October 2001

A View from the
S&T Start-Up

Climbing Steep Learning and Growth Curves in Biotech and Pharmaceuticals

Broad Collaborations on Small-Scale Research

Playing a Role in Science Education Worldwide

Summer of Discovery

Two Computer Scientists Who Follow the Roads Less Traveled

Download PDF

Back to S&T Home

KEEP US INFORMED:
Please send us your comments on this issue, ideas for future issues, and news about your professional interests and accomplishments.

Al Dorof, Editor
adorof@brynmawr.edu
info@brynmawr.edu

© 2003

 

Bryn Mawr College
A quarterly newsletter on research, teaching, management, policy making and leadership in Science and Technology

Two Computer Scientists Who Follow the Roads Less Traveled
By Dorothy Wright

Rebecca Mercuri
Photo: Peter Olson

The roads less traveled are the ones that beckon two new tenure-track assistant professors in computer science at Bryn Mawr: Rebecca Mercuri and Douglas S. Blank.

Rebecca Mercuri became interested in the esoteric field of electronic voting in the late 1980s, when she was a committeewoman in Bucks County, Pa. "The county commissioners in Doylestown were considering purchasing voting machines," she recalls. "Researching the technology, I got to know a lot of the people involved in computer security, including the noted risk expert Peter Neumann, and I became convinced I didn’t want electronic voting machines for our area."

Only weeks before the 2000 presidential election became ensnared in controversy over recounts in Florida, Mercuri had defended her doctoral dissertation, "Electronic Voting Machine Checks & Balances" at the University of Pennsylvania’s School of Engineering. As one of a handful of experts on the subject, she was sought after for her views.

Cited in numerous publications from the Wall Street Journal to the Los Angeles Times, Mercuri’s opposition to exclusively electronic voting methods is now well known. "People can program a computer to do pretty much anything," she says. "But while we can ascertain that a computer may be doing something correctly, we can’t say it is not doing something undesirable. If we could, then we would be able to eradicate viruses."

To prove the point to her Bryn Mawr students, last spring Mercuri had a class create bogus voting machines, which looked as if they were accepting votes as intended by the citizens but tallied them inaccurately.

The Need for a Paper Trail

Mercuri is particularly concerned about Internet voting. "The Internet in its present configuration is a very dangerous way of running elections because anyone can tamper with it," she asserts. "We really need a paper trail to confirm an electronic vote. Otherwise we have no checks and balances."

Recently Mercuri testified before the House Science Committee on voting system standards and consulted for the General Accounting Office on Internet voting. She is writing detailed comments on a proposed new standard of the Federal Election Commission.

A self-described eclectic, Mercuri earned two bachelor’s degrees, one in classical guitar from the Philadelphia College of the Performing Arts (now the University of the Arts) and one in computer science at Pennsylvania State University. At Princeton’s David Sarnoff Research Center in the 1980s, she was among the few people working in the area of interactive video. And, as owner and senior staff consultant of Notable Software, she worked with Immaculata College to pioneer multimedia software for application in arts therapy. "I guess I’m always looking at the next latest, greatest," she says.

At Bryn Mawr, Mercuri serves as acting chair of computer science while Chair Deepak Kumar is on sabbatical. She is teaching an introduction to computer science section and a course in data structures, and she hopes to offer an upper-level course in digital multimedia. She’ll also work with three undergraduates to research common Java programming errors.

A champion of women in math and computer science, Mercuri is excited about teaching at Bryn Mawr. "Helping women prepare for what has become the career of the century is very important to me," she says. "It’s great to have the opportunity to do it at Bryn Mawr."

A Different Take on AI

Douglas S. Blank
Photo: Paola Tagliamonte

Like many students of his generation, Doug Blank was inspired to become a computer scientist by Gödel, Escher and Bach, the 1980 Pulitzer prize-winning book by Douglas R. Hofstadter, which explored the human mind and the development of artificial intelligence.

With a bachelor’s degree in anthropology from Indiana University and an interest in the problem-solving mechanism of evolution, Blank had intended to do graduate work in anthropology. "Then I read Hofstadter’s book about what it means to be a person and what it would take for a computer to duplicate those attributes," he recalls.

Returning to Indiana University, Blank earned a second bachelor’s degree in computer science and a joint doctoral degree in computer science and cognitive science, a relatively new field that comprises psychology, philosophy and computer science. "It’s a fascinating area," he observes.

Blank is one of the few computer scientists who have remained solidly in Hofstadter’s camp. "Hostadter believes the ability to perceive analogy is at the core of intelligence. A lot of people try to do an AI project concentrating on only one aspect of the problem, but Hofstadter would say you must incorporate all aspects of intelligence. I’ve taken that to heart."

Intelligence as Emergent

Learning is as important to intelligence as the perception of analogy, Blank believes. "I am working on a lot of the same practical problems as other people in AI, but I am doing them in a very different way, using evolutionary techniques and neural metaphors," he says. "I’d like to do the whole enchilada — to create an intelligent system or robot — and that’s going to take a long time."

Blank says his basic philosophy also differs from others in his field. "I don’t believe we will be able to program these systems at all," he says. "Consequently, we won’t understand the fundamental way in which they work. That is, there won’t be a way to abstract what they do or to develop a model for intelligence. Intelligence simply would be emergent — it would just ‘be.’ I think that’s true for the human brain as well."

In addition to sharing Kumar’s duties with Mercuri, Blank will teach introductory computer science and Bryn Mawr’s first cognitive science course.

Blank is optimistic about women’s potential in the traditionally male-dominated field of computer science. "As an all-women’s college, Bryn Mawr offers an opportunity to help women find their ‘spot’ in computer science," he says. "I really believe this is going to be a great program."

About the Author

Dorothy Wright contributes news and feature articles on science, technology, engineering and general interest topics to a variety of publications, including Civil Engineering, Engineering News Record and Bryn Mawr Now.

Back to Top