In 2009, three faculty in the College of Computing and Digital Media — Radha Jagadeesan (professor), Corin Pitcher (associate professor), and James Riely (associate professor) — received two grants from the National Science Foundation (NSF). Here, Jagadeesan talks about one of the team’s projects — a novel approach to computer security that focuses on systems accountability — and the subtle reasons their research matters.

 

Q: Why the topic “systems accountability”?

“For a long time, people asked ‘What are the good things a computer can do?’  The world of computing was a more cooperative, trusting place.  Of course, as the computer became ubiquitous, new questions emerged: ‘What are the bad things people can do with or to computers? And how do we prevent those behaviors?’

“That’s the essence of security: people will do whatever the system allows.  One software program will exploit security holes in another; sometimes, programs can inadvertently interact in a way that causes damage. So, we’re exploring possibilities not in security per se but in accountability: ‘How might a system detect whether the rules have been broken?’

“My colleagues and I liken systems accountability to traffic lights. At a red light, a driver will choose to stop. Is that because a camera captures everything? No, that much security isn’t necessary to enforce the rule. It’s enough to put the fear of getting caught into the driver’s mind.

“In a computer system, a lot of tracking would mean better enforcement, but too much creates problems in privacy and performance. Our work is filling a gap by asking the theoretical question: ‘What balance is both possible and optimal?’  Ideally, people know the rules and the computer lets them do what they want; but when they do something bad, the system should have the means to audit and catch them. 

“Our research is all about making people accountable for their actions.”

Q: What are some potential applications of your work?

"Scientists don’t solve problems; we’re really just hammers looking for nails.

“The NSF expects researchers to advance the state of knowledge. So, while the world is interested in finding answers, we’re interested in the techniques for asking questions — in this case, mathematical logic and semantics. These two views live together: through science, our understanding moves from the world of pure technique to the world of useful information.

“That said, when we ask ‘how should the universe be?’ occasionally useful things come out of that! I can imagine our work eventually being used in many areas. For example, in medicine, electronic records should make it easier to trace accountability in the provision of care. Or in business, accountability is important in enforcing legislation like Sarbanes-Oxley whose intent is to enable accounting transparency.”

Q: How does your research affect what goes on in the classroom?

“The real value in what we do is not in finding particular solutions but in thinking about things —and training other people to think about things — in a certain way. The things we teach in class are not the things we learned as students; in this field, everything has changed in the past 10 and 20 years. Our research keeps us in touch: we learn, we teach.”