HP's Open Innovation Strategy: Leveraging Academic Labs

When Rich Friedrich of HP Labs looks into the future, he sees desks used as 3-D displays, printers that automatically tailor a newspaper to a reader’s tastes, faster and more secure cloud computing servers, and wireless nano-sensor networks that monitor the environment.

But he also sees that achieving these technologies will require tapping into resources beyond HP’s own intellectual property. It will require an embrace of “open innovation,” the idea that companies should make wider use of ideas and technologies that come from other sources—and allow their own technologies and ideas to be adopted by others.

Toward that end, HP’s Innovation Research Program, now in its fourth year, gives grants of $50,000 to $75,000 to university researchers. Each grant can be renewed for up to three years. The company is reviewing proposals for this year’s round of grants.

“This is not innovation by doing contract research,” says Friedrich, director of the Open Innovation Office at HP Labs. “This really about bottom-up ideas and inspiration and trying to understand how to apply those.” Existing projects include research at the Technical University of Berlin into improved ways to process search queries in the cloud, work at Imperial College London into building nano rods to make new display devices, and data-mining research at Tsinghua University in China.

“This is not a program for incremental innovation, nor will it show up in next year’s products,” says Henry Chesbrough, director of the Center for Open Innovation at the Haas School of Business at the University of California, Berkeley. “These program areas HP identifies are explicitly for longer-term, five-plus-years-out time frames.” Chesbrough adds that this focus on long-term strategic value is unique among the companies he knows of that do open innovation. And, he says, HP’s program—which last year gave 65 grants to 49 institutions in 14 countries—reaches a far wider circle than most.

HP Labs focuses on eight broad themes, including cloud computing, digital printing, and sustainability. Within those areas, the Innovation Program seeks proposals on 26 topics. For instance, the Sustainable Ecosystems Research group is interested in better methods of handling the energy demands of data centers and applying information technology to a city’s infrastructure. So if a researcher has an idea for a mathematical model that could describe how server farms behave, or ways to use analytics to predict leaks in water pipes, he or she could win a grant.

The company would like to develop 3-D video that doesn’t require users to wear special glasses, so it’s looking for new ways to design camera systems and new algorithms to handle the images. Digital printing, obviously, is a big area for HP, so it’s seeking ideas to help computers figure out the meaning of text, photos, and video, then print customized newspapers. Another topic area involves the behavior of inks made of nanometer-sized bits of metal and polymers, which could find applications in pharmaceutical manufacturing as well as printing.

The idea, Friedrich explains, is neither to come up with a slightly improved version of an existing product nor to do completely blue-sky research with no obvious practical value. Rather, the office is trying to identify some long-term goals and figure out what it takes to reach them.

Along the way, HP researchers and grant recipients have coauthored about 200 journal papers. At least 21 invention disclosures, the first step toward securing a patent, have been filed. HP has all grant winners sign an agreement detailing how they’ll share any intellectual property that comes out of the research. “At some point I do expect some of these to have influence and impact on our products,” Friedrich says. “But we’re really engaged in things that go beyond the product road map.”

Of course, some of these projects might hit dead ends. “The goal is for us to push the envelope enough that we uncover the boundaries of what’s possible,” he says. “Some people might consider that a failure. I consider it the edge of knowledge.”

One grant recipient, Alan Willner, an electrical engineer at the University of Southern California, has had his grant renewed twice. He’s working on better ways to handle signal processing in optical interconnects on computer chips.

These interconnects use multiple wavelengths of light to shunt data between chips at higher speeds than is possible with wires. By using one beam of light to transmit many signals simultaneously, they’ll be able to handle the massive amounts of data involved in cloud computing, while drastically reducing energy consumption at the same time. HP aims to double the performance of such interconnects in 2012 and increase it 20-fold by 2017.

Neil Savage (www.neilsavage.com) is a freelance writer based in Lowell, Massachusetts. He has written for IEEE Spectrum, Discover, and Nature Photonics.