Those who think network neutrality is the answer might be asking the wrong question

The recent announcement by the FCC Chairman of impending rules on network neutrality caused a lot of stir in the press, including our own coverage here. On the whole, we think network neutrality is a good idea. But it may not achieve the Chairman’s stated goal: to preserve the open character of the Internet that fostered so much creativity and innovation in its early days. Those times are slipping away. No set of FCC rules, however well intentioned, will bring them back.

We quickly forget how the Internet used to be. Early technologies of other kinds – 8-track tape players, manual typewriters, tail-finned cars – are easy to remember because we have still examples to look at. But the early Internet has vanished without a trace, save in the reminiscences of the one-time early adopters.

Ah, The Good Old Days

Before the Internet was the precursor ArpaNet, which linked a few dozen major universities and Government facilities. Mainly it was a way to transmit research data, with a pasted-on afterthought called “email” to help researchers coordinate their exchanges.

As students and employees left the ArpaNet sites, they worked out ways to stay linked to the system over ordinary phone lines. Some set up connections their friends and colleagues could dial into, thus becoming the first Internet service providers (ISPs). Using the Internet in those days took technical know-how. Conveniences like URLs and clickable links had yet to be invented. There were few graphics, mostly just text. Access was by typing into command lines. Needed functions like “finger” and “Archie” and “Telnet” were complex and hard to use. Also we walked five miles to school in the snow.

Then, in 1991, came the World Wide Web.

Today many people think the Web and the Internet are the same thing, but in fact the Web was a late add-on. Suddenly anyone, with or without technical aptitude, could learn to use the Internet in a few minutes. One needed only to type an address or click a link. Thousands of ISPs sprung up to serve all the new users. In part because the FCC’s earlier Computer II and III rules had assured ISPs equal access to the phone lines, competition among them was fierce. ISPs vied to bring the most content possible to their customers. People with obscure interests had a new way to find each other, to form on-line communities, and to trade conspiracy theories or antique car parts, as the case may be.

It is hard to remember now what a big shift this was. Before the Web, to reach a large audience required going through a book publisher, TV network, or some other powerful gatekeeper. But the Web allowed anyone with a computer and a phone line to make his or her views available to millions. Conversations surged around the globe on every conceivable topic, from the conventional to the bizarre, from cookie recipes to outlandish sexual practices. And yet, amid all this vigorous dialogue, there was almost no large-scale commercial content.   Big companies distrusted the Internet because they could not control it. The medium was too open, too accessible, too accommodating of contrary views.

When people talk about the good old days of the Internet, this is often the period they have in mind – roughly the early 1990s. The Internet itself was big, and getting bigger very fast, as new subscribers signed on in droves and new sites appeared by the millions. Amazingly, though, no one was in charge. The only form of control was a simple system for registering domain names. Any user could post any information and visit any site. Even if a company or government had wanted to limit information or access, there was simply no way to do it.

Large companies soon woke up to the fact that all those millions of computer screens were lighting up the faces of millions of consumers. Cautiously at first, companies began to test out the Internet, first as a marketing tool and then as a channel for conducting actual sales. One early use was for brochures promoting new cars and the like. Internet-only retailers such as Amazon appeared. Chains like CompUSA and Lands End began offering Internet sales in parallel with their physical stores and phone operations. And then came the insight that set off the dot-com boom. If you can sell something that need not be physically delivered in a box, like mortgages or insurance or travel services, then you can sell it on the Internet for little more than the cost of electricity.

As more big firms came to rely on the Internet, they took an interest in seeing it run reliably. The casual patchwork that served in the early days was gradually supplanted by a much larger and more expensive infrastructure run by behemoths like MCI. Other companies installed massive facilities around the country that replicated popular sites willing to pay for the service. A click to access would deliver the content quickly, from a server close by, while the site of a small, specialized bookstore across the country might take much longer to reach the screen. The simple domain-name set-up was replaced by a more complicated scheme that gave big companies a leg up over smaller ones. A plumber named Jim McDonald who wanted to use for his company website was plumb out of luck.

With commercialization, web-page design evolved from a hobby to a profession. Content became richer and more slickly produced, with more graphics and photos and sound and even videos. End users became impatient with the time it took these increasingly sophisticated pages to load. Once modem makers had wrung the last few bits out of ordinary phone lines, the only recourse was broadband. Consumers signed up for it in droves.

The Narrowing Effect of Broadband

Building out a new broadband connection to every home and business was prohibitive. Fortunately there were wires already in place, and ways to upgrade them. Adding components to the phone company’s gear made DSL possible over voice lines, while additions to the cable system allowed for two-way Internet along with one-way video. Other providers offer broadband over electric power lines, via satellite, and through fixed wireless connections, but none of these has made much of a dent in the market.

Where a dial-up customer could choose among a great many ISPs, a broadband customer usually had two at most: the phone company and the cable company.  That limited menu worried the FCC, which considered extending the principle of Computer II and III so competing ISPs could connect through the DSL and cable facilities. But the phone and cable companies insisted they would have no incentive to build out the networks if they had to share them with competitors. The FCC bought the argument, with the result that broadband ISP customers in most markets are limited to two providers at most. Only a few consumers live within range of municipal public Wi-Fi. Cell-phone modems are an option for some, although expensive. FIOS fiber-optic service is available here and there, but again is provided by a major phone company.

Protecting the Parent

When thousands of ISPs competed for subscribers, no one cared about network neutrality. Now we do. Why might the cable and telephone ISPs be more likely to discriminate than their predecessors?

The early ISPs were mostly small businesses, many of them literally mom-and-pop operations. Even the bigger players, like Netcom, were primarily ISPs. Today’s broadband ISPs, in contrast, are all in some other business. The ISP offering in each case started out as a sideline. While it has become a growing piece, the parent companies still have other interests to protect. 

That creates incentives for ISPs to discriminate against certain content. One category at risk is anything that directly competes with some other part of the ISP company. There was an outcry when Comcast blocked subscribers’ access to BitTorrent videos, possibly to protect Comcast’s own video-on-demand service. (We reported that here.) Also in danger is anything that might expose the ISP parent to even a remotely theoretical chance of criminal investigation. The major ISPs shut down vast segments of Usenet, a widely-used group discussion service, because a minuscule portion might have been used for distributing child pornography. Subscribers needlessly lost access to enormous amounts of fully legal material. Another category to watch is anything the ISP parent company fears might offend someone’s political sensitivities. Verizon once refused to assign a text-messaging short code to a political group solely because it feared the political content of the messages might be controversial.

In the early days, vigorous competition corrected any such inclinations. But today a home or business broadband subscriber who wants access to blocked content has few options. In principle there is more competition for wireless broadband services, because there are more companies in the market. They certainly tout the fact. “The wireless industry is ultra-competitive,” says their trade association website, “and that means you win with a wide variety of service plans, options, and features from which to choose.” In practice, though, the providers’ use of handset tie-ins and early termination fees makes competition largely ineffective. AT&T won’t let iPhone users run VoIP services over its broadband network, and Apple won’t let any other carrier connect to the iPhone, so iPhone users who want VoIP are stuck. For many of the Verizon users outraged by the company’s hindering their political message, changing carriers would mean paying early termination fees and buying new handsets.

Network neutrality rules will help. The ISPs and carriers no doubt will make the same bat-and-ball argument they have in the past: they built the networks, they own them, and they can run them as they please. In fact, though, the broadband networks could be viewed as a semi-monopoly, and we often take measures to ensure that privately owned monopolies treat their customers fairly.

But let’s be realistic. Although network neutrality rules will look good on paper, they will be frustratingly hard to enforce. The Chairman proposes to evaluate alleged violations “as they arise, on a case-by-case basis.” To anyone who has spent time around the FCC, this is the formula for a slow and expensive process. An innovator with novel content who runs afoul of discrimination by the broadband ISPs will not often have the time and money to litigate against a deep-pocket telephone or cable company.

Network neutrality rules may deter some of the worst abuses. But they will not bring back the anything-goes character of the early Internet. Anything went, in those days, because control was physically impossible. Now that a few companies do have control – or at least the capacity for control – all we can do is tell them not to use it. That may not be enough.