Eyeing IT spin-offs through an R&D lens

Are there any recent examples of proactive computing research that you've come across?

We're starting to see some of this with systems in IT for health care. You're starting to see a combination of machine vision and robotics to help with telesurgery. If the computer can anticipate the 10 or 100 things that you're about to do, that's a step forward.

My partners and I also are interested in using IT to improve energy efficiency. That fits with the use of the sensors. There are a lot of opportunities to improve efficiencies, particularly with our public infrastructure -- the highway system, water utilities, electric utilities.

We're increasingly interested in new medical diagnostic techniques. And somehow, I continue to be a sucker for new storage technologies. Most VCs have been burned sufficiently that they no longer want to invest in new storage ideas. I'm particularly interested in highly portable storage on laptops, handhelds, etc.

How did your scientific background help position you for your current job?

There's two parts to that. There's the technical background and having been in the work environment. Simply, I know what it's like to be a faculty member, and I remember what it's like to be an industrial researcher. You understand what's motivating them and what's motivating the business world. It's the experience of working in those technical environments.

Another aspect is having been involved in DARPA and Intel R&D. We touch on a lot of different technologies. So while there are a lot of things that can come up, chances are I have some familiarity with them. Not that I'm a know-it-all, but I'm at least aware of who else is doing work in a given area and what are the showstoppers to look out for. No matter what a researcher has come up with, chances are someone else has looked into this before.

What are some examples of recent research you've found interesting?

There's one company New Venture Partners has invested in that predated me, and they're working on sensor network nodes using low-power Wi-Fi. Most sensor network nodes use a new standard, [but] they're trying to figure out how to run Wi-Fi at a much lower power. The company is called GainSpan [an Intel spin-off in California], and it's an interesting project. It would have been tragic to me to just see that technology die.

There's another project that has been out there for a while called Planet Lab that I helped launch at Intel. It involves many companies and people worldwide. It's a system used right now by and for network researchers. There should be numerous commercial opportunities. I see that as a natural complement to grid computing.

I'm also interested in the autonomous vehicles used in the [DARPA] Urban Challenge. The vehicles themselves won't be used commercially in the near term. But a lot of the technologies could be -- i.e., how a vehicle deals with other traffic could be applied somewhere else.