ブロードバンド時代の制度設計

Now the competition of Internet service providers meant that ISPs needed to develop fast, cheap access to this network and more importantly could not behave strategically to attempt to control their customers in some way that was antithetical to their customers' access to the broadest range of Internet resources. Competition ensured that the control over the physical layer did not result in strategic behavior by the owner of the physical layer to protect itself against the emerging Internet and regulation guaranteed that access to the physical layer for narrowband Internet access would be preserved. This was an accident of regulatory history, I am suggesting, because no one who thought about the break-up of AT&T in 1984 thought the reason was to give birth to something like the Internet. Nobody had any conception that by liberating the physical layer of this telecommunication network to be applied to any use people wanted, they would be giving a spur of innovation to a radically different use of this network which at that time was just beginning to be developed in research labs around the world.

But because of the unintended effect of that break-up, there was an environment for innovation at the physical layer that supported the development of the Internet. Now the Internet is moving from narrowband access to broadband access, from telephone wires to cable wires in the United States and increasingly to wireless access regulated through a system of access granted through licenses by the government. But as Internet access moves from narrowband to broadband, the regulatory environment of that access changes dramatically.

Telephones were effectively regulated as common carriers in the United States. The essence of the regulatory regime governing cable is a fundamental right to discriminate. The essence of the cable owners right is the right to choose which content flows across their network and therefore the right to decide what kinds of content and applications will run on their network. And so as cable began to develop Internet access it brought to this technology the presuppositions that they had developed in the world where they delivered content to televisions. They thought, "When we deliver content to televisions we have a perfect right to choose how that content gets delivered. Therefore when we deliver content to computers we should have the same right to choose what content gets delivered and how it gets delivered."

So they increasingly push for the development of a physical architecture that supplements the code layer of the Internet in a way that gives them control over content on the Internet. They develop routers. Cisco has a router that permits policy-based routing that means routing where the router owner chooses which content goes quickly and which goes slowly; which content is permitted and which is not; which applications will be allowed on this network and which will not be allowed on this network. The policy-based router is like a television set where Channel 6 comes in clearly but Channel 8 is intentionally fuzzy because the owner of the television set decided Channel 6 is to be preferred and Channel 8 is to be discriminated against. This is their way of doing business carried over to the context of the Internet and they have developed a set of technologies that enables them to implement that control.

The point to see is that that set of technologies is fundamentally inconsistent with the original ideals of end-to-end design. Because unlike a design where the users of the network decide which content they will get access to and under what terms, this is a design where the network owners choose who gets access to what content under what terms. It is a design where the owner stands between the customers and therefore a design where innovators must decide whether their innovation is likely to be permitted by the cable owner. And if the cable owner is unlikely to permit this particular innovation, then the strategic costs facing that innovator are much higher and as you increase the strategic cost the innovators face, you reduce the range of innovation.

So for example, when AT&T's head of Internet services, Daniel Somers at the time, was asked whether AT&T cable services would permit the streaming of video to computers-after all they make lots of money streaming video to television sets-would they permit uncontrolled streaming of video to computers? Somers said, "We did not spend US$56 billion on a cable network to", quote, "have the blood sucked from our vein". This is AT&T, just like AT&T in 1964, expressing its view that it has a right to choose what kind of innovation will go on on the network. So if the innovation is streaming video to computers, this is the expression of the company that it has a right to decide that will not be allowed; that a competitor to its video services to television sets will not be permitted on this network and their right to exercise that control, cable companies believe, is guaranteed to them by the very nature of the regulations governing cable and maybe not just regulations.

For in a court in Broward County, Florida, a cable company succeeded in getting the district judge to agree that the right of a cable company to discriminate in access to content and services is not just a right guaranteed by a statute, but this judge said it is a right guaranteed by the First Amendment to the United States Constitution. That the First Amendment means that the cable owner must have the right to discriminate just like the First Amendment means that the owner of the New York Times must have the right to choose what content goes on page four. So because of the constitution, this judge believes that it guarantees cable providers the right to discriminate, this means that the physical layer of the Internet, to the extent that it comes controlled by cable, becomes an infrastructure which is subject to precisely the kind of discrimination that the end-to-end principle abhors. This physical infrastructure becomes an infrastructure to enable choices by the network owner over the types of innovation that would be permitted and therefore an infrastructure that undermines the opportunity for a broad range of innovation and creativity that the original Internet guaranteed. Now those are changes at the physical layer, and I hope later today we can talk about how wireless communication is an important part of that dimension of freedom or control.

But now I want to describe a much more troubling transformation that is occurring right now in the context of the Internet that is also likely to compromise the original architecture of end-to-end in a way that will undermine that type of control. This is a change that is brought about by people who I produce for a living by lawyers who are arguing for a particular conception of intellectual property that has, as its consequence, the effect of radically transforming features of the original Internet's design to guarantee that these owners of content have the power to control much about the innovation in the Internet of the future.

Now the core right which they assert is the right to control copyright although the changes in patent law too, are changes that will threaten much of the opportunity for innovation in the context of the Internet as software patents and business method patents begin to pollute this environment of innovation. But I want to focus first on the context of copyright. Now as lawyers should know, this idea of copyright is a very odd right. We describe this as intellectual property but the term "intellectual property" is a term, which emerged in the United States at the turn of the nineteenth century. When the framers of our constitution created the power that Congress has to grant copyrights they spoke of copyrights not as property but as government backed monopolies and they struggled with the question, should Congress have the power to grant government backed monopolies, because they knew against the history of English copyright law that the opportunity to grant monopolies like this is always an invitation to corruption. So they established under great objections from founders such as Thomas Jefferson, a limited right to issue these monopoly powers. They granted Congress the power to, "promote the progress of science." That was the power. And they could promote the progress of science by granting exclusive rights to authors for their writings for limited times.

Now the first statute in the United States that granted copyright protection in 1790 granted copyright protection consistent with that grant in an extremely limited way. The first statute granted copyright for maps, charts and books if the author registered for that copyright and for an initial term of 14 years. Between 1790 and 1799 though, 13,000 titles were published and a total of 556 registrations were made. Ninety-five percent of the creative work at that time fell into the public domain upon publication. The right that was granted for 14 years was a right that extended to control republishing of this initial copyrighted work. It was a regulation of publishers that said that you could not take this work and republish it for commercial purposes. In 1790 there were 127 publishers in the United States. And our outrage in the United States for China notwithstanding, we should remember that until 1891 foreign copyrights were not protected in the United States. We were born a pirate nation.

Our law said to publishers in the United States, focus your piracy on French and British texts and leave American texts alone. Now the framers of our constitution established this limited copyright not because they were communists, not because they had no idea of the importance in values of property, not because they had not figured out that the market would make prosperity work, but because they realized the creative and innovative process was a process that constantly drew upon the public domain in the creation of stuff that was new and the important function of regulation was to assure a sufficient return for innovators so that they would innovate and create but then to channel their creative work into the public domain as quickly as was feasible. The objective was to assure a rich public domain that people could build upon so that creativity generally could be diffuse and uncontrolled by important actors like the church. This was copyright of the Enlightenment. It was copyright that believed that the objective was to decentralize control over this resource and diffuse resources that many could draw upon was the source of innovation and creativity in their minds.

Now copyright law in the United States has changed dramatically over the past 210 years. Copyright law now covers not just maps, charts and books, it now covers essentially any creative act reduced to a tangible form. It extends not when someone exercises effort to register this monopoly right, it is created to upon the reduction to a tangible form automatically. It extends not just to protect against the republication of the thing that is copyrighted, not technically to just copies, but also to derivative work, stuff that builds upon the copyrighted work, stuff that uses the ideas in a way which is sufficiently close to the original copyrighted work, and therefore extends this monopoly protection to the original work and those works that it directly inspires. And it extends this protection not for an initial term of 14 years but for a term of the life of the author plus 70 years. Which in the case of authors such as Irving Berlin would mean a term of 140 years. This expansion of control has transformed this tiny regulation of publishers into a general regulation of all who would engage in copy of the creative process. And contrary to the vision of a system that would channel stuff into the public domain the United States Congress has now adopted a practice, which essentially assures that nothing ever falls into the public domain.

In a series of statutes, which we affectionately refer to as the "Mickey Mouse Protection Acts", the United States has 11 times in the last 40 years extended the terms of copyrights for works that have already been created. Contrary to the original vision that we give a monopoly protection as a quid pro quo for new creativity, these works continually give a monopoly protection as a quid pro nilo; something for nothing. A protection for works that have already been produced as a monopoly favor in an exchange for what is obviously an extraordinary amount of money contributed to political campaigns, especially to Democrats in the United States. Now this process is fundamentally inconsistent with our framers conception of the creative process. This is not a process for creating innovative works which then fall into the public domain and therefore are free in the sense that I described them, uncontrolled by particular owners. This is a process for concentrating control over culture and creative works of culture into the hands of a relatively small set of actors who then get to decide how this culture gets deployed. This is a process for concentrating control over the development of creativity rather than a process for defusing control over the development of creativity, the process that the framers of our constitution envisioned.