Lawrence Lessig on the future of patents (or lack thereof):

Who Should Own What?, by Todd Datz.

Also of particular interest is this mention of the MP3 patent, which media player developers will need to keep on radar.

That's the one that's most obvious and direct. Another threat is patents that we've been seeing recently, from the absurd British Telecom patent on hyperlinking to the way in which the MP3 patent is now being deployed against people who build players or record MP3s and to the way patents have been used in standards-making processes. You have these groups that get together to build a standard that other people can use and adopt. Secretly, one of the participants in the standards-making process early on files a patent for the basic idea. Nobody knows the patent has been filed because you don't have to reveal that information. Once the standard is out there and adopted, the company comes forward and says, "I have a patent on that standard and you've got to pay me to use it."

Here's the complete text of the article in case the link goes bad:

http://www.darwinmag.com/read/080102/lessig.html

LAWRENCE LESSIG knows how to stir things up. A professor at the Stanford Law School, he is also the founder of the Stanford Center for Internet and Society. His latest book, The Future of Ideas: The Fate of the Commons in a Connected World, published in 2001, details Lessig's view that commercial forces are threatening to close off the once-free and open Internet by attempting to control the software code and content. Such ideas have stirred up animosity among patent lawyers.

Todd Datz of Darwinmag.com spoke with Lessig about patents and the Internet earlier this year.

DARWIN: If the purpose of patents is to spur innovation that would otherwise not occur, how well does the present system serve that goal?

Lawrence Lessig: My objection has been that we have launched into this fairly massive regulation of the innovation process without any [clear evidence] that it will do good. As I say in my book, where are the Republicans when you need them? Did anyone do any impact statements to show that this kind of regulation for a crucial part of our economy was needed or would actually be productive?

Companies are now patenting everything under the sun. In fact, companies that don't keep up with the competition are at a disadvantage.

It's impossible to imagine people sacrificing themselves in order to create a world without patents. It is a rational thing in this world to patent everything under the sun—especially because it's not just legitimate competitors in the innovative process who will come and threaten you down the road. It's also "patent thugs" who seek any way they can to hold up legitimate innovators through the use of these patent claims. Given their ability to get an injunction against the release of a product or muck up an IPO, that's quite a significant cost.

I think there's got be real legislative reform. I've argued that during this period we ought to permit people to patent what they want but forbid the offensive use of patents in the productive process. People can build up their claims of what property they've got.

A second kind of change would be just to lower the stakes by reducing the term of certain kinds of patents, such as software and business method patents, as Jeff Bezos [CEO of Amazon.com] has argued. I can't remember whether it's two or three years, but it's a radically shorter term, which I think would be fine. Another way to lower the stakes would be to remove the use of injunctions as a tool for enforcing the patent. If you've got a patent that covers some sort of technology, especially software, and you want to enforce that right, then you can seek damages for the violation of the patent, but not resort to this nuclear weapon of an injunction to stop the production process from going forward.

Do there need to be changes made in the patent office?

Though I'm a critic of the patent system, I'm not a critic of the patent office. I'm not saying I think they do a great job. I think, in fact, they do a bad job. But they do a bad job because they're wildly undersupported, underfunded. If we're going to manage this kind of Soviet system of picking and choosing which innovations are going to be allowed in this system, at least the Soviet system ought to have the resources to do it correctly. And patent examiners shouldn't be under extraordinary pressure to process these applications in an extremely quick way, which inevitably is going to lead to bad patents being issued.

Can you talk a little bit about software and business method patents and why you think they're out of control?

First, all of this patenting behavior took off after it was encouraged by the Federal Circuit court. At that time [1998], all the money in the world was in software and dotcoms. There was great incentive to find a tool to buy or bully your way into that market. That tool was the patent. All sorts of money was spent on patents in order to tell dotcom billionaires, "your business is toast unless you pay us money."

Second, whenever patents enter a new field (one that most people didn't think was open to patent behavior), there's obviously going to be a land grab. There's lots of stuff to claim, especially when that field is undergoing significant innovation. It's a great opportunity for taking ideas and beginning to express them in patents. Furthermore, the traditional inhibitor on the issuance of patents is the patent office's reliance on "prior art" [the history of innovation in a particular field]. Prior art for software and business method patents is almost nonexistent in the patent office. They didn't spend the last 30 years gathering the data they would need in order to establish something as prior art. They don't even know when something is not sufficiently novel to justify a patent.

This is another stupidity of the system: The legal system does not impose a strong enough requirement on inventors to come forward with prior art. Basically, the rule is you've got to reveal the prior art you know of. But that, of course, creates an incentive for the ostrich effect—put your head in the sand and not make yourself aware of the [prior art] other stuff. That puts all the burden on the patent office. It's a fairly simple legal change to say you've got to come forward with all the prior art that anybody familiar with this field would know of. If inventors had that obligation, and if patents were invalidated on the proof that they didn't come forward with prior art that should have been revealed, then that would ensure that the patent office knew how to do its job. Those are the primary reasons why the system should be afraid it's doing more harm than good.

What is the argument used by those companies that say strong intellectual property rights are necessary on the Internet?

The arguments I've heard are about the need to defend their significant investment for a long period of time before it gets taken by others. It's not an argument well-tuned in the context of the Internet. All of the creative work in the context of software and its implementation to the Internet is how the idea is implemented, not really the idea itself. However, given the way patents function now, they become the patents of ideas and that creates a huge ability to block subsequent deployment of competing systems.

The standard response of any capitalist is to say, "Give me a monopoly or else I won't be able to compete." A standard response of wise government would be to say, "We don't give capitalists monopolies unless they're absolutely necessary, so we'd rather have you compete like hell than give you the power to use the government to stop people from competing against you."

Do you think there will continue to be more disruptors like Napster that sprout up from the cracks, or has the fear of being brought immediately into court served to quash that kind of innovation?

It's clearly having an affect on innovation. Five years ago, when the law was still unsettled, venture capitalists were willing to gather lots of money to fund these innovations. They didn't expect that the first thing that would happen when a company opened its door was a lawsuit from a competitor—primarily from the copyright industry, which was trying to use its power over copyrights to kill you.

The reality now is that every new innovation has got to not only fund a development cycle and fund a marketing cycle, it's got to fund a legal cycle during which you go into court and demonstrate that your new technology should be allowed in the innovative system. In that context, there's an extraordinarily high burden on innovation because the legal system is extremely poor. It's costly and it's inefficient in that it doesn't often produce the right results. It imposes a huge risk on the development process, which translates into a much stifled level of investment by venture capitalists.

What are some of the most pressing threats to the Internet right now?

The stifling of broadband deployment is one extraordinarily important problem to solve. Unfortunately, in my view, it's not going to be solved, but, rather, is going to be exacerbated by the deregulation of the telecom infrastructure. The problem in getting broadband out there and accepted, in part, is caused by overly restrictive IP [intellectual property] laws. As even Michael Powell of the SEC has said, there are 10 million channels but still nothing on. That's because people aren't making content available. They're afraid of piracy, but also they're afraid of competition to their existing business models. Giving them this amount of power over IP contributes to the stifling of broadband deployment.

That's the one that's most obvious and direct. Another threat is patents that we've been seeing recently, from the absurd British Telecom patent on hyperlinking to the way in which the MP3 patent is now being deployed against people who build players or record MP3s and to the way patents have been used in standards-making processes. You have these groups that get together to build a standard that other people can use and adopt. Secretly, one of the participants in the standards-making process early on files a patent for the basic idea. Nobody knows the patent has been filed because you don't have to reveal that information. Once the standard is out there and adopted, the company comes forward and says, "I have a patent on that standard and you've got to pay me to use it."

Has that actually happened?

This is a particular complaint Vint Cerf was describing at a conference in Boulder, Colorado, about two weeks ago. He was describing how patents in the standards-making process are really inhibiting the opportunity to establish uniform and open standards. I see the beginning of discussion of this in "RDF," a standard to enable a description of data to be used for all sorts of Web services. The opportunity for people to begin to assert and threaten patents in this context will significantly chill innovation.

After reading your book, I'm scared to death of the consequences of companies controlling the Internet, as well as the general stifling effect of the surge in patenting on innovation; scared of this notion that "only property matters." Should I be? You even quote Marc Andreessen saying that the innovation age "is over."

I spoke to him late in the process of writing the book. I was both reassured and deeply depressed that essentially everything I was arguing in the book he summarized in a couple of sentences.

Look at the history of innovation in the context of the Internet. The World Wide Web was built not by AOL or Prodigy, but by a researcher in Switzerland. Hotmail was developed by an Indian immigrant. ICQ, which was the beginning of the really persistent instant messaging system, was developed by an Israeli—or rather stolen by an Israeli from his son—and deployed outside the United States and sold to AOL for $400 million.

All of these innovations were enabled by people outside the dominant, powerful industries at the time. They didn't need the permission of those industries to develop and deploy their innovations. That was what Andreessen was talking about. He said there was this window of time when you could become a desktop software maker because of the birth of the Internet. Before the Internet, everybody thought that the industry was dead because they thought software was in the control of Microsoft and other large companies. But the Internet radically transformed that. There was a period of time when an extraordinary explosion of innovation occurred. And once the explosion occurred and people came in to try and change the architecture of the environment, Andreessen says the opportunity for innovation disappeared.

You say that the rules governing the freedom to innovate are best characterized as a constitutional question. Explain what you mean.

The core argument of my book—an argument I've completely failed to get people to see—is that architecture is the politics of the Internet. The architecture of the Internet sets the values and possibilities for the Internet. In that sense, it functions like a constitution functions in a legal system. The Internet's architecture forms the background terms in which people experience opportunity on the Internet, and these are terms that are selected by designers of the network. So it functions like a constitution.

It's like a constitution in a second sense, too. Because the initial values it embraced are extraordinarily important and worth defending, then like a good constitution, they ought to be defended when they are threatened by changes in the environment that undermine its core. It's a metaphor to say it's like a constitution but it is the most significant constituting force in the context of the Internet.

In your opinion, what should be done to protect the freedom of the "Internet commons" from increasing control? What role might government play?

At the physical layer, the government has to take steps to induce much greater competition in the deployment of broadband. It can do that by opening up and ensuring a much broader opportunity for wireless technologies to deliver the last hop to the home. It could fund the deployment of basic infrastructure. If it's a national information superhighway, then we should spend money like we spend on regular highways and make sure it gets deployed in a neutral way. But most importantly, the government needs to architect the Internet's neutrality.

In the context of the content layer, the obvious thing to do is for the government to take steps to weaken the overly strong intellectual property rights. One way to do that is to change the term of IP, so instead of the term being essentially forever, the way it's become, it should be five-year renewable terms for copyrights to a maximum of 75 years, and people should be forced to register and renew their copyrights in order to get the benefit of the government monopoly.

This is a part that's bizarre to me. When I suggest this, copyright holders are outraged at the idea they should have to register and renew their copyrights in order to get the benefit of government protection. You can imagine a system where this registration and renewal only takes as long as it takes to buy a book on Amazon.com—practically nothing. The idea that you can't require people who are getting this very significant government benefit to take a little bit of effort to get that benefit seems crazy. Those steps ought to be taken so stuff doesn't become copyrighted forever in a way that's not even desired by the copyright owner.

There are other more affirmative things. We're building something called the Creative Commons, a kind of IP conservancy that will try to encourage people to donate IP into a conservancy for other people to draw from. Congress could take steps to encourage more of this. That would be a great encouragement to expanding the public domain.

Are there any groups working to lessen the length of copyrights on software? Is Congress aware of the difficult issues regarding this?

I'm not a politician, I just write books and try to make arguments of what I think is true. I just observe that in our system that doesn't seem to be enough. So I'm not really optimistic.