Harvard Picks Up Pace In Computer Sciences

Last fall, Harvard's Division of Applied Sciences undertook an extensive survey of its roughly 300 undergraduate concentrators to discover whether they felt the division's computer science program made sense.

Currently, students who want to focus their study on computers and programming must concentrate in either Applied Mathematics or Engineering and Applied Sciences, two venerable options within their division with requirements that have come to be considered unwieldy for computer students.

Evidently, the students' answer was no, for more than half the students surveyed said they considered themselves primarily computer scientists, although the division does not officially recognize that label.

Because of the survey's results and the sense that the current applied requirements are too cumbersome for computer specialists, the division is contemplating creating a formal concentration before the Faculty later this spring.

"Computer science has come into its own as an intellectual discipline," says one of the concentration's leading proponents, McKay Professor of Computer Science Harry R Lewis '68.

At the same time, the field has also soared in popularity, both at Harvard and around the nation Lewis notes that in the past several years, the number of undergraduates focusing their studies on computer science has increased at an annual rate of about 20 percent.

Lewis and other computer science specialists at Harvard point to two main reasons for the growth of interest in the discipline the first is simply the fast-paced evolution of knowledge in the field.

"Some of the advanced courses I took as an undergraduate are now the beginning chapters of today's introductory textbooks, notes Nathan Goodman, assistant professor of Computer Science.

But Goodman, like other specialists, points to what seems an even greater reason for the field's spurt in popularity: the extraordinarily lucrative possibilities of employment in private industry. Starting salaries for qualified computer scientists in the private sector range up to $40,000.

"People can get jobs in computers after a bachelor's degree that compare to jobs they could get after graduate degrees in other fields," says Lewis.

But the financial allure of computer science has proved a double-edged sword for Harvard and other institutions. Combined with a relative shortage of Ph.D.'s in the field, it has made faculty recruitment at both the senior and junior level extremely difficult for almost all universities in the field.

This problem is especially critical at Harvard, which specialists assert fell behind in the field in the '60s and '70s.

In the early 1950s, recalls Dean of the Divi-

sion of Applied Sciences Paul C. Martin '52, three students were at the forefront of the nascent computer science discipline: the university of lllinois, the University of Pennsylvania and Harvard.

But all three institutions "blew it" he says, explaining that they failed to keep pace with the swift development in the field. Academic leadership in the discipline is said to have been transferred to such universities like Carnegie-Mellon Institute, MIT, and Stanford University.

"Computer scientists today are very, very choosy, and computer scientists are not impressed by Harvard's name like physicists," says one Ivy League professor who insists on anonymity.

Harvard "doesn't rank anywhere nearly as high as it ought to," says another, Yale computer scientist Alan J. Perlis. Harvard has "fumbled around" in the field for 20 years, Perils says, adding. "They do use out good people, but that's largely a consequence of the rest of the place."

But Perlis, like other scholars inside and outside the University, does add that Harvard has improved considerably in the last several years.

He specifically refers to a recent batch of widely heralded tenure appointments tat has given Harvard for the first time an important "Critical Mass" of theorists in the field. These theorists include professors of Computer Science Michael A. Rabin, Leslie G. Valiant, and Lewis.

Harvard computer scientists say the University also hopes to attract in the next several years a number of specialists in the non-theoretical side of the field-system, or programming.

But specialists say this task might prove trickier than building up a group of theorists because of the small size of Harvard's computer science faculty, which numbers four at the tenured level.

By comparison, institutions such as MIT or the University of California at Berkeley have 50 to 100 computer specialists on their faculties.

Since programming specialists often work together on broad "synthesis" projects, they may be discouraged from joining a group as small as Harvard's.

"It is possible to have a first-rate small theory group, but it is harder to develop a first rate small systems group," says Peter Elias, associate department head for computer science in MIT's Department of Electrical Engineering and Computer Science.

Harvard officials, however, stress that departments here are traditionally on the small side. Dean of the Faculty Henry Rosovsky for example, points to other disciplines such as physics, where Harvard's of specialists is comparatively small but very highly regarded