Three Unconventional Sources for High-Bandwidth Internet Growth

by Andy Oram
August 17, 2001

Business leaders are traditionally independent-minded and circumspect about asking government to launch broad social policies. So it was particularly noteworthy that, toward the end of the 1990s, industry conveyed to Congress the alarming message that both content and conduit on the Internet were going to stagnate without intervention on the policy side.

Congress responded by passing two massive, ground-breaking laws. Both revolved essentially around incentives. The laws reflected a realization that current incentives were not fostering adequate Internet growth, along with the economic benefits it could provide in the coming century.

The realization was correct, but the remedy was not. And the current Internet slump shows that the country needs a radically creative approach to Internet growth.

The two mega-laws that failed were the 1996 Telecommunications Act and the 1998 Copyright Act. And while they are impossible to summarize—both consisting of multiple provisions covering many activities by different industries and running into dozens of pages—the key flaw underlying Congress’s mistake was the same in both cases.

Basically, Congress knew that current industries (the telephone and cable companies in one case; the publishing, recording, and software industries in the other) were not stepping up and providing the investment or innovation needed to bring the benefits of high-speed Internet use to everyday life. The mistake was the assumption that investment or innovation would need to come from these industries. Congress compounded the mistake with the trust that it could bring about a broadband revolution by steeply increasing the rewards it gave the industries.

As of summer 2001, the telecom industry has no far-reaching initiatives to boast of (only some jerry-rigged broadband offerings in cable modems and ADSL that would probably have happened even without the Telecom Act) while the recording industry has used the Copyright Act to constrict rather than expand technological opportunities. In this paper I will summarize a few key clauses of both acts and explain why their effects were negative. Then I will go on to the much more interesting question of where we might find new sources for innovation and investment. How can the public drive a public broadband Renaissance despite the recalcitrance of the old-guard industries?

The Telecom Act: A Response That’s All Static

The key issue leading to the Telecom Act of 1996, at least so far as the Internet is concerned, is the famous “last mile” problem. Current copper wires are unsuitable for high-speed access, and stringing new wires (let alone optical fiber) is far too expensive for the average user.

The Telecom Act addressed this gap by trying to increase competition in “advanced services,” within a larger agenda of increasing competition throughout the telecom sector. But Congress’s concept of a “competitor” was extremely impoverished. They figured that the investment needed to create new lines had to come from enormous companies with deep pockets, so they relaxed rules about cross-ownership, mergers, and entry into new markets in order to weigh down the pockets with more gold.

Many cynics say the true intent of the laws was to let the major telecom and TV players enrich themselves; the goal of competition becoming just a fig leaf. While cynicism is usually justified where major business-oriented bills are concerned, it is still worthwhile taking the justification for the Telecom Act at face value and seeing why it was flawed.

The first hope of Congress was that local phone companies and cable companies would come out of their corners fighting. They even created a new category of competition called “Open Video Systems” in the quaint expectation that phone companies would upgrade their lines for high-speed access and earn the rewards of offering TV and movies to their customers. While a few highly-publicized trials of this service were unveiled in the 1980s and 1990s, the phone companies couldn’t bring it to a state of commercial readiness; by now the idea has less heft than ocean-traveling dirigibles. (The one major new cable company to arise since the Telecom Act, RCN, started out with OVS status but quickly switched to regulation as a conventional cable company.)

Competition on the cable side is significantly more successful. In many parts of the country, both high-speed Internet access and local phone service are available over cable lines, sometimes from two different companies. Yet the offerings fall short of what broadband should be: geographic coverage is far from universal, bandwidth is asymmetric, heavy restrictions are placed on usage (no servers, no long video downloads), and reliability is no better than cable TV as a whole.

Cable modems are an impressive success only because the phone companies’ response (ADSL) is so problematic. The design of the cable system—with its straggly upstream connection and shared bandwidth—was just never meant to support advanced Internet applications like two-way videoconferencing, and no one should ever expect cable modems to provide a long-term solution.

The long-term promise of the Internet is to give every user a voice and provide rich media upstream as well downstream. (In fact, the very terms “up” and “down” betray old-fashioned thinking that is unsuited to the Internet.) Neither owner of the main wires into the home—local telephone companies or cable TV companies—has a strategy for providing this promise.

Congress harbored another hope for competition, both in POTS (plain old telephone service) and in high-speed Internet service. They hoped that small telephone companies would gradually invade the market owned by local monopolies (the baby Bells), cherry-picking the easiest and most lucrative customers first, then challenging the incumbents at all levels of residential use.

But this incremental strategy required the Bells to let competitors interconnect with the Bell networks at convenient points. If a customer of the competitor could talk only to other customers of that competitor, it would be asking people to go back to the 19th century before interconnection (and the founding of the Bell monopoly). So Congress required the Bells to interconnect with competitors, a commitment that carried with it a host of requirements like selling lines at low rates and handling service quickly.

There is no reason to regale readers here with the five-year history of legal obstruction and anti-competitive behavior on the part of the Bells. The outcome is all too clear. Some competition exists for the high-volume business market, but practically none for residential users.

I should, however, mention the notorious quid pro quo that Congress offered in Section 251 of the act: if Bells could prove to the FCC that they had a competitive local market, they could start offering long-distance service in that same geographic region. The very presence of this clause has affected competition negatively. Whereas the 1984 break up of AT&T categorically excluded the baby Bells from long-distance, now they saw a chink in the armor and started prying away at it.

Many hundred-page filings later, while competition is obviously non-existent to anyone who has eyes to see, the Bells are breaking down the will of the FCC and cracking a few long-distance markets. The Bells’ friends in Congress have continually tried to bend the delicate agreement in the Telecom Act in favor of the Bells, most recently through the Tauzin-Dingell broadband bill that would open long-distance markets to them (but supposedly only for “data,” not voice communications.) Meanwhile, their only high-speed offering at a price consumers can consider is an ADSL that is very geographically limited, hard to obtain, and overpriced compared to cable modems.

Wireless Internet must also be mentioned as a source for hope. Ignored in the Telecom Act, it has been grudgingly allowed a few slices of bandwidth by the FCC. Intrepid technology activists and ISPs have jumped in to offer community networks that suggest intriguing possibilities for the future.

What the community networking movement needs is a wide swath of clean spectrum. Unfortunately, what the FCC has given them is the dirtiest and most demeaned bands, where a wide range of devices from microwaves to baby monitors can interfere. Power restrictions keep the range low. The spectrum policy instigated by Congress and pursued by FCC may prove ultimately to be their worst sin against the American people in the area of telecom. Under the spell of 3G hype, enormous ranges of bandwidth have been sold to private companies who haven’t been able to build a viable business with them. The FCC has even raided and pillaged the spectrum used by non-profits for broadcasting in an attempt to sate this corporate beast. An enormous public resource is going to waste.

The moral is: don’t expect the rich to be moral. If you give them more money in the hope they’ll do the right thing, they’ll just take the money and keep on doing more of what they’ve been doing all along.

The Copyright Act: A Story Line That Never Took Off

According to the common wisdom, content and conduit are two sides of the same Internet coin. Both are required to bring its benefits to a wide public. Without interesting things to do online, few people are willing to pay for high bandwidth. And without high bandwidth, few interesting things will come online.

While some researchers challenge the primacy of formal content (see, for instance, Andrew Odlyzko’s “Content is Not King”) I’ll follow the story line regarding content for now and return later to an expanded notion of what matters on the Internet.

The 1998 bill that changed copyright law was introduced with such opulent promises for the future good of America that its sponsors called it the Digital Millennium Copyright Act. In actuality, it ensures that content providers can wait another thousand years before doing anything to exploit new technology.

Proponents of the bill, both in industry and in government, played two hands at the same time. While they claimed the DMCA would unleash a wealth of digital entertainment and information for the delight of future Internet users, they spoke quite modestly of the actual provisions in the bill, claiming that they merely closed some gaps in copyright law that hampered the application of copyright in digital media. Certainly, as the MP3 revolution has shown, the Internet makes a mockery of the traditional barriers to distributing content. But the bill gave copyright holders much, much more than a weapon against infringement.

To copyright holders, every unanticipated use of content is a problem. It’s a problem for librarians to share their holdings with users. It’s a problem for teachers to display a movie or circulate an article among their students. The original plan for the DMCA would have made these common activities difficult or impossible, along with dozens of other everyday types of access that currently fall under the safe harbor of fair use.

During the legislative bickering that goes on for years before a major bill passes, librarians and teachers happened to win limited exemptions from the harshest provisions of DMCA. But the seriousness of its attack on new technology became clear during the well-known prosecutions and threats surrounding the DeCSS decryption program, and even more after the outrageous arrest of Dmitry Sklyarov. Web sites were prohibited from making links to forbidden content, and a professor was intimidated into withdrawing an academic paper about the technology under debate.

It has become clear from these cases (and from the prosecution of MP3.com and Napster, and from the attempt to suppress “The Wind Done Gone,” and from many other incidents) that the spread of knowledge and the public interest are the furthest things from the minds of large publishers, studios, and software manufacturers. But while they suppressed new technologies that they considered a threat to old distribution channels, they had no success in creating new channels. A rule is emerging in the new technological era that applies to both telecom companies and the recording industry: any company that can’t find sufficient employment for engineers had better hire a lot of lawyers.

The Secure Digital Music Initiative (SDMI) was the recording industry’s big hope for offering music online. But it has practically come to a standstill, suffering all the worst drawbacks of standards committees. As for movie manufacturers, the only technological innovation they have been able to think up is the ridiculous “use once and throw away” DivX format, which died in the marketplace. Unwilling to abandon this idea, they are fumbling around with a digital distribution system that would force every computer’s disk drive to include copy controls. The book publishing industry has recently had to admit that its e-book initiative is a flop among consumers.

Thus, the DMCA shows all the signs of failure and no signs of success. Like section 251 of the Telecom Act, one section of the DMCA—the one on circumventing effective technological measures that prevent access—is doing serious harm. The law has opened up chances for a grievous degradation in the public’s right to information, with no compensating innovations from the companies it benefits.

New Hopes For Implementing a High-Bandwidth, Content-Rich Internet

Regardless of Napster’s and MP3.com’s legal problems, their popularity—along with Linux, the Web, and other nontraditional sources for conduct and conduit—shows that the public is looking beyond entrenched oligarchies to seek new thrills. What we need is some blue-sky creativity in terms of how to finding new content that is so compelling that it drives investment in fiber or wireless. My solutions involve the familiar ideas of disintermediation—that is, going directly to a source rather than handing someone else the power to aggregate and filter it—and replacement—that is, justifying an expensive Internet connection by using it to replace an even more expensive form of physical travel. A massive demand for symmetric, high-bandwidth connections may foster an underground industry that can bypass the incumbent telecom network.

I’ll suggest three possibilities here; I’m sure others will emerge if people use their imaginations.

Teleconferencing from the home

Imagine you run a business and have just found the perfect, high-powered, dynamic individual to carry out one of your key functions. She lives 300 miles away and doesn’t want to relocate, but she’s so well-suited that you’re willing to let her telecommute. Still, fluid relations between your staff are so important that you require her to drive in once every two weeks and spend a couple days at the office.

Adding up the costs of transportation, food, housing, and office space during her visits, how many months would have to pass before you’re willing to invest in a high-speed line direct from her home to her ISP? Could full video teleconferencing replace face-to-face visits? Video teleconferencing is not ideal, but it’s much closer to an intense personal meeting than a phone conversation or email. Many kinds of consultations and personal support are viable once participants get comfortable with the medium. It may be an attractive proposition for lots of businesses.

High-speed Internet is also an ecological issue. We’re using too many cars and planes, and soon our airports just won’t be able to handle all the business travelers trying to get to corporate meetings. We need video teleconferencing, and if it means spending several thousand dollars for a T1 line, some companies will break down and pay for it. Tax breaks would be a major boost for this environment-friendly policy.

Computer-mediated education

Experiments in offering courses online are proceeding slowly. Some of them make creative use of the richness offered by immediate interactivity, such as online math and science courses that involve simulation software. But most computer-mediated education, unfortunately, is just an imitation of some other model that is not well-suited to education. Thus, some offerings are just old-fashioned correspondence courses delivered more cheaply, others are like mailing lists (although these can produce useful results when participation is intense and the instructor has the energy to coordinate it), and still others are just televised lectures.

I have confidence that experiments will pay off and eventually produce entirely new experiences that can sometimes replace classroom education. The goal is difficult because no solution can be cloned exactly to fit another topic and professor. But the stakes are worth the effort, because if done right, distance education can save a staggering amount of money. Businesses and individuals may both be willing to pay for high-bandwidth connections if they can save hundreds or thousands of dollars in travel costs for each course. It’s worth noting that some Canadian schools are paying to string their own fiber to backbone providers.

Like telecommuting, computer-mediated education is a social issue as well as an economic one. Computer scientist Alfred Bork suggests in a series of papers that, with so few teachers and so many people around the globe in need of education, online delivery is the only way to meet the world’s needs.

Given the cost savings in travel and the potential world-wide applicant pool, online education should be able to compete at the educational high end. It is puzzling that so many college administrators try to justify their institutions’ entry into online education as a cost-saving measure. Their projects are usually quite uninspired and seem to be based on the idea that a professor’s insights can be digitally copied with the same degree of perfection as an MP3. Instead, computer-mediated courses should be able to charge what they’re worth.

A successful track record with innovative, high-quality models would assuage those educators who worry about the weakness of remote interaction vis-a-vis traditional classrooms. A sign of success would be online offerings so good that professors choose to use them to supplement face-to-face classroom instruction.

Low-budget entertainment

Remember David Bowie and Michael Jackson and all the rock spectacles of the 1970s and 1980s? They seem to be dying down now. High budgets do not always mean better entertainment.

Perhaps we’ve also reached the limits of spectacle in film. It has a long history (just look at the seminal work of Sergei Eisenstein) but there’s just not much new that you can do with explosions, crowds, and scenery.

Spectacle won’t meet its end in a bang, by striking an iceberg or succumbing to a massive air attack. Instead, the public can be enticed away from it by innovative, online multimedia experiments. Even over low bandwidth, artists can distribute interesting animation and video effects. It’s up to the artists to find low-budget ways to use the channels available on the Internet—channels not controlled by a handful of studios that want to see ten million dollars in projections before they release anything. Interactivity is a key competitive advantage of the Internet, and artists who figure out how to harness it could lead the revolution that’s long been owed us—the revolution that’s supposed to rival the printing press in social impact.

There have been experiments in musicians jamming over the Internet. Performing artists, unfortunately, have rarely been known as sources of massive investment for infrastructure. But perhaps if the Internet can provide low-cost rehearsal and even performance space, actors and musicians will find it worthwhile to invest in more bandwidth.

Re-examining content

I have listed just a few areas that are ripe for a cost-effective broadband revolution. The point behind my suggestions is that content indeed can be king. But it’s not the kind of content Congress was thinking of in pushing the DMCA—content that’s discrete, that’s easily commodifiable and commercialized, that’s tied up with access restrictions and parceled out in a miserly manner. Instead, the users generate their own content. Perhaps we should speak of “culture” rather than “content.” The former term conveys more successfully the range of input that the Internet can transmit and foster.

In short, it’s up to the Internet’s users to find a way out of the current logjam. Concerned network engineers and application programmers can certainly help. If Congress wakes up and decides to stop pandering to the dinosaurs of the old media, they might be able to help on the policy side. Realizing the benefits of connectivity and democratic interaction requires a stretch on everybody’s part.


Andy Oram is an editor at O’Reilly Media. This article represents his views only. It was originally published in the online magazine Web Review.