CHAIRMAN REED HUNDT FEDERAL COMMUNICATIONS COMMISSION INET '96 CONFERENCE Montreal, Canada June 28, 1996 (Delivered by Blair Levin, FCC Chief of Staff) A+B = C Access + Bandwidth = Communications Revolution Here we are at last smack dab in the digital age; an age of promise, an age of possibility; and for many an age of anxiety, apprehension & alarm. The embodiment of this era is, in all respects, the Internet. This a technology every bit as revolutionary as the invention of the telegraph was over 150 years ago. The Internet grew out of a small project started by the U.S. Department of Defense to link up four military research centers. Similarly, the telegraph got its start as a small government grant -- $30,000 awarded in 1843 led Samuel Morse to the first test of a telegraph line from Washington, D.C. to Baltimore in 1844. The Internet can help us achieve many goals we've had since that first city-to-city click. Even though it's been 150 years since Samuel Morse's invention, much of the world does not yet have basic telephone service, millions of American households do not have active phone service, and our kids spend each day in classrooms, 90% of which are unconnected to the world of information. The growth of the Internet provides new ways to bring all children and all Americans into the information age. Today I want to talk about the questions we face in assuring that we can realize the full promise of the Internet. In preparing for this speech, I asked my staff to explain to me how the Internet works. They said I wouldn't understand. I told them I was the first FCC Chairman to have a computer on my desk. That's our point, they said. I persisted -- so in desperation they resorted to charts. Then they showed me a picture of a plate of spaghetti. The Internet, they said, looks like this. And they showed me pictures of trains in a switch yard. It works like that, they said. Then they showed me curves that sloped up like the ascent of Everest. It's growing like that, they said. Then they told me that, based on current projections, in the year 2010 there will be more Internet users than people on the planet. That's a little hard to get your mind around. But I imagine that at least by then Slate will be making money. Here's what I took away from these desperate, nearly hopeless efforts to explain the Internet to a lawyer. The Internet changes everything. Its digital code turns a zero into a window on infinity and a one into a unifier of economies and society. It turns the world upside down. In the words of Jimi Hendrix, a six becomes nine, and I don't mind. Jimi Hendrix is, I know, a historical figure who needs explanation for the Internet group. He is a person whose accomplishments in the musical field are memorialized in a Seattle museum owned by Paul Allen, cofounder of Traf-O-Data, a small Seattle business that changed its name some years ago to, if I remember correctly, Microsoft. And because the Internet changes everything about the way we communicate, it necessarily changes all our governmental communications policies. To drill down a little deeper: it would be more accurate to say that the Internet gives us the opportunity to change all our communications policies. George Washington Plunkett, the famous 19th century politico in New York explained the secret of his success this way: "I seen my opportunities and I took them." So what should we do with the opportunity to change communications policy that the Internet gives us? I see five major questions that need answering at the get-go. 1. How can public policy promote or at least not deter expansion of bandwidth to power up the development of the Internet? 2. What rules can we get rid of and what rules should we write to promote the development of the Internet? 3. Should we be concerned that the economics of pricing on Internet -- influenced as they are by current out-of-date regulatory policies -- won't in fact sustain development of the Internet? 4. How can we make sure that Internet reaches all Americans, especially kids in classrooms? 5. How can we make sure the Internet reaches across the globe? Don't we need government policy in the United States and worldwide to guarantee that Internet access becomes a truly universal product? Don't we need to guarantee that everyone gets access to the common network of networks? We at the FCC have seen the benefits of the Internet first hand. One of my first actions as Chairman of the FCC was to speed the replacement of the outdated FCC communications equipment and mainframe computer systems with a state of the art internal computer network, ISDN phones, and PCs on all staff desktops. Last fall I unplugged our 25 year-old mainframe because we can now do much more with networked PCs and workstations than we ever could with that behemoth in the basement. If anyone is looking for a good deal on a used Honeywell, come talk to me after the speech. I'll even throw in a bunch of rotary phones that we still have lying around. When we put up our World Wide Web site in 1994, we initially got about 100 hits a day. That reflected our low-budget operation and the small community of lobbyists who normally visit us. But our goal was to open up our doors on the Net and let all Americans participate. We're getting there. We now get 50,000 hits a day and transfer almost 400 megabytes of data. We get visits from more than 50,000 unique hosts every month. To some of you, this may not sound like a big deal, since there are now many sites that get over a million hits a day. But, for the first time, members of the public can get copies of FCC information shortly after its released -- or months later -- and for no charge. We couldn't afford to duplicate and mail copies of our decisions and resource materials to every home, or even every public library in America. But we can put everything online, where anyone in the world with an Internet connection can get access to it. The explosion of external communication with the FCC is nothing compared to what has happened inside our walls, up and down our linoleum-lined halls. Not only are we on the Net -- we are accommodating our own internal communications and networking needs via an intranet and remote access connections to our servers. What this shows is that an open, distributed, interoperable packet-switched network is an incredibly powerful force for change. The success and openness of the Internet is attracting a proliferation of new services. Streaming video & audio may alter TV and radio as we know them in the long run. But in the short run, these and other new services give us three major objectives -- bandwidth and access, more bandwidth and more access, and even more bandwidth and even more access. That's what our policies should promote, facilitate, encourage, or at least avoid deterring. The traditional communications industries don't think in this way. They think they are in the telephone business, or the broadcast business, or the satellite business. They don't think they are in the access and bandwidth business -- but they are. And when people used to think of the FCC, they thought about universal service, affordable rates and quality service. Now they recognize, I hope, that we're in the access and bandwidth business. Or, to be honest, when people think about the FCC, they think about radio shock jock Howard Stern. But all these issues -- yes, even Howard -- are really about bandwidth. Our policies have too long and too often been focussed on services: how to keep local telephone prices or cable rates low, or how to get lower accounting rates on international telephone calls. Bandwidth and access are not goals that presuppose any particular service. They are means not ends, and over time our focus on services can diminish even as our zeal to maximize bandwidth and access is unabated. After all, if the bandwidth and access are there, the services will be invented. Indeed, visions of services are dancing through the heads of you in this room. Bandwidth on networks accessible to all will make those visions possible. Of course, no one can afford to build, on their own, the spaghetti plate of networks that is not only the Internet but also the underlying telephone network. Government is necessarily involved in the promotion and development of bandwidth and access. As Lincoln said, "The legitimate object of government, is to do for a community of people, whatever they need to have done, but can not do, at all, or can not, so well do, for themselves in their separate and individual capacities." Building a network to some people -- like the FCC's intranet -- is something that some communities can do for themselves. Building a network to all people -- which is what the Internet should be -- is something that needs to be done, but that none of us can do acting alone or so well as if we act through government. I want to dispel the myth that bandwidth is infinite and access to everyone is guaranteed without collective action. The FCC, like similar agencies in other countries, was created because there are limits on the availability of bandwidth. Over-the-air transmission requires spectrum, of course. Spectrum is bandwidth. By definition spectrum is a scarce resource. If it weren't, we wouldn't have collected $20 billion in spectrum auctions. People don't pay that kind of money for a non-scarce community. Spectrum is infinitely inexhaustible, but it is finitely available. You can't use spectrum up by using it, but you can't make more of it than God gave us. Of course Manhattan is a small island but tall buildings permit more people to use it. Similarly, new compression techniques make more use of bandwidth than previously thought possible. I just read that a single swatch of six megahertz for digital television need not be limited to a single channel of high resolution picture but can be used for 24 channels of standard definition. That means that the 10 licenses that Congress is ordering us to give to the broadcasters of Washington, D.C. can be used for 240 terrestrial over-the-air digital channels. That trumps the satellite industry and even cable. It is a tremendous force for competition. Compression doesn't prove that spectrum is no longer scarce. Even George Gilder would agree with that (except possibly in the Keynesian long run -- defined as when we're all dead). But compression does prove that market forces, if unleashed, will make more of spectrum than any government policy is likely to do. That's why the idea that digital broadcast licenses should be regulated to produce just one high definition channel may not be a crime, but, in the words of the French diplomat Talleyrand, "it is worse than a crime, it is a blunder." Bandwidth over wired networks is also a scarce resource, because it costs money to build those wires. We have historically allowed companies to monopolize bandwidth in one region in exchange for the authority to regulate the way those companies offer their services. In 1984, we opened up the right to offer national bandwidth connections -- long distance service -- to competition. In the Bandwidth Act of 1996 -- excuse me, I meant the Telecommunications Act of 1996 -- Congress gave the FCC a mandate to take the next step and open up local telecommunications markets to real competition. Our mission is to give consumers and businesses more services, more choices, and more control over those choices -- in other words, more bandwidth. Congress also asked the FCC to guarantee universal service, especially to kids in classrooms -- in other words, more access. The future of the Internet, and the ability of service providers to offer more bandwidth to their end users, will thus depend on how successful we are at promoting competition in local telecommunications markets and truly universal service. This gets to our second goal: rulewriting or rule-unwriting. We need to write rules that open up the local exchange market. We need to unwrite written rules and undo unwritten rules that inhibit competition. In August of this year, I hope the FCC will issue interconnection rules. I'm going to get technical now -- these should permit the spaghetti strands of competing networks to connect at fair prices and terms to each other. And by the end of the year, all states will impose fair terms and conditions on Bell companies and AT&T and MCI, among others. These terms will translate our rules into the text of the interconnection arrangements available for all network providers and users. But while we need to write interconnection rules, there are other rules I'm not convinced that we should write. The FCC has received a petition from the America's Carriers' Telecommunications Association asking that we restrict the sale of "Internet phone" software, because the providers of that software do not comply with the rules that apply to telecommunications carriers. Similar issues are being discussed in other countries, including Canada. We've just finished getting comments on that petition. We're in the process of reviewing those comments now, but I would just note that the National Telecommunications and Information Administration, the Administration's telecommunications expert, has filed very thoughtful and well-reasoned comments with us asking us to reject this petition. I am also strongly inclined to believe that the right answer at this time is not to place restrictions on software providers, or to subject Internet telephony to the same rules that apply to conventional circuit-switched voice carriers. On the Internet, voice traffic is just a particular kind of data, and imposing traditional regulatory divisions on that data is both counterproductive and futile. Even if most of the FCC wasn't working around the clock on implementation of the Telecommunications Act, I can't imagine that we would have the time to keep track of all the bits passing over the Internet to separate the "acceptable" data packets from the "unacceptable" voice packets. More importantly, we shouldn't be looking for ways to subject new technologies to old rules. Instead, we should be trying to fix the old rules so that if those new technologies really are better, they will flourish in the marketplace. For years, some engineers have been telling us that voice over a packet switched network wasn't possible. The latency periods were too great, they said, and you'd never get acceptable quality. Well, it is going to be possible, and in the short period of time since the first commercial products became available, the quality has been rapidly improving. But the last thing we want to do is stop that improvement by thoughtless regulation. Internet telephony may well become, in time, a competitive alternative to traditional circuit-switched voice telephony. After all, as the growth of the cellular industry demonstrates, people are willing to give up a significant level of quality in exchange for other benefits. In the cellular case the benefit is the ability to make a call from virtually anywhere, in the case of Internet telephony the benefit is a vastly lower price. This is especially true, for example, for international telephone calls. And now the third goal: an economically-sustainable regime. Economics is about the marketplace sending signals that compel users and service providers to do the right thing, to utilize the network in the best and most efficient way, and to encourage the network to be built out in ways that, for example, avoid brownouts by building in redundancy. When we have competition among network providers, many of the pricing problems we now address through regulation will be resolved by competition instead. Consumers will be able to choose a communications company that gives them the best deal on subscription charges, service availability, and charges for making calls. Although competition is seldom perfect, we look forward to being able to rely on it. One problem that competition might not resolve, though, is that, under anything like today's network architecture, someone who wants to call me has no choice but to deal with the carrier that I have chosen to provide my connection. There's a monopoly relationship there, however many carriers vie for my choice. The monopolist on terminating calls to my home doesn't charge me but charges those who make the calls, so I may not care very much (or even know) how much it charges. So, as the econo-guys would say, there's a monopoly problem exacerbated by a principal-agent problem. What then is the appropriate economic model that can sustain the Internet? With the number of users and host computers connected to the Internet roughly doubling each year, and traffic on the Internet increasing at an even greater rate, the potential for congestion is increasing rapidly. The growth of the Internet, and evidence of performance degradation, has led some observers, including Bob Metcalfe, the inventer of Ethernet and founder of 3Com, to predict that the network will soon collapse from excessive usage. Because the Internet interconnects thousands of different networks, each of which only controls the traffic passing over its own portion of the network, there is no centralized mechanism to ensure that usage at one point on the network does not create congestion at another point. Because the Internet is a packet-switched network, additional usage, up to a certain point, only adds additional delay for packets to reach their destination, rather than preventing a transmission circuit from being opened. This delay may not cause difficulties for e-mail, but could be fatal for real-time services such as video conferencing and Internet telephony. Moreover, at a certain point, Internet routers are simply unable to handle the load and will "drop" packets, causing network "brownouts." Such brownouts are already occurring. A group of high-energy physics researchers recently sent a memo to the Federal Networking Council complaining about "catatonic" connections that made it impossible to effectively share scientific data over the Internet. It's like Yogi Berra's old line about a popular restaurant: "Nobody ever goes there any more; it's too crowded." Even if individual providers upgrade their networks to achieve sufficient capacity, end users may still experience congestion delays when their traffic must traverse another backbone provider's network, or the smaller networks to which the receiving computer is connected. Standards bodies are developing new standards to reduce congestion and make possible more reliable connections for bandwidth-intensive applications, but these standards must be widely deployed within the network to have a significant effect. The increasing levels of Internet use are also beginning to affect the telephone network. Carriers engineered and deployed their switches based on the characteristics of voice traffic, where a conversation lasts an average of three minutes and an average customer attempts about three calls per busy hour. Internet users, however, typically engage in far longer calls than voice users. As a result, Internet usage is placing unexpected demand on local exchange carriers' switches, to the point that switch congestion is threatening service quality for voice users of some switches. This might not be such a big problem if we had an all-digital phone network that was based on a packet-routing multimedia technology such as ATM. But we don't, yet. We've got hundreds of billions of dollars of plant in the ground that works fine for voice, but that gets strained by the requirements of data communications. So the hard question is: if there are costs for upgrading the network to support the explosion of Internet and other data usage, who pays those costs? The FCC decided in the early 1980s that enhanced service providers, which include Internet service providers, should not be subject to the interstate access charges that long-distance carriers pay to local phone companies for originating and terminating calls. ISPs are therefore treated as "end users" and can purchase lines that have no per-minute usage-based charge for receiving calls from their customers. The phone companies argue that the absence of usage charges means that ISPs do not provide the revenue to cover the additional costs they impose on the network. I don't know what the full answer is to this problem. But I'm inclined to believe our best guidance is to let technology, competition, and access reform make the problem go away. We are working to open markets so that these forces can operate most effectively. I'm also aware that as the Internet becomes a part of daily life for more and more people, reliability and service quality will be increasingly essential. Whose job is it to minimize the likelihood that the network will go down? Last week, Netcom's 400,000 customers lost connectivity to the Internet for more than 13 hours. I don't want to single out Netcom because they aren't the only company that has had this kind of outage, and because this is ultimately an industry-wide issue. How would we feel if Montreal lost its telephone service for a day? I might be happy that no one from my office in Washington could reach me, but I would be concerned about being unable to get in touch with my family. We have a Network Reliability and Operability Council in the United States that addresses these issues for the telephone network. It reports regularly to the FCC, but it's an industry-led group. I know the Internet has the IETF which does a very effective job in setting standards. I hope that an appropriate body can be developed for addressing the issues of reliability in the context of the Internet. And now the fourth goal: Ensuring the availability of bandwidth to all Americans, especially kids in classrooms. We have an obligation -- legally, morally, and economically -- to promote universal service, especially for those with limited resources. The value of networks increases exponentially as more people are connected. In addition, the Internet rides on top of the existing public switched telephone network, and the rapid growth of the Internet is driven in part by the universal deployment of traditional telephone networks. Ensuring that schools, libraries, health care providers, and poor, rural, and insular communities benefit from the Internet and the new communications revolution it represents is one of the great challenges of the beginning of the twenty-first century. Government can no longer assume that it can meet public policy objectives by directing regulated or public companies to invest in a certain manner. But the private sector must also contribute its fair share if we are to escape from what the econopeople call the tragedy of the commons. So today, I challenge the Internet community to provide two years of free Internet access to classrooms and libraries. On this and other issues we find that challenge is to find ways to work together -- governments, businesses, service providers, researchers, and educational institutions -- to solve common problems. Because it's in everyone's interest to have more bandwidth. It's in everyone's interest to have ubiquitous access. And it's in everyone's interest to minimize the artificial barriers placed on the free flow of information. As the Federal Court in Philadelphia properly recognized two weeks ago in enjoining enforcement of the Communications Decency Act, the Internet is a new medium. It encompasses some aspects of traditional media such as broadcast and telephony, but it is much more than that. Whether or not it was a wise decision to pass the Exon Amendment -- and the Court has said it was not -- we still need to recognize that if kids have access to the Internet at home and in classrooms, then parents and teachers need software filters and other tools that empower them to make choices. I've talked to Tim Berners-Lee about this, and he said he and his very able team are working on addressing this very problem with the PICS system. If we want, as I do, for our children to ride the information highway from Carthage, Tennessee to the Library of Congress, we can't permit the virtual school bus driver that takes them on that field trip to travel through the red light district. It's legitimate to challenge the Communications Decency Act in court but again, it's not enough to just say no. Your policy goal and our policy goal has got to be to fulfill the promise of the Internet, not just protect the problems of the Internet from governmental solutions. My final point is that our Internet policies must be international, because the Internet is an inherently global medium. Although the largest concentration of Internet users is still in the United States, the fastest growth is occurring in other countries. Internet users may not even know where in the world content providers are located. Material that is legal in one country may be illegal elsewhere. It doesn't make sense to subject providers in one country to the laws of a different country simply because, unbeknownst to the provider, someone in the second country downloaded its material from the Internet. To the extent that there are legitimate, agreed-upon governmental objectives -- such as privacy and protections against fraud -- that may be adversely affected by activities conducted over the Internet, governments should look to international, and wherever possible industry-driven solutions, rather than acting unilaterally. A second critical area in which we must have global solutions is standards. The Internet has been successful because it is based on open, interoperable standards. In addition to technical standards for the transmission of data across the network, frameworks are needed in areas such as electronic payments, intellectual property, commercial codes governing contractual agreements, security, and privacy. As the Internet grows and becomes more commercialized, however, existing mechanisms may not be sufficient. There are many administrative functions that are essential to the smooth functioning of the Internet, which are handled by a variety of different governmental and non-governmental bodies. We must consider whether any of these functions can be coordinated in a more efficient way, without losing the flexibility and openness of the existing processes. When I say coordinated, I do not mean government-controlled. Although there are occasions where it is useful for governments to participate in the standards process, the process should be international and driven by the private sector. The efforts of Tony Rutkowski and the Internet Law and Policy Forum on issues such as domain name assignment are an encouraging example of how such a process might work. So we have many challenges before us. I've outlined what I think are some of the most difficult problems because I think they are also the most important. And all of this is important precisely because of what the Internet means for tens of millions of people today, and what it could mean for hundreds of millions more in the next decade. This came home to me recently when I traveled to Johannesburg for the meeting of the G-7. While there, I visited the township of Alexandra. Surrounded by 350,000 people from all over the world, I thought about how many of these people came from terrible war torn areas to this frightening township, a place without clean water, electricity or police or jobs. But while I was there, I looked up and saw around Alexandra Township the towers of the world's second largest digital mobile communications system. These towers deliver telephone calls. They can also deliver the power and the promise of the Internet. I was proud to be there because in Johannesburg we Americans announced the Mickey Leland initiative to bring the Internet to Africa. If you really believe in the theory of the Internet, in the democratization of this medium and the need and the ability to reach all people, then you should be the biggest proponents of expanding their access to the Internet. You should support this initiative as well as Net Day and any other effort designed to connect us all. Bandwidth and access in an information world should be on a par with clean air and water -- we are all benefitted when these fundamentals are available from Canada to South Africa from Alexandria, Virginia to Alexandria, Egypt to Alexandra, South Africa.