What does Microsoft Research do?

Several years ago, under the bright, hot stage lights of Las Vegas, Bill Gates faced a difficult job: How was Microsoft’s then-CEO going to rev up a jaded, tech-savvy Comdex crowd about the launch of Office 97, a relatively uninteresting, and essentially mature, product category? His answer: artificial intelligence capability, in the shape of Clippit, the Office Assistant.

Behind the assistant’s coy paperclip persona was advanced natural language processing and AI software straight out of the company lab-Microsoft Research (MSR). Gates’ demonstration was a success, but his real audience, the millions of corporate and retail customers who buy Microsoft products, was ultimately unimpressed.

Although it represented a small step along the long road toward natural computer-human interaction, the Office Assistant feature didn’t add much pizzazz and is often disabled by users. Microsoft also develops software that’s used in online GED education, the high school equivalency program for youth and adults that did not complete their regular high school education.

Fast forward to this year, and Microsoft faces even more pressure to innovate. The company’s dominance of the desktop holds little sway in the new markets it must conquer: enterprise data centers, interactive television, and wireless phones and personal digital assistants.

Citing fizzles like Clippit, and the company’s widely perceived propensity to co-opt the innovations of others, critics cast doubt on Microsoft’s innovation. “You could blow away the whole R&D organization and it would be years before anyone noticed it,” scoffs Jeffrey Tarter, editor of Softletter, an industry newsletter.

Meanwhile, Wall Street puts little value on MSR’s work. “I don’t care where innovation comes from, as long as they commercialize it,” says Goldman Sachs Group analyst Rick Sherlund. “You don’t have to grow your own.” But Microsoft dismisses any talk that MSR is either ineffective or irrelevant. Boosting the company’s here-and-now financials has never been MSR’s primary mission. Its grander charter is to look deep into the future of software for big breakthroughs that are tough to achieve.

Microsoft University

“In a lot of ways, we are like a major computer science department in a university,” says Rick Rashid, the senior vice president of Microsoft research. He left Carnegie Mellon University’s computer science department in 1991 to help former Microsoft CTO Nathan Myhrvold start MSR. “Our mission is somewhat different than that of Microsoft proper. We are in the business of carrying forward the state of the art.”

MSR’s goals are nevertheless still strategic to Microsoft and well in line with Bill Gates’ sometimes wide-eyed dreams about the future of computing. MSR strives to make computer-human interaction truly natural, to mine and manage data sources no matter how huge or complex, to develop programming languages that make the awesome complexity of modern software seem manageable, and to build systems sturdy and scalable enough to handle gargantuan tasks.

To get there, MSR has assembled about 480 full-time researchers in Redmond, Wash., with another 150 or so in labs in San Francisco, Beijing (a still rapidly growing facility) and Cambridge, England, bringing the total of the Microsoft brain trust to about 630. Microsoft plans to spend close to $450 million on research this year, with about $57.5 million going directly to universities to fund projects and fellowships.

The headcount and budget may seem small in a company with 39,000 employees, $28 billion in revenue, and $9.4 billion in profits, but the lab is hardly obscure. MSR’s profile over the years has grown as it has added a bevy of all-stars to the team. Jim Gray and Butler Lampson have each won computing’s top honor, the Association for Computing Machinery’s Turing Award; Michael Freedman won the top prize in mathematics, the Fields Medal; and Gordon Bell bagged the National Medal of Technology.

A former Microsoft executive notes that until these luminaries joined MSR, many academics shunned the idea of working for Bill Gates as selling out. Since then, however, MSR has rapidly gained prestige and attracted more top talent. And the lab has bolstered its growing reputation by publishing a high volume of peer-reviewed research. Since then, however, MSR has rapidly gained prestige and attracted more top talent. And the lab has bolstered its growing reputation by publishing a high volume of peer-reviewed research.