously allowing software developers to profit modestly from their innovations. More importantly, it would keep them innovating.

Maybe because it was at the end of the book, or maybe he was having an off day, but for whatever reasons, Lessig’s treatment of software copyright may cast doubt on all of his other arguments. Get this: “Worse, the copyright system protects software without getting any new knowledge in return. When the system protects Hemingway, we at least get to see how Hemingway writes. We get to learn about his style and the tricks he uses to make his work succeed. We can see this because it is the nature of creative writing that writing is public. There is no such thing as language that doesn’t simultaneously transmit its words.”

This is so wrong as to be laughable. First of all, when I purchase a license to a piece of software that is protected by copyright, I most surely get new knowledge. Arguably as much new knowledge as I get from the last book I purchased. Why else would I spend money on either software or books? More importantly, the notion that we get to see how Hemingway writes when we buy one of the books he authored is just wrong. We don’t see the process Hemingway employed in his writing, any more than we see the process the software author used. We see only the end-result of Hemingway’s collaboration with an editor and publisher; we see what the publisher wants us to see—the same as we see with software protected by copyright.

The solution is simple: the same five-year renewable terms for copyright are just as appropriate for software works as they are for literary works.

My grandma taught me to “give good weight” by underpromising and overdelivering. While open source software has an incredible amount of potential, it has—with very few exceptions—failed to live up to that potential let alone its hype. Open source proponents have, for the most part, overpromised and underdelivered. This is most likely because all of us like to get paid for our work. And the idea that there is somehow something wrong with that is preposterous.

This hypnotism by open source software notwithstanding, The Future of Ideas is an important work, well-researched and well-written. Highly recommended.