Overdrive, page 18
Gates immediately went to work on damage control. He promised that China would be involved in the development of the next Chinese version of Windows. Then Charles Stevens, Microsoft’s Far East vice president, sacked much of the company’s China management team, and began courting Chinese software companies. Finally, Microsoft initiated talks with the Ministry of Electronics Industry. By the time Gates returned to China with his wife and his pal Buffett, Microsoft had all but locked up the Chinese market for Windows. The Chinese version of Windows 95 would become the standard.
While all this was going on, however, the world had begun moving to Internet time, where Microsoft did not set the standard. While Gates was visiting China in late March, his executives in Redmond were preparing for an important off-site retreat that Gates had called to discuss the Internet and how Microsoft should respond. The retreat was scheduled for April 5. On April 4, Jim Clark and Marc Andreessen formed a startup company in Mountain View, California, which, after some legal battles over its name, would be called Mosaic and finally Netscape Communications. Gates had nothing to fear from the merger of Novell and WordPerfect. The real threat would come from Netscape, a company that Microsoft took little notice of until it was almost too late.
On his last day on the job at Silicon Graphics, the company he had founded in Mountain View, California, James Clark sent an e-mail to a young inan who had recently arrived in the Silicon Valley after graduating from the University of Illinois at Urbana-Chnmpnign.“Any way that we might be able to collaborate would be of interest to me,” Clark wrote 22-year-old Marc Andreessen, whom he had found while clicking through a series of Mosaic pages on the Internet. In his e-mail, Clark explained that he was thinking of starting a new company and wanted to kick around some ideas. Could they meet and talk?
At the time, Clark knew next to nothing about the Internet or Mosaic. But he had heard of Andreessen and his work at the National Center for Supercomputing Applications (NCSA). And Andreessen had heard of Clark; in fact, he had been involved in an effort by IBM to take over Clark’s turf. In 1990, while a student at the university, Andreessen worked for two semesters for Big Blue in Austin, Texas, on a project to develop 3-D graphics, with the intent of wresting the market from Silicon Graphics. IBM’s product, however, turned out to be slower and more expensive than the machines produced by Silicon Graphics, and Andreessen returned to the University of Illinois to complete his undergraduate degree.
At the time Andreessen was completing his studies in December 1993, the Mosaic development team was dismantling. There was dissension within the team; some had begun to resent management when an article about Mosaic by New York Times reporter John Markoff on December 8 ran with a photo of only Larry Smarr, the center’s director, and Joseph Hardin, who was in charge of the center’s software development. None of the team who had worked so hard on Mosaic were in the picture.
Although Hardin had told Andreessen that he could stay on at the center after graduation, he also told him he could not continue working on Mosaic. Hardin thought the big blond kid from New Lisbon, Wisconsin, was claiming too much of the credit for Mosaic. Since Andreessen didn’t much care for Hardin or for Champaign, which he would later describe as mostly “corn fields and pigs,” when he graduated in December with a bachelor’s tierce in computer science, he didn’t stay around long enough even to pick up his diploma.
He headed for California and the Silicon Valley, where he had been recruited to work as a programmer for a Palo Alto company that was designing security software for Internet transactions. There he rented a cheap apartment, bought a red 1994 Ford Mustang, and went to work at Enterprise Integration Technologies, which is where Clark’s e-mail found him.
Clark was a self-made software millionaire who was looking for a second chance in life and hoped interactive television and the information highway were going to give it to him. Clark had grown up poor and fatherless in rural Plainview, Texas, a panhandle town halfway between Amarillo and Lubbock. In school, Clark was something of a troublemaker, and was suspended several times for such antics as taking a bottle of whiskey on a band trip. He dropped out of high school to join the Navy, where he changed his bad-boy ways when he discovered electronics and passed his high school equivalency test.
After his stint in the Navy, Clark went to Salt Lake City, where he completed a doctorate in computer science from the University of Utah, a greenhouse for talented undergrads like Clark who wanted to study computer graphics, a field then in its infancy. The department’s renown was due in large part to the arrival in 1966 of David Evans, a computer professor from the University of California at Berkeley. Evans, whose specialty was computer graphics, had attracted some very bright undergrads who would go on to make names for themselves in the computer industry. In addition to Clark, the group included John Warnock, Nolan Bushnell, and Alan Ashton.
At the time, there was no shortage of research money for the work being done by Evans and his students. The Pentagon was Very interested in the military applications of these developments and supplied a generous funding stream through the Pentagon’s Defense Advanced Research Projects Agency (DARPA), the outfit that in 1969 would fund a computer network linking universities and military research labs, a network soon to be known worldwide as the Internet.
In 1978, Clark took his talent and his PhD to Stanford University, where he joined the faculty as an assistant professor. Four years later, he left, to form Silicon Graphics. His goal was to create realistic 3-D graphic simulations on powerful workstation computers rather than on supercomputers. His company acquired chipmaker MIPS and began building the powerful computers that Clark had dreamed about. It didn’t take long for Hollywood to discover Silicon Graphics when in the late 1980s filmmakers began using the company’s technology to create lifelike special effects, from the water creature in The Abyss to the dinosaurs of Jurassic Park. Suddenly Silicon Graphics was hot; revenues were soaring.
But Clark was restless, and he squabbled with other company executives over the direction of the company. Clark wanted Silicon Graphics to develop low-cost hardware for the information highway. But he had met with strong objections when he promoted the deal with Time Warner for its interactive television project in Orlando, Florida. “I got tired of pushing against an immovable object,” Clark told Fortune. “I felt like I wasn’t having any influence.” In frustration, Clark resigned in early February 1994, and in the process walked away from $10 million in stock options. But the money didn’t matter. He still had a 1 percent stake in Silicon Graphics that was worth about $37 million.
When Clark sent his e-mail to Marc Andreessen, he envisioned a company that would produce software for the television set-top box, which Clark believed was the true consumer computer of the future. “There are $100 billion worth of companies out there wondering how they are going to use this data superhighway,” Clark told reporter Lee Gomes of the San Jose Mercury News when he announced he was leaving Silicon Graphics. “And if you develop relationships and help them connect to this network, there’s no reason why they shouldn’t want to give you money.’’
Clark knew that he would face competition from well- established companies, particularly Microsoft, which was making deals left and right to position itself on that data highway. Clark disliked Bill Gates intensely, and so was looking forward to the competition. “Everyone I’ve spoken to looks askance at Microsoft,” Clark told Gomes. “Why should I let them control this market, considering how they have used their power in the market they are in?”
After his initial e-mail, Clark invited Andreessen to his Atherton home, and to go sailing on Clark’s yacht. For the next month thereafter, they exchanged e-mail daily about the kind of company they should start. One possibility they discussed was an on-line network for the game company Nintendo. Andreessen even wrote a white paper outlining the project. He also suggested a company built around Mosaic, though even he wasn’t sure about the future of the browser. “The Internet was a toy then,” he would later tell Fortune. “That was when the whole interactive TV thing was peaking; one of those frenzies the industry indulges in when it loses sight of what’s happening in the real world. So I wasn’t sure about the browser.”
Clark’s vision of developing interactive software began to fade. He was now willing to take the gamble that the Internet would be the ticket. Initially, he and Andreessen called their start-up Electric Media, but when they learned there was another company with the same name, it was rechristened Mosaic Communications. Clark named Andreessen vice president for technology. They set up shop in an office in downtown Mountain View. Clark invested about $4 million of his own money, supplemented by several times that amount from venture capitalists with Kleiner Perkins Caufield & Byers.
Clark quickly went to work hiring the talent that they needed to turn their vision into a reality. He brought in several friends from Silicon Graphics, but the core of company was made up of several programmers from the Mosaic development team at the NCSA, whom Andreessen had alerted about the planned start-up.
Dave Thompson, a full-time employee of the center, recalled talking to Aleksander Totic shortly before Totic was hired by Clark. Totic had developed the versions of Mosaic for the PC and the Mac. “Aleks was telling me that Marc had plans for starting a company and that he was just waiting to be called. He didn’t know when that would be,” said Thompson. “A couple weeks later, Aleks was gone.”
Their most important hire was Eric Bina, the gifted programmer and full-time NCSA employee who had done much of the coding for Mosaic. Clark invited Bina to meet with him on his yacht. When Bina declined, Clark and Andreessen flew to Champaign to talk with him and several other members of the Mosaic development team. They all met at the University Inn. The full-court press worked. Clark hired five of them, including Bina and Totic.
On Tuesday, April 5, the day after Clark and Andreessen formed Mosaic Communications Corporation, two events occurred on opposite sides of the country that showed that Microsoft was still wavering as to what course they would chart along the information highway. In Kirkland, Washington, not far from Microsoft’s headquarters in Redmond, Bill Gates and his top managers met at a 1909 mansion for an Internet briefing by Steven Sinofsky, whose e-mail to Gates two months earlier prompted the meeting. Cross-country in New York City, the Penguin Group announced that it had purchased the rights to a book by Gates about the information highway for an advance in excess of $2.5 million—a book that would not yet embrace the Internet.
Gates was writing his still untitled book with Nathan Myhrvold and Peter Rinearson, a former reporter for the Seattle Times who had done some other writing for Microsoft. Gates had been shopping the book with New York publishing houses since early in March. He had no literary agent and had not met any publishers himself; instead, the negotiations were conducted by a team from Microsoft that included Jonathan Lazarus, Microsoft’s vice president of strategic relations. Typically, the Microsoft team played hardball. In a letter to about a dozen publishers who had expressed interest in the book, Gates had set down his terms. He demanded an advance of at least $2.5 million with a guaranteed first printing of 500,000 copies. He also expected the publisher to set aside $125,000 for marketing. In his letter, Gates wrote that he and Myhrvold planned to donate their share of the book royalties to charity.
This was not the way it worked in the book-publishing business: nonwriters or unknown authors didn’t set the terms; publishers did. Nevertheless, Bill Gates’s demands were met, although Peter Mayer, Penguin’s chief executive officer, said that only Gates’s “minimal requirements” had been met. The book was scheduled for publication later in the year.
Like Windows 95, the book would be late. By the time it was published in November 1995, the information highway envisioned by Gates and Myhrvold had changed significantly since they had set Microsoft on an ambitious course they believed would position it as the information highway leader. In a state-of-the-company memo, Gates had told employees that Microsoft was placing a “major bet” that the information highway would become a significant growth factor for Microsoft.
Although the proposed venture with TCI and Time Warner did not work out, Microsoft was continuing to make deals intended to take the company away from the desktop and into new markets. In March, Microsoft announced two projects with TCI. One was an interactive cable system capable of delivering a wide range of services, including movies and home shopping. It was to be tested on a few hundred TCI and Microsoft employees in Scuttle later in the year. The other joint project was a cable TV channel for computer users.
Despite such announcements, interactive TV was still more hype than fact. Time-Warner had recently stated that it had postponed its high-profile Full Service Network in Orlando, Florida, because the technology for the set-top boxes from Silicon Graphics wasn’t ready, proving that Gates was right when he predicted in 1993 that the industry was moving ahead too quickly, anticipating technology that was not yet in place. Microsoft’s own entry into the interactive arena, the project code-named Tiger, was still under development.
Microsoft, nevertheless, continued to scout new alliances that might give it an edge on the information highway. Gates had been talking with Robert Allen, chairman of American Telephone and Telegraph Company, about a possible joint venture, including interactive television. The two had been brought together by Craig McCaw, who was in the process of selling his Seattle-based cellular phone empire to AT&T for $12.6 billion. Gates and McCaw were good friends, and had been working secretly for several years on a project that was, literally, out of this world: a celestial information highway that would bring video, voice, and data transmission to millions of people in the most remote parts of the world—at a cost of about $9 billion and 840 satellites in low Earth orbits. In late March, Gates and McCaw, two of the country’s most accomplished entrepreneurs, announced that they had formed a third company, called Teledesic, to build the satellite network communications. McCaw Cellular owned 30 percent of the new company; Microsoft had invested $30 million for an 8.5 percent stake, and Gates and Paul Allen each put up another $10 million of their own money.
Teledesic was scheduled to be up and running by the year 2001. But the many technical and financial hurdles that had to be leaped made it all sound like pie in the sky. Besides, there was already a cheap and easy way to link the world and satisfy its craving for information: the Internet. Which was exactly what Steven Sinofsky was determined to make Gates and others understand at the company’s April 5 executive retreat.
It was an ironic coincidence that Microsoft’s first off-site conference on the Internet was held one day after the official incorporation of Mosaic Communications. No official announcement accompanied the formation of that company; in fact, it would be May before the first stories appeared in print that Clark and Andreessen were in business together, and that they had drafted some of the Mosaic development team from the NCSA for their own.
The off-site Internet conference had been organized shortly after Sinofsky returned from Cornell. Gates believed that his top staff focused best on a critical issue when they got away from the Microsoft campus for a day or two. This retreat, as it was called, was held at the Shumway Mansion, a four- story, 10,000-square-foot estate in nearby Kirkland.
All of Microsoft’s top executives attended the Internet conference, along with James Allard and Russell Siegelman. Sinofsky presented a 300-page report on the Internet and what Microsoft should be doing to respond. One executive who attended said, “It was a very good first step, and it got everyone thinking about the Internet. But frankly, I did not see how Microsoft was going to get much of a return on any investment it made with regard to the Internet. Where would the revenue streams come from? Bill asked the same question: How was the company going to make money?”
The executive staff did come up with a list of Internet- related projects that would at least get the company moving in the right direction. High on the list was Microsoft’s on-line service. Siegelman agreed that Internet communications protocols should be built into the service to enable users to access the Net. But he argued against making the service Web-based, which is what Allard wanted, because it would have meant a total redesign. Also on the list was a project to give users of Word, Microsoft’s best-selling word processing application, the ability to create documents for the World Wide Web.
Later in April, during his semiannual “think week” at his secluded second home on Hood Canal on the Olympic Peninsula, Gates spent much of his time thinking about the Internet and writing a memo titled “Internet Strategy and Technical Goals.” He sent the memo to Microsoft’s managers on April 16. “We want to and will invest resources to be a leader in Internet support,” Gates wrote. But he was not yet convinced that Microsoft should make an all-out push.
David Marquardt, a longtime member of Microsoft’s board and a general partner at Technology Venture Investors in Menlo Park, California, later told Business Week he did not understand why Microsoft was not making a bigger commitment to the Internet. He had asked Gates as much during an April board meeting. “His view was that the Internet was free,” Marquardt said. “There was no money to be made.”
The cash that fed the Microsoft monster came from its bread-and-butter programs like DOS and Windows, and from its applications like Word. The only danger to that revenue stream was if the Justice Department broke up the company, something Gates had vowed he would never let happen.
Trustbuster Ann Bingaman had not forgotten about billionaire Bill. Since the Justice Department had taken over the investigation from the FTC in August 1993, there had been a constant flow of newspaper and magazine articles speculating on the kind of action Bingaman might take against Microsoft, and when that action might come. But Bingaman had kept quiet, refusing to comment.
But as her staff painstakingly went about its investigation of Microsoft, there was growing concern among some in the Clinton administration that the probe might do damage to the country’s brightest high-tech star. Some very powerful people within the administration, according to sources, wanted the matter settled out of court. “There was a lot of concern about what was going on at Justice in the Microsoft case,” said a high- ranking official in the White House office of the U.S. Trade Representative. Some of those concerns, the official said, were being communicated quietly to Bingaman and her boss, Attorney General Janet Reno.
