New developments in the antitrust face-off between Microsoft and the U.S. Department of Justice keep on coming. On August 17, Bill Gates' company failed in its efforts to delay any more action in the case until the Supreme Court decides whether to consider Microsoft's request to dismiss the suit. That was bad news for the company, since the next major step would be to decide what "remedies" will be imposed. Then, on September 6, the DOJ announced that it would no longer seek a breakup of the company—and, more surprisingly, that it would drop its claim that Microsoft had illegally "bundled" separate programs. But the other charges remain, and it is clear that Microsoft's enemies will surely urge the court to impose every possible restriction on the company's ability to adapt to changing conditions—particularly the diminishing importance of the personal computer and the growth of Web-based computing.
It has been six years since Microsoft introduced Windows 95, the operating system that, by "bundling" itself with a Web browser, prompted the government's first antitrust suit against the company in 1997. Put another way, six years have gone by without Microsoft suffering any penalty for its supposed misconduct—unless, of course, you count the expenses and negative publicity it has incurred fighting the Justice Department. When Windows 95 debuted, Microsoft's critics and competitors made many predictions of the unpleasant things that would happen if the company kept doing business without new restraints. It's past time to see whether those predictions have come true.
It is also past time to take an even longer historical perspective: to look at the government's earlier adventures in antitrust and see how they compare with the Microsoft case. The results are very telling—not just with regard to Microsoft, but to antitrust law in general. Indeed, when one looks closely at the ground-breaking government actions taken against Standard Oil, the Aluminum Company of America, and AT&T, it becomes clear that something other than preventing harm to consumers—the stated goal of federal antitrust legislation—is the motivating force behind applying the law. Misinterpretation of these cases lies behind the claim that Microsoft, unless punished, crippled, or otherwise injured, will achieve a "chokehold on the Internet" or somehow undermine the entire computer industry.
What follows is a medley of what might be called antitrust's greatest hits and an analysis of how the lessons of history are being misapplied to the Microsoft case.
The Oil Standard
From 1906 to 1911, antitrust authorities prosecuted Standard Oil, a case that culminated with John D. Rockefeller's company being forcibly broken up into several smaller businesses. The Microsoft wars have often been compared to the Standard Oil case, and the analogy is apt—though not in the way it is usually intended.
Like Microsoft, Standard Oil was pilloried for practices considered legitimate when used by other companies. Since Standard Oil was such a high-volume customer, railroads gave it special discounts in exchange for planning shipments in ways that enabled railroads to use their lines and railcars most efficiently. Standard Oil's competitors complained bitterly about these discounts (called "rebates"), which the railroads kept secret from other oil companies.
Also like Microsoft, Standard Oil may have harmed its competitors, but it helped its consumers. Rockefeller's chemists developed 300 different byproducts from oil and created production and distribution processes far more efficient than those of other companies, allowing it to underprice them and to buy many of them out.
Standard Oil began in 1870, when kerosene cost 30 cents a gallon. By 1897, Rockefeller's scientists and managers had driven the price to under 6 cents per gallon, and many of his less-efficient competitors were out of business—including companies whose inferior grades of kerosene were prone to explosion and whose dangerous wares had depressed the demand for the product. Standard Oil did the same for petroleum: In a single decade, from 1880 to 1890, Rockefeller's consolidations helped drive petroleum prices down 61 percent while increasing output 393 percent. He eventually built Standard Oil of New Jersey into a trust composed of 18 companies operating under a single board of directors.
Standard Oil used resources with legendary efficiency, introducing many new labor-saving devices to its factories and locating sophisticated facilities at key points in its distribution system. Yet Rockefeller paid wages well above the market level, believing that high wages and good working conditions would save money in the long run by averting strikes and by encouraging loyalty among employees. Before Standard Oil revolutionized oil derivatives by lowering prices and improving quality, the high prices and limited supplies of whale oil and candles prevented all but the wealthy from being able to work or entertain after dark. Thanks to Standard Oil, families could illuminate their homes for just one cent per hour. And he saved the whales.
The federal government filed suit against Standard Oil in 1906 for violating the Sherman Antitrust Act, and in 1909, the company was found guilty; the Supreme Court affirmed the finding in 1911. Standard Oil, claimed the courts, evinced an "intent and purpose to exclude others"—demonstrated, ironically, by its many mergers, acquisitions, and business alliances. No one brought forward evidence of consumer harm, and the government never showed that Standard's specific actions, as opposed to its alleged intent, were illegal.
For several decades following the verdict, economists and legal scholars viewed the Standard Oil case as a classic example of "predatory pricing"—a monopolist's attempt to underprice its competitors out of the market so it could raise its prices later. In fact, just as the threat of new entry into the operating system, browser, and applications markets has kept Microsoft from ever exercising its supposed "monopoly power," so did new sources of competition keep Standard Oil from raising its prices. Neither the federal district court nor the U.S. Supreme Court found that Standard Oil's practices made kerosene prices higher than they otherwise would have been. If Microsoft Windows actually were a monopoly (that is, essential for anyone who wants to use a computer), the proper price would be about $900 a copy. Microsoft doesn't price this high because it knows that if it does, consumers will flock to Linux and Macintosh, and other companies would enter the operating system business, with products much cheaper than $900.
There's one more important parallel between the Standard Oil and Microsoft cases: Technological change made the Standard Oil decision obsolete by the time it was resolved. Of course, the Microsoft case hasn't resolved itself yet, but as we'll see, changing technologies are changing market conditions in the software world as well.
The oil business was opening fields in states such as Kan-sas, Oklahoma, Louisiana, California, and especially Texas, where Rockefeller had failed to invest. All those fields were far away from the Ohio/Pennsylvania/New Jersey corridor that was the base of Standard Oil's power. Also, the national kerosene market had declined, as home lighting shifted from kerosene lamps to coal-generated electricity and as fuel oil replaced coal and wood as the major fuel for home heating. In 1899, kerosene had accounted for 58 percent of all refined petroleum sales, and fuel oil for 15 percent. By 1914, kerosene had plunged to 25 percent, and fuel oil had risen to 48 percent.
Rockefeller was slow to switch from kerosene to gasoline, and with only 11 percent of the nation's oil production in 1911, Standard Oil could never hope to dominate the new market. Throughout the energy business, new technologies and new efficiencies were creating new and stronger competitors from industries previously distinct from the oil industry. Those competitors were far more powerful than the kerosene companies Rockefeller had defeated decades before.
Some observers have noted that in the years after Standard Oil was broken into smaller regional companies, the stock prices of those smaller companies rose, leading to speculation that breaking up Microsoft might have a similar positive effect on the total value of Microsoft stock. This is a misreading. Nearly all oil companies' stock went up in that period, not because of the breakup but because of rising demand and technological breakthroughs. Nor did the breakup have any discernible impact on oil production or oil prices.
The government's victory against Standard Oil had a long-term effect on the oil industry that is seldom discussed by those who see parallels with the Microsoft case. Only six years after losing the antitrust case, Standard Oil dramatically changed its attitude toward Washington, moving from hostility or avoidance to a very warm embrace. Company chief A.C. Bedford served as chairman of the War Services Committee, an agency created to mobilize the nation's supplies of gasoline and diesel fuel for military use during World War I. After the war, federal control never retreated, transforming what economist Dominick Armentano has called "a virtual textbook example of a free and competitive market" into "what had previously been unobtainable: a governmentally sanctioned cartel in oil." The legacies of this transformation include higher prices for consumers and the "energy crisis" of the 1970s. Deregulation in the 1980s finally restored some measure of competition to the industry.
The Standard Oil case teaches some important lessons about competition, innovation, and antitrust law. We see the difficulty antitrust has dealing with highly innovative companies. We witness the vagueness of antitrust law, which allows prosecution on the basis of alleged intent rather than specific actions. And we see how the Standard Oil case ultimately failed to benefit consumers or investors. Instead, it laid the groundwork for collusion between industry and government, bringing about many of the very ills the "progressive" proponents of antitrust said they were fighting.
Too Good to Be Allowed
In 1937, the U.S. government filed suit against the Aluminum Company of America, alleging over 100 violations of antitrust law. The government lost the case and appealed. The matter was finally decided eight years later, in 1945. This case is remarkable because it held that a company could be prosecuted under antitrust laws for being too efficient and responding too quickly to consumer demand.
The Aluminum Company of America (later Alcoa) grew out of the Pittsburgh Reduction Co., founded in 1887 by Charles Hall, the man who discovered and patented the technology for producing commercial quantities of aluminum. At the time, aluminum ingots sold for $5 a pound. By the time of the antitrust suit, the price was down to 22 cents per pound.
Alcoa dominated its industry from the start. It not only invented nearly all the tools and techniques required to lower production costs and raise the quality of the aluminum it produced, but also played a major role in creating markets for the new metal. While many companies entered the business of fabricating products out of aluminum and collecting and recycling used aluminum, none attempted to compete with Alcoa by producing virgin aluminum ingots. This was not because Alcoa restricted access to inputs such as electricity or aluminum bauxite, both of which the courts ruled were available to potential competitors in ample supply. Nor, by the time of the suit, did Alcoa deny others access to the manufacturing techniques it had patented: Those patents had expired in 1910. Alcoa was dominant because, as Armentano summarizes the situation, "users of ingot or sheet, and ultimately the consumers of fabricated products made from aluminum by Alcoa, were being served at degrees of excellence, prices, and profit rates that no one could equal or exceed."
The lower court found Alcoa innocent of all counts of anti-competitive behavior, even while acknowledging that it controlled 90 percent of the market for virgin aluminum ingot. (The other 10 percent was imports.) District Court Judge Francis G. Caffey reasoned that the Sherman Act forbade activity aimed at monopolizing markets, but did not outlaw the common business practices of companies that held dominant market shares due simply to the absence of competitors.
The appeals court agreed with Judge Caffey that the government had failed to show that Alcoa engaged in anti-competitive behavior or charged higher prices than it should. But Judge Learned Hand, writing for the majority of the federal Court of Appeals, held that Alcoa's de facto monopoly was itself a violation of antitrust law. Alcoa, he wrote, "insists that it never excluded competitors; but we can think of no more effective exclusion than progressively to embrace each new opportunity as it opened, and to face every newcomer with new capacity already geared into a great organization, having the advantage of experience, trade connection and the elite of personnel."
One is reminded of those police officers who sometimes pull over drivers late at night for moving at exactly the speed limit and staying in the middle of their lanes, on the grounds that this kind of careful conduct may be evidence of overcompensation by a drunken driver.
Having found no evidence of specific actions that were illegal, the court could hardly remedy the situation by restricting Alcoa's ongoing business practices. Nor, since the judges recognized the firm's outstanding efficiency and service to consumers, did it seem right to break up the company. Instead, the court settled for prohibiting the company from bidding for government aluminum plants which had been built to meet World War II military needs, and which were being sold off. Those assets were subsequently sold to Reynolds Metal and Kaiser Aluminum.
In 1948, Alcoa and the federal government asked the federal District Court for New York to reconsider the 1945 decision. Alcoa sought to be relieved of the scarlet M-for-monopoly that effectively criminalized its common business practices; the government, on the other hand, wanted to force Alcoa to divest some of its holdings. The district court, under the direction of a different judge than in 1937, once again found the government's case without merit, and this time ruled that Alcoa was not a monopolist.
A Real Monopolist
Besides Standard Oil, the case most touted by advocates of the Microsoft prosecution is the 1982 breakup of AT&T, which was overseen by federal judge Harold Greene. But while both cases involve information technology, there are important differences.
AT&T was indisputably a monopoly. From the beginning, the company lobbied for, and won, government protection against competition. It maintained its market share thanks partly to an array of legal prohibitions on other companies entering any part of the telephone services market, be it local or long-distance service—or even selling telephones and other equipment that could be attached to a phone line. The company's first president stated its strategy succinctly: "If there is to be state control and regulation, there should also be state protection to a corporation striving to serve the whole community…from aggressive competition which covers only that part which is profitable." Obviously, Microsoft has not called for similar protections from its competitors, nor is it today similarly protected.
Another difference: The AT&T divestiture undid acquisitions from decades before, in which AT&T had swallowed local phone operating companies. Microsoft, by contrast, has expanded primarily through internal growth. Because AT&T had capital and employees dispersed all over the United States to serve its customers, it could therefore divest itself relatively easily of the local telephone companies. These were then organized into seven "Baby Bells" to provide regional phone service. Microsoft, with its capital far more centralized and with much less need to have people "on the ground" in geographically defined areas (except for sales), would be far more adversely affected by such a legal order.
The settlement that led to the AT&T breakup also liberated the company from a 1956 antitrust consent decree that prevented it from entering and competing in non-regulated businesses, such as data processing. In exchange, AT&T voluntarily acceded to divestiture. Thus, the AT&T breakup was a consensual step toward deregulating a part of the economy that had long been regulated under the public utility model. A Microsoft breakup, by contrast, would represent a major increase in the government's intervention in this part of the economy.
At any rate, the AT&T breakup has been far from a complete success. One part of the agreement created a competitor in the long-distance market, free to introduce new technologies. This seems to have been relatively successful, with AT&T moving into cable, wireless, and other data transmission arenas and competing with a variety of businesses around the globe. (Of course, AT&T doesn't always compete successfully, as demonstrated by its huge stake in the floundering cable-modem system Excite@Home, which has been teetering on the verge of bankruptcy for most of this year.)
Much of the old AT&T was left behind as the local Bell companies, which were forbidden to manufacture telephone equipment or design new telephone products. The theory was that keeping these Baby Bells from equipment manufacture and design would prevent them from using their profits from local telephone service to subsidize new businesses. Instead, the arrangement created local phone monopolies that have been slow to innovate or to let competitors into their captive markets. Lucent, the technology company formed out of the breakup, is itself mired in financial and legal troubles.
Judge Greene's supervision of the telephone companies continued from 1982 until 1996, when an exasperated Congress finally dissolved the consent decree. In the intervening period, hundreds of applications for waivers—usually by local Bell companies wanting to sell or license a new technology—sat on Judge Greene's docket for an average of four years.
Antitrust is sometimes said to be superior to formal regulation, in that antitrust does not require continuing government oversight of the company's business. But the AT&T case demonstrates that enforcement of antitrust laws can generate as much or more intervention. Like the Standard Oil case, the AT&T case reveals a pattern of government control expanding over time, first to manage prices and avoid "unhealthy" competition, then approving and disapproving of mergers and acquisitions, and ultimately ruling on whether to allow innovations in products and services.
The Microsoft Panic
And Microsoft? If the assault on this company is to do more good than the partly successful breakup of AT&T—let alone the utterly unjustified wars on Standard Oil and Alcoa—then one would at the very least expect the suit's rationale to survive the passing of time. But it hasn't. More than half a decade after the first loud warnings about the awful world to come if Microsoft isn't stopped, the company's critics have been proven wrong at almost every turn.
In the year before the introduction of Windows 95, Microsoft announced it would start its own online service, to be called Microsoft Network (MSN). An icon for MSN would appear on the screen of every computer that shipped with Windows as an operating system; this was expected to be a huge advantage for gaining customers. At the time, Microsoft had a market share of exactly zero in the online services business. AOL promptly ran to the federal government to complain that Microsoft's plan was "anti-competitive." Technology journalist Steven Levy wrote an article in Newsweek warning that because of MSN, "One day, dollar bills may be replaced with Bill Dollars, and a piece of every online transaction could go through Microsoft's bulging coffers."
In Upside magazine, Gary Reback, Brian Arthur, and other devoted Microsoft critics wrote, "It is difficult to imagine that in an open society such as this one with multiple information sources, a single company could seize sufficient control of information transmission so as to constitute a threat to the underpinnings of free society. But such a scenario is a realistic (and perhaps probable) outcome." Business Week worried that Microsoft might "leverage" its operating system dominance to "corner" markets such as "networking, home software, and online services. In short, it might largely take control of the information superhighway."
Later, a group of Microsoft's competitors—Netscape, Oracle, Sun, and MCI—urged government action so that Microsoft would not "gain control of the Internet," arguing that suppressing Microsoft would "ensure the accessibility and affordability of information technology and the Internet." Netscape's Jim Clark offered a similar warning regarding Microsoft's Web browser, Internet Explorer: "If Microsoft owns the browser as well as the operating system, there will be no Yahoo!, no Infoseek, no Excite, just Bill standing at the gate, pointing out where he wants to go. Microsoft will be the one and only 'portal'." Sun's Scott McNealy fretted: "How are you going to compete if Microsoft won't put you on the Microsoft Shopping Center—which will be the opening screen of everyone's computer?"
Ohio Attorney General Betty Montgomery warned that unless Microsoft was stopped, it would turn the "information superhighway" into a "toll road." In 1997, the misnamed Council for a Competitive Electronic Marketplace warned that with Windows, Microsoft would be able to capture customers for online services for products such as insurance, banking, real estate, and local entertainment. A year later, an advocacy group called ProComp (which had been created to promote restrictions on Microsoft and is funded by Microsoft's business rivals) warned of "the very real potential that Microsoft will become virtually the sole gateway to the digital marketplace."
Similar warnings were made when Windows 98 made its debut with Channels (a soon-to-fail version of a "favorite links" list). As late as April 2000, after AOL announced it would choose Netscape as the AOL browser, the Department of Justice was warning that Microsoft might "add proprietary features to its Internet Explorer browser to tighten its control of the main on-ramp to the Internet for millions of consumers."
The government did not abolish MSN, nor did it suppress Channels, nor did it outlaw "bundling." While the pressure of the antitrust case may have forced Microsoft to stop enforcing some terms in contracts with some of its business partners, and may have distracted the company's leaders from producing new and better products, those setbacks were surely minor in light of Microsoft's supposedly immense market power. Microsoft's sinister power has had years to grow since the DOJ filed its suit. So what happened?
Windows 95 made its debut with the MSN icon intact, and MSN went on to become the most expensive failure in Microsoft's history. The network's content was weak, the interface was horrible, and the installation routine was lengthy and error-prone. Meanwhile, AOL made its interface better and better, and marketed itself incessantly through free sign-up disks and by paying computer manufacturers to include an AOL icon on the Windows desktop screen.
Microsoft Network no longer exists as an online service. It has been replaced with a free Web portal, similar to the Yahoo! or Excite portals. Microsoft's Internet service provider currently serves about 5 million customers. AOL has 35 million.
The fuss over Microsoft Network shows that antitrust action was not needed to maintain competition, even though MSN was on every desktop of every Windows 95 computer. MSN was an inferior product, so it failed. The same events illustrate the power of technological change to eliminate incipient monopolies. The growth of the Internet made online services much less important than they used to be.
Despite MSN's failure, however, allowing Microsoft to compete in the market for online services produced enormous benefits for consumers. When MSN was introduced, AOL was charging $54.20 for 20 hours of use a month. MSN was priced at $19.95 for that same amount of time. Thanks in part to the competition created by MSN, AOL eventually dropped its price to between $19.95 and $24.95 for unlimited use, and most other online services and Internet service providers followed suit. The same story of falling prices and rising usage has been repeated in virtually every area where Microsoft's entry was predicted to reduce competition and harm consumers.
After a brisk start, Microsoft sold its much-touted Sidewalk sites, which operated as local entertainment guides. Its real estate site, HomeAdvisor.com, trails Homestore.com and is being forced, like many other e-commerce sites, to reconfigure its business strategy. The Microsoft Expedia travel site was spun off into another company, and is now owned by USA Networks, not Microsoft. Microsoft's automobile Web site is doing pretty well, but hardly has a chokehold on its market.
But the fact that Microsoft neither dominates nor even still attempts to compete in various online businesses has not stopped Chicken Littles from warning that Microsoft's new XP operating system—scheduled for release in late October—will take over digital commerce.
Why No Monopoly?
One easy conclusion is that Microsoft's ownership of Windows and Internet Explorer is not enough to give it control of online commerce. Microsoft competes with traditional brick-and-mortar companies as well as Web sites, with other portals and online services that have millions of users, and with companies specializing in e-commerce. Even though Microsoft supplies the starting point for much Web surfing, the rest of the Net is just a click away.
More fundamentally, the idea that a Web browser could be used to control Internet content was hardly believable in the first place. One might as well believe that Sony would be able to control television programming if it sold 40 percent—or even 95 percent—of new television sets in the United States. A browser, like a television, is just a tool for reaching content. A television or a browser that interferes with access to content is, by definition, an inferior product. It is not going to have a viable economic future, much less become a market leader.
In 1997, Microsoft executive Nathan Myhrvold said the company wanted to get a "vig" (a bookie's share) of every Internet transaction that used Microsoft software. But this was unrealistic. The Internet vig was possible only in theory, not in the real world. Stanford economist Robert Hall offers the following scenario for what would happen if Microsoft made the attempt: "Yahoo! will ally with a manufacturer of cheap small computers and a national Internet service provider to produce an entire system that is Yahoo!-branded, defaults to the Yahoo! portal, but also provides access to the entire Internet with an open standard browser such as Netscape or Opera. The hardware would be cheap enough to be given away, like cellular phones or cable boxes, and all of the profit will be made from advertising, monthly fees, and transaction fees."
Bill Gates had hoped his company could at least make money from banks which used Microsoft software for online banking. That too failed, as banks ditched Microsoft networking software, and instead offered banking services via the World Wide Web.
What about Web servers—the computers that serve up the Internet's content to Web surfers? Could a company leverage a huge market share for its browsers into control of the market for servers? As it happens, one business tried to do precisely this. The company was Netscape, during its period of early dominance on the Web. But Netscape offered miserable support for developers and priced its product extremely high—thus creating an opportunity for Microsoft and other competitors. Today, the leading Web server software is Apache, a Unix-based program, which is free, and which is on 63 percent of servers. Microsoft's IIS is second, with 20 percent. Netscape's Enterprise has 7 percent.
Even if Microsoft achieves a high share in the server operating system market, it is likely to have little market power, because barriers to entry are low. Server software, including the operating system, carries out a limited range of functions. The software provides only the simplest user interface, which is the source of much of the complexity in full operating systems.
And what if Microsoft were the only browser company in the world? Could it then introduce a browser with Microsoft-only features and force the rest of the world to buy Microsoft server software, by making IE incompatible with every other company's Web server? An allegation to this effect was made in the spring of 2000 by the Department of Justice, although the court never heard evidence on the subject. According to the allegation, Internet Explorer included proprietary extensions of Kereboros (a security program that prevents hackers from entering a Web site) that work best with the Microsoft Web server.
The first practical obstacle to such a strategy is that the users of Internet Explorer would be cut off from any Web site that did not fall in line with Microsoft's program. This would be a major competitive defect, to say the least. Older versions of Internet Explorer and any remaining copies of other browsers on the market would still be able to gain access to those sites. An immediate market would emerge for new browsers able to reach Web sites that did not adopt Microsoft's server software. Major Web sites, particularly portals, would give away such browsers to ensure their sites could be reached. AOL, as owner of Netscape, would be in a particularly good position if Microsoft altered Inter-net Explorer to make it incompatible with AOL and other Web sites. Microsoft's hold on the browser market could never be strong enough to let it extract significant value from the server side, despite Microsoft's important roles in providing both a browser and server software. Individual users would not have to play a major role in opposing Microsoft. A few key Web sites—valued in the stock market at tens of billions of dollars—could do it on their own.
Because the Internet is still developing so rapidly, reporters and politicians are easy prey for manufactured panics. It would be much more difficult to create such a fright over a more familiar product, such as automobiles. Nobody would believe today that if General Motors opened its own chain of filling stations, GM would take over all American transportation. But on the Internet, folks who can't tell the male end of a dongle from a TCP stack are often suckers for silly claims about chokeholds.
Merely asserting that a company is a "monopolist" has allowed many of Microsoft's competitors to get a free ride from reporters and policymakers who ought to know better. For example, Jim Barksdale, then-CEO of Netscape, said this to Congress in 1998: "I was struck by the fact, in the response of Mr. Gates to the question about whether or not he was a monopoly, he talked about how short-lived the products were, and we all understand that. That doesn't negate whether or not it's a monopoly though. Even if it went away six months from now, it is a monopoly today." The hypocrisy of Barksdale's claim is astonishing—since Netscape's browser at its height held a larger market share than Microsoft ever had for Web browsers or for operating systems.
The scaremongers appear not to have suffered any loss in credibility. Steven Levy, the writer who warned that we'd all be using "Bill Dollars" by now because of the Microsoft Network, is still sharing his expertise with Newsweek's readers. In June 2000 he penned a cover story advising Bill Gates to capitulate to most of the government's demands. Similarly, Sun's Chairman Scott McNealy applauded Judge Thomas Penfield Jackson's Microsoft breakup order as a tool to "protect Internet technologies from becoming the proprietary presence of any one company."
A close look at antitrust's greatest hits—the cases of Standard Oil, Alcoa, and even AT&T—reveals a pattern of arbitrary rulings, disregard for consumers, and political interference with the administration of justice. The much shorter history of the Microsoft case has exposed the same injustices, along with the series of embarrassing exaggerations and falsehoods espoused by Microsoft's critics. Where are Microsoft Network, Channels, and Sidewalk today? All have disappeared, become irrelevant, or been radically transformed by competition and changing technology. The Internet remains free and decentralized, and for good reasons: Microsoft cannot "leverage" its dominance in a few markets into control over Internet access or content. To claim otherwise might sell newsmagazines or flummox congressmen—but it is hardly realistic.