What does Sun Microsystems want to be when it grows up? That’s the question the company’s board of directors is probably asking newly appointed CEO Jonathan Schwartz, who this week replaced cofounder Scott McNealy. The former CEO’s mantra of “the network is the computer” fit nicely into the pages of business magazines – but never quite gave an identity to the company.
While most tech-savvy businesspeople and consumers have heard of the Santa Clara, CA-based firm, few could say what it actually does. Schwartz – who’s held seven positions within Sun since joining the company in 1996, most recently as president and chief operating office – hopes to rectify that.
Known outside the company for his technical acumen – as well as a ponytail and outspoken blog (he was one of the first high-level technology executives to start a public blog, in June 2004) – the 40-year-old Schwartz wasted little time laying out his vision for Sun’s future on April 24 during a conference call where he was introduced as the new CEO.
Schwartz expressed enthusiasm for a few specific technologies: open-source operating systems for large corporate computing systems, “grid” or “utility” computing, and the idea of selling computing time and power by the CPU-hour over the Internet. But he failed to bring up two keys areas in Sun’s past: semiconductors and microprocessor architecture.
First, though, to the passing of the reins. McNealy, who was CEO for 22 years, emphasized that it was his decision to turn the reins over to Schwartz, and the time was ripe for the handoff, given that Sun’s long-uncertain finances are stabilizing. The company has $4.2 billion in cash on hand and racked up revenue growth of 21 percent in the third quarter of its fiscal 2006. Schwartz did little to dispel that notion, saying observers should not expect any drastic changes in Sun’s strategy, and minimizing his personal and philosophical differences with McNealy. “We have different haircuts and sports preferences. He drinks Budweiser, I don’t,” quipped the new CEO.
When it comes to Sun’s role in the future of computing, Schwartz’s philosophy and strategy differ little from McNealy’s. Nonetheless, the protégé said it was time for a comprehensive review of Sun’s activities – which he said will mean “adding areas where we can grow and pruning areas that aren’t yielding return.”
Schwartz has already launched parallel “comprehensive reviews” of the companies’ product line, marketing spending, resource allocation, and financial planning. Although that process is an annual one at Sun, this year’s reviews will involve “a little more scrutiny” of areas where the company can grow, Schwartz says.
Asked to say which low-return areas might be pruned to make Sun’s focus clearer, the new CEO would not be specific. But he said the company would shy away from “lots and lots of individual products that don’t stitch together well.” As noted, his list of growth technologies did not include semiconductors or microprocessor architecture – a big focus for Sun over the last several years, as it has developed new generations of its UltraSparc server chips to compete with similar products from Intel, IBM, and Hewlett-Packard. The latest UltraSparc chips feature multiple processors to increase computing capacity.
Instead, growth will come mainly from the Internet, Schwartz said, which will continue to increase in power and scope spread “for as long as I’m on this earth.” The question for Sun, he says, is whether it can take advantage of that growth more effectively than its key competitors. “We’re not worried about demand,” he said. “We’re worried about intercepting that demand.”
Schwartz outlined four technology markets where he believes Sun can grow the fastest.
Corporate operating systems that need large data centers to operate database-driven websites will continue to be a driver, according to Schwartz. Sun’s Solaris operating system has been holding its own in this arena against Microsoft Windows and Red Hat Linux, and Schwartz believes that the company’s decision last year to make the code behind Solaris open source – like Linux – will attract a large community of software developers who will write more programs that run on Solaris or make Solaris servers more efficient. Already, developers have downloaded five million copies of open-source version of Solaris – ten times more than the company expected, Schwartz said.
Java, the mini-operating system, or “runtime environment,” that Sun has been adapting lately for use on mobile devices such as cell phones, will also continue to receive attention. A growing number of new cell phones carry Java, which allows users to run powerful software, such as video players and GPS navigation programs. Many of these applications interact with servers on the Internet, and work especially well with Sun servers running Solaris, creating a package of networking technologies that Sun can sell together. “Whether it’s handsets or Internet-protocol TV or automobile dashboards or ATM machines, there is a new world of network clients emerging, all of whom are delivering very strong demand for network infrastructure,” Schwartz says.
Privacy and the concept of “network identity” – a single technological standard for sharing and protecting Internet users’ personal information – are also crucial components of Schwartz’s vision. Through the Liberty Alliance, an industry consortium, Sun, Nokia, IBM, and other companies are working on specifications that would, for example, allow cell-phone users to authenticate their identities and transmit secure credit-card information when making Internet purchases using a cell phone. Regulations requiring such control over data, including the Health Insurance Portability and Accountability Act of 1996, will increase demand for Liberty-compliant operating systems, servers, storage platforms, and software development tools, according to Schwartz.
Finally, the new CEO underscored his interest in utility computing, which he sees as the Internet-age equivalent of the electrical grid. “Our belief is that all [Internet] technology is ultimately becoming a service,” says Schwartz. The company has launched the Sun Grid, a vast supercomputer comprised of hundreds of individual Sun Fire servers running Solaris. “For a buck per CPU-hour, [the Sun Grid] allows you – whether you’re modeling a protein or rendering a movie or doing a financial simulation – to purchase computing as a service rather than having to build out your own infrastructure,” he says.
That broad-stroke outline may help to sharpen the public’s image of Sun, which has lacked a clear identity. Microsoft has Windows and Office, IBM has database, server technology, and consulting, and Google and Yahoo have search and personal information services. Sun has a single slogan: “the network is the computer.” But that hasn’t led customers, investors, or journalists to see how the company’s products work together to achieve a distinctive take on information technology.
Schwartz will likely not be as quotable a helmsman as McNealy, but he has a reasoned plan, which may ultimately be as productive. “I’m a lot more verbose – that’s why I write a blog,” he says. “Where Scott has a one-liner, it takes me seven screenfuls to get to the same conclusion.” And now more people will probably be reading him.