June 21-July 12, 2002
Back Issues
IN THIS ISSUE
U.Va. offers to share traffic costs
Job talk -- myths and realities
To the point -- with David Evans
Students create Rotunda mosaic

U.Va., World Wildlife Fund sign agreement

In Memoriam
Notable -- awards and achievements of faculty and staff
Dante Germino dies in train accident
Season tickets available
Employees: How to get back to school
After Hours -- Gayle Noble
Study discovers effects of exercise in women on HRT

To the point
with David Evans
SEAS-Computer Science

David Evans
Photo by Jenny Gerow
David Evans

David Evans, assistant professor of computer science, works to keep a sense of fun in his classes, even as his research tackles such serious challenges as code-breaking and software dependability.

He founded and directs the U.Va. RoboCup Team, which participates in international competitions that use a soccer simulation exercise to research group behavior. His students also enjoy the “Jeopardy”-style games and team competitions he uses at the end of the semester to review for exams.

“Dave is extremely innovative in using a multitude of techniques in his classes,” said John Stankovic, computer science department chair. “The students find this intriguing and challenging, and everyone benefits. Dave is a good example of the commitment to excellence our computer science faculty pursue in both teaching and research.”

Evans received a joint bachelor’s and master’s degree from Massachusetts Institute of Technology in 1994 and completed his doctorate at the MIT Laboratory for Computer Science in 1999. He joined the faculty of the School of Engineering and Applied Science the same year.

How do you keep students excited about computer science?

Computer science is both an exciting intellectual endeavor and an amazing tool. Most of computer science is about creating and solving puzzles, inventing and using languages, and designing and observing interactions. These are things that nearly all human beings are naturally drawn to, once they discover them.

Why aren’t more women interested in computer science?

Part of the reason is that computer science education — especially in our middle schools and high schools — focuses on math and video games, fields that women don’t generally find interesting. On the other hand, most of what’s interesting in the field of computer science has to do with language, logic, music and biology, areas that generally do interest women. Also, at U.Va. the College doesn’t offer a computer science major, so the classes offered by the Engineering School are filled mostly with engineering students, most of whom are already planning careers in the field. If the College offered a bachelor of arts degree in computer science, it would open up the field to many more people, especially women. The women who do go into computer science usually excel at it. I see it in my classes and in the many important contributions women have made to the field.

What is your research focus?

I’m particularly interested in computer security and swarm computing. Some of my work in computer security involves creating a software tool — “Splint” — that can identify security flaws in programs. I started working on it when I was a grad student at M.I.T., and my research group here is making it more powerful and easier to use. We’re also working to expand the tool’s scope. It’s good at detecting silly implementation problems that lead to a lot of security vulnerabilities. But detecting problems that relate to how the system is engineered around the concepts of trust and sharing is much harder. It’s not just a question of correcting errors in a few lines of code. We’re looking for a way to automate the process of catching higher-level design flaws.

Swarm computing sounds like it should relate to bees.

That’s where the concept comes from. Swarm computing is what nature does well — a group of bees form a swarm, a collection of cells form a complex organism. The individual bees or cells are not smart, but the system is smart. It’s the challenge of designing a system, made up of a collection of small and simple devices, harnessed together to solve global problems. We think we can learn a lot about how to build robust computer systems that behave predictably as they grow by looking at how nature does things. And we believe that building biologically-inspired computational models will help us learn more about biology.

—Charlotte Crystal


CURRENT ISSUE

© Copyright 2002 by the Rector and Visitors
of the University of Virginia

UVa Home Page UVa Events Calendar Top News UVa Home Page