ICANN's Boondoggle

Modern societies depend on systems of unique identifiers. If you’re a U.S. citizen, for example, it’s vital that your Social Security number be yours alone, or somebody else could start picking up your checks after you retire. Similarly, it’s crucial that no two phones have the same number, no two neighborhoods have the same zip code, and no two products at the supermarket have the same bar code. When an address system expands—say, when the phone companies introduce a new area code—it’s almost always because the community of users has outgrown the existing scheme; it means that the old identifiers either are in short supply or aren’t specific enough.

But that’s not what’s prompting a huge expansion of Internet domains right now. As you may have heard, the relatively manageable list of “generic top-level domains” (gTLDs) that we’ve all mastered over the last couple of decades, such as .com, .net, and .org, is set to expand dramatically starting next year. You could soon find Amazon at amazon.book and Google at google­.search. And there may be hundreds more new top-level domains—the proposals now under review range from .aaa to .zulu. This expansion isn’t happening because we’re running out of unique Web addresses under the existing set of gTLDs. Far from it. It’s happening because the body in charge of these things—the Internet Corporation for Assigned Names and Numbers, or ICANN—thought it would be fun and profitable.

That may sound flip, but it’s the simplest explanation for the coming chaos. ICANN is a nonprofit organization that was created during the Clinton administration to, among other things, loosen the control that one company, Network Solutions, exercised over the domain registration process. But now ICANN is itself a monopoly. It has the freedom to mint new gTLDs and incorporate them into the “root zone file,” the master list that matches human-readable URLs (such as www.technologyreview.com) with the numeric Internet Protocol addresses that are used to route packets between computers. And it’s employing this freedom to orchestrate the biggest land rush in the history of the Internet.

During a four-and-a-half-month application period that closed on May 30, ICANN collected more than 1,900 proposals for new gTLDs. As expected, hundreds of companies applied for gTLDs corresponding to their brand names—.aetna, ­.barclays, .mcdonalds, and the like. But applicants also asked for the rights to hundreds of generic terms, such as .health, .mail, .music, and .pizza. The first such domains could be activated by next spring.

There is no general shortage of Web addresses. If there were, we might have seen businesses flocking to other new domains ICANN has already introduced over the past decade.

ICANN says it’s opening up these domains to promote competition and choice in the domain-name industry. But confusion and profiteering are the more likely results. Say you come across a URL like shoes.buy. How will you know who’s behind it? Amazon, Google, and three other companies have applied for control of the .buy domain; whoever gets it could sell subdomains to someone else. For example, Amazon could sell footwear at 
shoes.buy and charge the Gap for the rights to shirts.buy. Or say you’re planning to visit Hungary. Should you go to Budapest.hu, currently the city’s official site, or risk going to a new site like tourism.budapest and hope that you don’t stumble onto a phishing site in the process?

ICANN has built some limits into the program, including a seven-month objection period during which companies will be free to challenge aspiring trademark squatters. Even so, there’s a lot of money to be made now, starting with the fees that marketers, lawyers, and consultants familiar with the domain-name business have already begun to extract from big brands. Then there are the registration fees each new gTLD owner will be able to charge for access to the new domains. Corporations cite these fees as a major worry: they’re concerned that they’ll be forced to defensively register their brand names and related terms across hundreds of new domains.

Through all this, ICANN could also cash in. Plenty of names are in contention—11 companies applied for .home and another 11 for .inc, for example—and in many cases where the parties can’t settle their competing claims themselves, ICANN plans to hold auctions for the domains and pocket the proceeds. That’s on top of the $357 million in application fees that the Los Angeles–based organization has already collected, at a whopping $185,000 per domain. (The organization claims it needs all that money to pay for the evaluation of gTLD applications and prepare for potential litigation.)

What amazing new benefits will all this spending bring to consumers? None whatsoever, at least in the eyes of venture investor Esther Dyson, who served as chair of ICANN from its inception in 1998 until 2000. Dyson once supported the idea of allowing companies to create arbitrary top-level domains, but she says she came to believe that the change would be unnecessary and confusing for the public.

“I don’t think it’s illegal, but it’s wasteful,” she says. “One version of the future is: a lot of people spend a lot of money marketing [domain names], and a lot of new consultancies are created, and a lot of lawyers are very busy protecting and enforcing property rights, and there is no net benefit to anybody.”

It’s true that it’s getting harder to find a great .com domain name—hence the profusion of nonsense words like Xamarin as startup names. But there is no general shortage of Web addresses. If there were a genuine clamor for additional top-level domains, one might have expected to see businesses flocking to .biz, .info, .name, and the handful of other gTLDs that ICANN has introduced over the past decade.

Opening up the Internet to a flood of new gTLDs might also have the unintended consequence of making consumers even more likely to stop thinking about URLs at all and turn instead to Google, Bing, or Baidu to locate the sites they need. It’s already easy to surf the Web without ever typing a URL. As Dyson points out, “the search bar and the address bar are almost merging in browsers like Chrome.” As fewer and fewer consumers resort to direct navigation, owners of new gTLDs could soon find that they have “spent large amounts of money on something that is fundamentally worthless,” she says.

What amazing benefits will all this bring to ­consumers? None whatsoever, says Esther Dyson, who once chaired ICANN. She says the new plan is wasteful and ­unnecessary.

Bogged down

Who gave ICANN the power to make this mess? The U.S. Department of Commerce, which oversees the Internet Assigned Numbers Authority (IANA), the ultimate keeper of the root zone file. The story starts back in 1998, when Network Solutions, under contract to the U.S. military and the National Science Foundation, was still the sole registrar of new domain names. “It had become a giant business, and they were throwing their power around too much,” says Dyson. Under pressure from the global Internet community, the Commerce Department announced that it would hand over responsibility for the IANA—and with it, the power to coördinate the whole system of unique identifiers on the Internet—to a new nonprofit group representing stakeholders across the industry.

“The moment they said that, all hell broke loose, because everybody wanted to control that body,” Dyson recounts. The only person everyone trusted was Jon ­Postel, a researcher at the University of Southern California who had, up to then, been the IANA’s primary administrator. Dyson calls him a “saint” who was widely perceived as “the heart and soul of the Internet.” The coalition that formed around Postel took the name ICANN and won the contract to run the domain-name system. Dyson, who was seen as an impartial outsider, was asked to be its chair.

But the organization got off to a rocky start, she says: “We really didn’t know what we had gotten into, and were not too sensitive. None of which would have mattered if Jon Postel had not had heart surgery and up and died before our first board meeting.” Without Postel as its peacekeeper, ICANN had trouble raising money and was forced to survive on fees from Network Solutions and a new crop of domain-name registrars that grew up under ICANN’s oversight, Dyson says.

Thus was born a financial conflict of interest that continues to this day: ICANN subsists on the very industry it purports to govern. Dyson says she “lost any faith, over time,” in ICANN’s ability to regulate the domain-name business.

After years of planning, ICANN’s board voted in 2011 to move forward with the latest gTLD scheme. The Association of National Advertisers and other corporate lobbying groups oppose the program, which is why Dyson predicts that “the whole thing is going to be mired in litigation for a long, long time.”

Meanwhile, imagine the consequences if your city council proposed auctioning off street names to companies at $185,000 apiece. It would be a creative way to fatten the city’s coffers, and many companies might be interested. But others would feel forced to cough up preëmptively—IBM wouldn’t want an address on Microsoft Street, after all. On top of that, every resident would have to learn the new names, and every map and street sign would have to be changed. In short, it would be a loony idea with many hidden costs. Yet it is much like the scenario ICANN has penned for the Internet.

Wade Roush, Xconomy’s chief correspondent and editor of Xconomy San Francisco, is a former senior editor at Technology Review.