Follow us on Instagram
Try our daily mini crossword
Play our latest news quiz
Download our new app on iOS/Android!

Q&A: David Dobkin

David Dobkin, chair of the computer science department, will take on a new role as the University's dean of the faculty starting July 1.

'Prince' reporter Josh Brodie asked Dobkin about his past research.

ADVERTISEMENT

'Prince': About the time when you left grad school, computer science was a nascent field. Tell me a little bit of what it was like to be there at the beginning.

Dobkin: This wasn't quite the beginning, but it was close enough that from this perspective it felt like the beginning. I was a theoretical computer scientist, which means that I wasn't implementing anything. I probably didn't even have a computer on my desk . . .

But people were looking at all flavors of problems one would do on a computer and analyzing them theoretically to come up with good algorithms to solve them. Probably five years or ten years before my time, people would spend an afternoon inventing a concept that has now affected every computer that came since. It was exciting, and who knew where the field was going to lead. It was sort of an interesting time, because here we were at the relative beginning of a new field which could have turned into something or could have turned into nothing.

P: So you took the plunge into computer science. Is that what brought you to Princeton?

D: No, I went and became an assistant professor at Yale in 1973.

I came in and started by teaching courses on the more theoretical side of the computer science department: courses in algorithms. It was partly related to what I had done in my thesis, and partly I was learning as I went. After a few years there they asked me to teach the beginning courses. I used that as an opportunity to learn basic computer science. Initially I was teaching the beginning courses for non-majors and then I worked my way up to knowing enough computer science to teach the beginning courses for majors and used that as a lever to actually learn some computer science.

ADVERTISEMENT

By the time I left Yale I probably had a sophomore level of understanding of computer science.

P: Where did you go from there?

D: I left Yale in 1978 swearing to never live on the east coast again and never be in an Ivy League school again and to just be away from all this. So I went to the University of Arizona and lived in the desert.

Arizona, at the time, was a quite practical department so there were a lot of people I could learn from, because they really knew things different from what I did; I was really the first theorist to come into the department.

Subscribe
Get the best of ‘the Prince’ delivered straight to your inbox. Subscribe now »

During my second year there, they asked me to be acting chair . . . During that time I tried to recruit a friend of mine who was on the faculty at Berkeley. While I was in the process of recruiting him, Princeton also started to recruit him. So he said to me, "Instead of you recruiting me, why don't I recruit you?" So he came to Princeton in 1980 and convinced me to come in 1981 . . .

When I left Arizona I was on sabbatical and I spent a semester at Xerox Parc. This was at the height of Xerox Parc . . . That was the first time I saw what would qualify as a personal computer, the first time I saw a laser printer, the first time I saw an ethernet, the first time I saw networking. Many, many things existed there that hadn't existed anywhere I had lived before. Xerox Parc was the first time I saw the Arpanet, which has since evolved into the Internet . . .

When I came in '81, by then I had reinvented myself as someone in computer graphics. One of the research themes I had picked up in later years at Yale and at Arizona was the field of computational geometry — a more theoretical study of the kind of things that go behind doing geometry on a computer, geometric algorithms. The logical application of this was to computer graphics . . . When I came to Princeton I started teaching a graphics course. There was a certain amount of learning as I went. I think at that stage I had never taken a course that I taught — often I was learning the course as I taught it to the poor students who didn't want to get too many weeks ahead of me.

P: What first drew you to graphics?

D: Some combination of midlife crisis and other things. Here I had been doing theoretical things where I was saying the application was computer graphics, and it seemed it was time to learn computer graphics to see if that was actually true. Even today when you look at the computer graphics group here, I am on the more theoretical side of the people in the group. But if you look at theoretical computer science or computational geometry, I'm very much on the practical side.

It's been a bit of a progression throughout from pure math to applied math to computer science and within computer science from more theoretical to more practical.

P: How has your work changed? What are you working on now?

D: As things evolved, I got more interested in actually building systems and doing things that would allow me to implement and test ideas rather than just developing ideas . . . One of the applications we looked at was an application to morphing: If I have a three-dimensional mesh that represents your head and a three-dimensional mesh that represents my head and I want to morph from one to the other and do it smoothly, people know how to do that in 2D where you're just dealing with pixels, but in 3D it gets a little bit more complicated because your head could be turning . . .

A simple application is to be able to take your hair and smoothly fit it to my head so I look like I have hair.

The other project I've worked on is a project building a search engine on the web for three-dimensional shapes.

P: This past fall you taught a freshman seminar with sociology professor Paul DiMaggio called Sex Money and Rock and Roll: Information Technology and Society. How did you come to teach that course?

D: One of the things that I started when I became chair of the department was a move to move computer science to a more central role in the University. It seemed to me technology, particularly information technology, was having more and more of an impact on all sorts of aspects of society and all sorts of aspects of academia. One natural place to look was in policy issues.

We're developing this technology, is anybody looking at its impact on society, is anybody looking at how it changes things, and is anybody looking at the laws that are being made and whether the laws make sense?

It just so happens that Paul DiMaggio and I used to take our kids to the same bus stop to get the school bus every morning. Over time we started to talk more and more about where the overlap in our interests was . . . After a year or two of [discussion as part of the "Group Without Acronym"], Paul and I looked at each other and said, you know there's a course here. It would be really interesting to balance off each other and take some of the topics GWA has talked about and turn it into a course. We decided the appropriate venue for trying this would be a freshman seminar.

P: What particular goals did you have in mind for the seminar?

D: This is a tough call because with a course like that . . . there are one or two issues that become particularly relevant, because you open The New York Times and suddenly there is an issue that is very relevant. We had a sense of things that had to be covered, but we also had a sense that as we were developing the course some things would evolve and change.

P: How would you apply some of the discussions from the course to look at the recent suit brought against Dan Peng '05 for running the Wake website? It's not Napster, and it's not Kazaa.

D: These are delicate issues because peer-to-peer is a paradigm that comes up clear across computer science these days . . . Peer-to-peer is, in a different context, very similar to what's called grid computing [which] is a very valid research topic.

The research in doing the connection might be the same in both cases and so suddenly it's necessary for computer scientists to be trained in different ways than used to be the case. It's not just go out and build it and it will be wonderful, it's go out and build it and be very careful of how you choose where you do the prototype to make sure that you're legal.

P: Do you ever worry about the effects laws like the recent Digital Millennium Copyright Act have on academic research?

D: It's politically correct to say yes. But as far as we've come in this interview, one could argue that the set of laws were really comparing to are laws psychologists have to deal with in use of human beings . . . We all agree those [laws] are good things. The problem I see with the DMCA is that it's a law that's made to guess at technology, and it's a law that in places is probably created by people who don't understand the technology as it is or moving forward. The struggle there is to keep the laws up with the technology.

P: You've studied and worked at several institutions. What do you see as the advantages of the kind of community at Princeton?

D: I firmly believe that somebody back when, I don't know whether this is 50 years or 100 years back, found the niche for Princeton that's wonderful. The idea of having a small school which is a cross between an excellent college and a first class research place and getting rid of the professional schools and focusing on one thing or two things and doing them well and integrating them well — I think was brilliant, I think it works and I am happy to be at Princeton where we can have that focus. I think there are a lot of advantages to the faculty and the student body that come out of that.

P: What do you see as some things we could improve upon? For instance, some students have brought up a lack of intellectualism. There is the constant battle for size and funding among the departments.

D: I think a lot of those are necessary evils. I have a daughter who is a freshman at Princeton and when I talk to her about her life and I think back 30 some years to my life as a freshman, I think there's a lot of similarities. It's great to complain about intellectualism, and it's great to complain that all the students are doing is drinking and that they're just drunk all of the time. I think that's a sham. I think that Princeton admits very smart students and that the students are very intellectually curious by and large and it's easy to whine about it, but I'm not sure I've seen evidence of a lack of intellectualism.