By Steve Lohr
The New York Times
August 6, 1998
On April 6, 1994, William H. Gates summoned 20 employees of the Microsoft Corporation to a daylong retreat. The gathering included senior executives and a small group of younger workers who had become convinced that the Internet would revolutionize the computer software business.
Gates left no doubt that he agreed with the young evangelists, even though the Internet was not yet a household term or an industry. One person who attended the meeting recalled the Microsoft chairman saying, "We're going to make a big bet on the Internet."
In its sweeping antitrust suit against Microsoft, the Justice Department contends that the company's aggressive tactics in opening its deep pockets and wielding its market muscle in the Internet software business were intended to protect and expand its monopoly -- by crushing a new challenger, the Netscape Communications Corporation. But in a detailed reply to the Government's suit, which must be filed in Federal court no later than Monday, Microsoft will argue that the 1994 retreat, as well as other Microsoft discussions and documents dating to late 1993, show that the company's Internet plans were under way before Netscape rose to challenge Microsoft. The retreat occurred just two days after the founding of the Netscape. Microsoft allows that its tactics did eventually hurt Netscape, but it says that was a byproduct of its main intent, which was to improve its products and benefit consumers.
Few concrete plans were set that day in April 1994 at the Shumway Mansion, a three-story private residence near Seattle, in Kirkland, Wash., that had been converted into a conference center. But several participants say there was agreement on one strategy: that the Windows operating system, Microsoft's mainstay product, should have built-in access to the Internet. The concept, which was then still vague, was referred to variously as an Internet viewer, interface or browser.
"People walked away from that April 6 meeting knowing in broad terms what they had to do," recalled J. Allard, who is now general manager of Microsoft's Internet and data-access unit but then was the program manager for Internet activities.
The next day, on April 7, 1994, Gates told a group of corporate executives in Chicago that Microsoft planned to build Internet access into the next generation of Windows.
It was at the Kirkland gathering, Microsoft says, that it became a matter of corporate policy to put Internet browsing technology into Windows.
"The Justice Department can't be right," a senior member of Microsoft's legal team said this week. "Netscape wasn't even on anybody's radar screen back then."
The Government's case, to be sure, does not rest solely on Microsoft's intentions toward Netscape. The Justice Department says that Microsoft engaged in a series of illegal practices, including contracts with personal computer makers and others that were meant to thwart competition, and it accuses the company of eventually trying to persuade Netscape to divide up the market for Internet browsers -- all of which Microsoft denies.
A senior Justice Department official said this week that evidence in the Government's case showed a pattern of anti-competitive behavior by a company that was "simply hellbent on driving a competitor out of the market." Microsoft's defense, he says, is a "grand exercise in revisionist history." The Government's accusation that Microsoft's Internet software plans were designed not to improve its products or to benefit consumers but to destroy Netscape provides the context for the case, a context that seeks to explain Microsoft's actions.
"Intent is relevant because it provides a road map for the court to follow," said Stephen M. Axinn, a leading antitrust litigator in New York.
In court, Microsoft will argue that the Justice Department's road map is distorted. And interviews with current and former Microsoft executives and business partners, plus internal documents from 1993 and 1994, suggest that the company will at least be able to cast doubt on the Government's evidence of intent.
On Dec. 7, 1993, for example, nearly two years before Microsoft introduced its Windows 95 operating system, Steven A. Ballmer, the company's second in command, sent an e-mail message to his lieutenants after he had visited Harvard College and observed the students' use of the Internet. At the time, Windows 95 was code-named "Chicago." Microsoft, Ballmer wrote, could "really help popularize Chicago if we could say that Chicago is the greatest front end to the Internet."
The next day, Brad A. Silverberg, head of the Windows 95 product group,
replied to Ballmer: "I see a big opportunity here. Windows as the gateway to the information highway."
But the most informed thinking about the Internet was being done farther down in the Microsoft ranks. On Jan. 25, 1994, Allard, who was then 25, wrote a 16-page memorandum titled "Windows: The Next Killer Application on the Internet." In the memo, Allard wrote, "The Internet provides an incredible opportunity for Microsoft to effectively explore large-scale networks from many levels: customer needs, technical challenges, quality-of-service issues, electronic commerce and information-browsing technologies."
Mr. Allard's memo, which was widely circulated within Microsoft, also said the company should make putting Internet software into Windows 95 a priority. "When the next generation of Windows includes support for standard Internet protocols," he wrote, "the Internet becomes a natural extension of the Windows end-user experience, and the favored way to explore."
Mr. Allard emphasized that he was not talking only about putting internal Internet software in its operating system. Rather, he wrote, "Windows becomes the global infostructure explorer."
A day later, on Jan. 26, 1994, David Pollon, a young product manager, distributed an 11-page memo titled "Microsoft and the Internet: Strategic Direction." In it, he said that a key goal for the company was to "be an innovator in providing the easiest-to-use Internet interfaces." The interface he described was a primitive concept of a browser, which he argued Microsoft should make part of its Windows operating system. "PC users," he wrote, "will adopt Windows as their easy-to-use interface to the Internet."
Netscape had not yet been founded and thus was not mentioned in either of these e-mail memos. But other rivals were mentioned, including Sun Microsystems and other companies whose hub computers, or servers, ran on the rival Unix operating system -- the technology on which Internet pioneers had built its infrastructure and the favored platform for developing Internet software.
Today, many of Microsoft's rivals fear that it is using its market power to take over the lucrative market for software that runs on Internet servers, the powerful computers that individual users tap into from their desktops to send and receive e-mail, view World Wide Web pages, transfer files and download audio or video programming. Microsoft's industrial-strength operating system for this market is Windows NT. Microsoft's practices to increase the use of Windows NT are not part of the Government's antitrust case, but the Justice Department is investigating them.
Yet back in January 1994, based on Pollon's memo, the Windows NT strategy for the Internet was already taking shape. Among the goals he recommended was "supplanting Unix with Windows NT as the dominant back-end platform for Internet services."
In the antitrust case, Microsoft is expected to argue that it certainly did use every means -- every legal means, the company says -- at its disposal to try to defeat Netscape, the early leader and its main rival in the browser market.
But Microsoft will also argue that its powerful push into the Internet software market -- especially incorporating the Internet Explorer browser in the operating system -- also made it easier and less costly for many millions of customers who use Windows-based personal computers to participate in the global computer network. And the nation's antitrust laws, Microsoft's lawyers say, are intended mainly to protect consumers from practices like price gouging rather than to protect competitors.
The senior Justice Department official says, however, that Microsoft's version of events ignores some significant details. For example, he notes that Microsoft did not license the core browser technology around which it built Internet Explorer until December 1994, eight months after Netscape was founded.
To hasten its entry into the browser market, Microsoft licensed its early browsing software from Spyglass Inc. of Naperville, Ill. The first meeting between the two companies was in April 1994, recalled Douglas P. Colbeth, the president of Spyglass. In subsequent meetings later that year, Colbeth said it was becoming clear that Microsoft felt a sense of urgency about Internet software.
Netscape achieved celebrity more than a year later, in August 1995, when it went public and many of its employees became instant millionaires thanks to Wall Street's discovery of the Internet. But within the software industry, Netscape had become known as a hot start-up in 1994.
"All the attention Netscape was getting clearly prodded Microsoft and accelerated its timing for some steps, like its announcement in 1995 that it would give away its browser," Colbeth said. "But from our earliest discussions with Microsoft, the message was loud and clear that the browsing functionality was going to be put into the Windows operating system."
He added: "Whenever you license technology to Microsoft, you have to understand that they may well do it themselves and drop it into the operating system. That's Microsoft's business model, expanding Windows."
Regardless of the legal outcome, the way Microsoft awakened to the opportunity and threat of the Internet is a textbook case of the way the company works. A small group of Internet enthusiasts, mainly in their 20's,got the attention of Microsoft's top management including Gates.
It helped that one of those young enthusiasts was Steven Sinofsky, who at the time was a technical assistant to Gates, a job that Sinofsky described as "largely just being an extra set of eyes and ears to learn about the newest, latest technology and bring it back to Bill."
Mr. Sinofsky had read Mr. Allard's January 1994 memo, "Windows: The Next Killer Application on the Internet." And the two soon got together. "It was clear he had already figured this thing out," Sinofsky recalled.
Mr. Sinofsky and Mr. Allard arranged the one-day brainstorming retreat in April 1994, including handing out a three-inch-thick pile of advance reading for every participant, a hefty selection of Internet-related writings. Much of the discussion focused on how Microsoft might make money off the Internet and what might be the big money-making product, or "killer app," in the industry's shorthand.
Mr. Allard's formulation, then and now, was simple -- both a distillation of Microsoft's overarching strategy and a business practice that sends shivers of fear through its rivals.
©Copyright 1998 The New York Times; Reprinted by permission.