Rising and SettingBy Eric Nee | Posted 11-01-2004
From the beginning, Sun Microsystems has seen itself as the youthful renegade, the irreverent outsider taking on the establishment. Chief Executive Scott McNealy relishes his role as the bad boy of the industry, challenging conventional technology, conventional wisdom, even conventional behavior.
During Sun's early years, this iconoclastic approach worked in the company's favor. Sun was one of the first companies to embrace disruptive technologies such as the microprocessor, Unix and the Internet, each of which helped it achieve years of rapid and profitable growth, often at the expense of larger competitors. But the world is changing, and rather than changing with it, like IBM, Dell and Hewlett-Packard are, Sun is resisting the future. The company that prided itself on being the young upstart now finds itself in the uncomfortable position of being the last of the old-style computer firms.
The rest of the worldfrom computer vendors and software developers to corporate userslong ago started shifting away from proprietary computing platforms to industry-standard platforms based on Intel and AMD processors, and Microsoft and Linux operating systems. This has helped vendors, developers and users save money by simplifying computing environments and increasing compatibility. Sun recently began offering industry-standard platforms as well, but it still touts computers built around its own SPARC processor and Solaris operating system as the superior technology. But now Sun's stubbornness is losing the company salesand money.
Unless Sun makes radical changes, fast, it is likely to meet the same fate as any number of once innovative companies before it. Sun may hang around in some diminished capacity, like Wang (now Getronics) or Silicon Graphics, or it may simply vanish into the bowels of a larger firm, like Tandem and Digital Equipment, both of which ended up as part of Hewlett-Packard. Whatever the future holds, it is all but certain that Sun will never return to the prominent position it once held.
"Sun is where IBM was in 1990 when it got most of its profits from mainframes," says Marco Iansiti, the David Sarnoff Professor of Business Administration at Harvard Business School. "Sun remains one of the most vertically integrated companies in the industry." In fact, Sun now holds the dubious distinction of being the only company still attempting to offer a broad line of proprietary computer systems, from the smallest desktop (the Sun Ray 100 Ultra-Thin Client) to the most powerful server (the Sun Fire E25K).
Sun has made some belated adjustments to its "I'll do it all myself" approach to developing computers. The company now offers Solaris on computers using AMD's Opteron processor. And for companies standardizing on Linux, Sun is offering that as well. But it remains as focused as ever on computers made from SPARC and Solaris, believing that it can out-innovate Microsoft, Intel, IBM, Dell and the rest of the competition, and provide superior products.
Even today, Sun brags on its Web site that "The SPARC architecture is a key component of the intellectual property portfolio that differentiates Sun from the system repackagers. Sun is making strategic and dramatic investments in the future as evidenced by the size and experience of our microprocessor design team which is 1,400 engineers strong and includes some of the brightest chip designers in the world." And even more engineers are busy updating Solaris.
The result of Sun's intransigence is that the company continues to spend a large portion of its $1.9 billion in annual R&D on its proprietary processor and operating system, long after customers have made it clear that they prefer industry standard platforms. Sun would be much better off taking the money it now spends on SPARC and Solaris and devoting it instead to emerging technologies such as utility computing, where it already has efforts underway.
Sun has proven in the past that it is capable of developing breakthrough products in new areas. Think about Java, a technology that helps make it easier to develop software that works across a wide variety of computing platforms. Or Network File System (NFS), which helped make it easier for disparate computer systems to exchange files.
Sun claims that all of its R&D spending helps the company build better computer systems than it could if it relied on other companies for processors and operating systems. One could argue that point all day. What matters is whether IT departments buy the argument, and in Sun's case, they don't. The company's product revenues have dropped steadily, from $9 billion in 2002 to $7.4 billion in 2004, and the end is not in sight.
"When you've been as successful as Sun has, it's hard to change," says Iansiti, who recently co-authored a new book from Harvard Business School Press titled The Keystone Advantage: What the New Dynamics of Business Ecosystems Mean for Strategy, Innovation, and Sustainability. But change is possible. Other technology companies have gone through wrenching changes. Intel was once the largest supplier of memory chips, but it abandoned the market to focus on microprocessors when the Japanese proved to be formidable competition. IBM was the largest provider of proprietary operating systems, but was the first to embrace the Linux operating system. H-P invested billions of dollars in its own PA microprocessors, but shifted its support to the Intel family. Each of these companies was willing to tear up its game plan.
Sun needs to do the same, because the vertically integrated model of creating computers is dead. The world has moved on to a new model of innovation, one in which different companies contribute interlocking pieces of technology to the entire computing platform. Rather than trying to do it all, each company focuses on what it does best. Intel and AMD concentrate on processors, Dell, IBM and H-P focus on computer systems, and Microsoft (and the Linux community) on operating systems.
This new model of doing business isn't just for the technology industry, it's one that all types of industries are embracing. Retailing, banking, even automakersindustries long known for being dominated by vertically integrated firmsare moving toward this new model. Take a close look at the innards of Ford Motor's new 2005 Ford Mustang. The climate-control system, sound system, power-steering system, instrument panel, rear axle and drive shaft are all designed and made by Visteon. The Jeep Grand Cherokee, Chrysler Voyager, 9-3 Saab convertible, and Mercedes-Benz G-class and E-class vehicles are all assembled under one roof in Graz, Austria, but by Magna International, not by DaimlerChrysler, and not by General Motors (which owns Saab).
The move away from vertical integration and toward business ecosystems has been going on for some time, but it has accelerated in recent years. The globalization of business, advances in computer technology, and the widespread use of the Internet have made it easier, and more imperative than ever for companies to focus on what they do best rather than trying to do it all. "In the past it was an option whether a company wanted to participate [in an ecosystem]," says Iansiti. "Today it's a requirement." Let's hope someone over at Sun is listening.
Eric Nee, a longtime observer of Silicon Valley, has served in a variety of editorial positions at Forbes, Fortune and Upside magazines. His next column will appear in January.