This article is published in the November 2009 issue.

Breaking the Image

Musings from the Chair

For many institutions, early fall features Family Weekend events, when parents and families return to campus to visit their children and hear about research and educational activities occurring within the department. Especially for parents of first-year students, it is also an opportunity to hear about directions and opportunities in the field, as their children make decisions on a degree major.

Unfortunately, it has also evolved into an event at which departments must deal with dispelling myths and rumors. From parents, I regularly hear variations on questions like: “Aren’t all the jobs in computer science going overseas to India or China?” (This one comes up perhaps less now than in previous years, but still recurs.) Or “My daughter is really interested in providing better health care for everyone, and I have been telling her that computer science has nothing to contribute.” And from students, I regularly hear variations on questions like: “I really like computing, but I don’t want to end up stuck in a job like Dilbert™ or working with people like Dilbert™.” 0r “Don’t you have to have been hacking since you were six if you want to major in computer science?”

It is frustrating that as department heads we still have to fend off these misperceptions. In response, I often point parents to recent U.S. Bureau of Labor Statistics studies, showing that computer-related sectors have job growths faster than any other sector and nearly three times the general growth rate. I also tell them that starting salaries for graduates of my department are higher than any other department at my institution (and much higher than biology-related disciplines). I point out that roughly one-third of the faculty members in my department have significant research endeavors that intersect the life and medical sciences. And I try to talk about the huge range of opportunities to leverage computational modes of thought in other disciplines—medicine, finance, patent law, management, transportation, environment, energy, and social networks. But the fact that I have these discussions suggests that as a field we still have an image problem.

So how do we break this image problem? Clearly there already are efforts at the national level to tackle it. By coalescing viewpoints expressed by many within the field, Jeannette Wing at NSF has been championing the notion of “computational thinking” as “a universally applicable attitude and skill set everyone, not just computer scientists, would be eager to learn and use.” The Computing Community Consortium (CCC) has been engaged in a range of activities aimed at answering questions like: “What attracts the best and brightest minds of a new generation? What are the next big computing ideas, the ones that will define the future of computing, galvanize the very best students, and catalyze research investment and public support?”

Through events like the “Computing Research that Changed the World” symposium, CCC helps to inform policy makers about the game-changing impact of computing research on the world. A recent CACM article by Cameron Wilson of ACM and CRA’s Peter Harsha clearly outlines the challenges and activities aimed at radically revamping computer science education in the K-12 arena, in part with the aim of better communicating to future students the great opportunities encompassed by computer science research and development. Moshe Vardi’s forthcoming column in CACM provides an insight into our image issue. All are worth considering as we think about our image.

These are excellent starts to challenging the image problem. But we can’t rely on “fixing the problem” from the top down, through policy changes by funding agencies or educational bodies. We also have to examine our own individual institutions. Many of us have started taking very serious looks at our curriculum. Are we showing the many connections between computer science and other disciplines? Can we draw on examples from the life sciences, from finance, from modern engineering challenges, to motivate and frame the discussion of our basic tools and techniques? Are we willing to address the question of whether every computer science student must have a foundational grounding in all aspects of computer science (architecture, operating systems, networks, security, algorithms, complexity, and so on), or are we willing to provide students with some flexibility in designing a curricular path that fits their goals and interests? (Perhaps, if we really believe in abstraction, we would be willing to accept that a student can succeed in computer science without having to know all of the details of the computational engine?)

A number of institutions have already tackled these challenges, and there are several excellent examples available against which institutions could match their own curricular needs. Examples (among many) include Georgia Tech’s Threads approach, Cornell’s curricular redesign to provide computation-based degrees in Engineering, Arts & Sciences, and Agriculture & Life Sciences, Stanford’s new curriculum based around tracks, and MIT’s streams approach.

To demonstrate the breadth of impact that computation can have, to attract today’s students to the field, and to flexibly support their career paths, we need a bottom-up effort in which departments carefully examine and evolve their curricular offerings. We expect that this issue will be a focus of this year’s Snowbird Conference. And hopefully in a few years we won’t be fielding those annoying questions about why students should major in computer science.

Eric Grimson who recently became Chair of CRA’s Board of Directors, is the Bernard Gordon Professor of Medical Engineering and head of the Electrical Engineering and Computer Science department at MIT.

Breaking the Image