The 2009 Creativity 50: Gary Flake

When Blaise Aguera y Arcas, an architect at Microsoft Live Labs, demonstrated Photosynth at the TED conference in 2007, the audience gasped, then burst into applause. The software is capable of stitching together virtual spaces out of unruly photosets, and at TED, Aguera y Arcas demonstrated a breathtaking walk-through of Notre Dame Cathedral made up entirely of images scraped from Flickr. As Gary Flake, director of Live Labs, is quick to point out, however, Photosynth was itself a synthesis that combined Seadragon—software developed by Aguera y Arcas and acquired by Microsoft—and research being done at the University of Washington. The mission of Live Labs is to create an environment in which such synthesis is possible, according to Flake, who observes "that new innovations are unexplored because they look infeasible or impossible from the vantage point of at least one discipline."

In addition to Photosynth, Live Labs has developed Thumbtack, an online bookmarking application, and Social Streams, an application for tracking buzz in social networks. Flake, meanwhile, has become a frank and vocal defender of Microsoft—the oft-maligned giant—which he says "is rediscovering a big part of what has made it historically successful." At TED, Aguera y Arcas joked that he never imagined he'd go to work for Microsoft, but then things like Photosynth have people taking another look at Redmond.

Flake, on fostering innovation: "For us, innovation means doing something valuable that isn't obviously valuable or obviously possible. Everything about the Live Labs environment is a consequence of starting with that definition and then backing out the requirements for what it takes to succeed. For example, if things aren't obvious—either in value or in execution—then we need to accept a certain amount of risk. When you tease apart all of the subtle relationships between innovation and repeatability, you ultimately end up with the realization that not only will mistakes happen, but that they are an essential part of the process. The trick is to make mistakes that are cheap, easy to recover from, and full of lessons. Our entire philosophy for how to make innovation repeatable boils down to celebrating failures as well as successes and learning from both."

Read more Creativity 50 profiles.