Interview with Peter Lee

Friday, Friday 6th, 2009
Interviewers: Betty Cheng (graduate student, LTI), Lucy Li and Yang Shan (undergraduates, CSD)

What similarities and differences did you notice most from being the Associate Dean for undergraduate education and head of the CS department?

Actually, I was Associate Dean and then a CS professor. I then left the CS department completely to work as vice provost. I got to learn about research across the entire university (water, chemistry, material science, etc). But then Jeannette Wing got called away to work at the National Science Foundation, so I got asked back to work in the CS department as department head by Randy Bryant. I was having fun working as vice provost, and so my wife advised me against leaving that position, too. After thinking about it though, I decided to accept the position.

Even when I was vice provost, I still had some really good PhD students. Unfortunately, I didn’t have time to help much with research. And if I continued being vice provost, I wouldn’t have time to be involved in research anymore. Then there’s always the question… do I want to devote the rest of my life to research and teaching? Or be an administrator? Also, I was never the number one person, always the number two. Being the department head would give me a chance to experience being the leader.

So on the roles of being an Associate Dean… I was responsible for the undergraduate program and curriculum in CS. I had to make sure that there were enough TAs for the courses and that students don’t spend too much time only in the classroom but do other things like independent research.

As for being department head, it’s a much more complicated job. The main focus is on the quality of research that we do in the department, which goes down to the quality of the people – students and faculty. My main job is to make sure we have the best possible people in the department. And so to get good people, we need good funding. Really good researchers don’t want to be here if they don’t find enough funding. And really good graduate students don’t want to come here if we can’t offer them research assistanctships.

Another thing is reputation. We want others to know how great this place is. The whole process of how we select people – all of our admissions and recruiting processes – are very thoughtfully and carefully managed because we recruit the best people.

The department head has to do a lot of nuts and bolts administration. It’s my job to manage our budgets correctly, that we don’t overspend or under-spend. My job is to make sure faculty are reviewed correctly, promoted when they deserve to be promoted, and tenured when they deserve to be tenured. Even though we have the very best people here, I believe even the best people won’t do the best work if they feel like they aren’t part of a big team. There’s sort of a feeling of being a coach or a quarterback by trying to create a team atmosphere and make everyone feel empowered to make a difference. It’s really an interesting and comfortable job because I’ve been in this department for my whole career. I know the people here very well. My boss is Randy Bryant, and I’ve known him for 21 years, going on 22 years. So I’m very comfortable working here, and I trust everyone and everyone trusts me.

What is it like to be department head? How much power is involved, and how much work does it take?

[Laughs] Department head has no power. I’m joking there, but in a real sense, one of the wonderful things about Carnegie Mellon is that the students and faculty here are really empowered. Some department heads feel like they’re herding cats, but I feel like I’m herding lions and crocodiles, but very friendly ones. One of my roles is to talk to a lot of people and try to synthesize all of that discussion and how to make the department better. For example, faculty might have an idea about how we may do admissions for our graduate program better. So I’ll get ideas from 3 or 4 people. Then I can take those ideas and synthesize them into one idea that might work and then communicate that to the whole faculty and try to make a good decision. One thing I can’t do, and I think this is good thing, is to just be a dictator. We’re not exactly a democracy or dictatorship – we’re a community that tries to work together. In that community, I just try to be a good communication channel.

As for what it’s like being a department head, I can compare it to being vice provost. Being a vice provost is much busier because a vice provost is talking to the entire university. I used to get a lot more email. However, the emails I get now as department head are much more meaningful to people’s careers and lives. Vice provosts affect the strategy of the university; the emails I get here regard whether a person will get promoted or tenured, whether this person will get a chance of getting that crucial grant. So the issues that I deal with now have much more direct impact on individual people. Someone made a joke to me that the difference between the two jobs is that being a vice provost is being a grandparent, and that department head is like being a parent. As a parent, I have to change a lot more diapers, but it’s a much more satisfying and fulfilling job.

How is the economy affecting your role as department head?

Carnegie Mellon, because of good financial management, is better off than other universities. Having said that though, we will be having some budget cuts. I’m working very hard right now to figure out the best way. I haven’t come up with one yet but I’m discussing it with the whole faculty. So, for us here in CSD, I view this not as much as a budget cut because the actual amount we’re cutting is very small. However, one thing that’s happening right now is that there’s lots and lots of great people finishing their PhDs wanting to become professors, and lots and lots of people expressing an interest in coming here for graduate and undergraduate studies. We have the best people here, so the main impact of the budget cut is that it’ll make it hard to think about expanding our size. We have great opportunities to grow a little bigger, but we might not have the budget for it. The main discussion has been: maybe if we make enough cuts elsewhere, we can have some more program here. Our own PhDs right now are having a hard time finding research jobs because other universities are having a hard time hiring. We’re hoping it’s possible to help them stay here as researchers for a few years until the economy gets better. So right now, I don’t know if, or how, we can do these things. But there are tremendous opportunities and I’m trying to find a way to take advantage of that.

Are you looking forward to moving to the new Gates Building? Have you been closely involved in the development?

[Very enthusiastically] I’m so excited about the Gates building. The project has been with the community for four years now, so just to see a project nearing its end – that’s just tremendously exciting. I had several tours of the building, and it’s the most amazing building that I’ve ever been in. One of the things I love to brag about is the student lounge. The building has 9 floors, and the lounge is on the 6th floor. It will rain and snow into the 6th floor. There’s a glass shaft called the impluvium which opens up to the sky and goes all the way through the building to the “collaborative commons.” It’ll be open to both undergraduates and graduates. There’s also a huge helix, where in the middle hangs something called the rock that has two classrooms in it. It’s just a spectacular place. On the 3rd floor, there’s a huge café cluster with lots of low comfortable furniture and places to plug in your laptop. It’s just architecturally very amazing.

One thing does scare me though. One of my jobs is to assign people to their offices and labs. I’ve discussed this with fellow department heads that have gotten new buildings. What I ended up doing was rehiring Sharon Burks out of retirement. Right now, she is interviewing every faculty member and graduate student. I’m working with Sharon, but she’s really doing most of the work. She’s known all the faculty and staff for years and years, so she really understands their needs and desires and is trying to come up with a plan that works for everybody.

So there’s the high-end cluster on the 3rd floor; there are computerized classrooms, conference rooms, and study areas dedicated to students, particularly undergraduates. There are five or six undergraduate study areas, some for the size of one to two students, others big enough for eight to ten students. There are a total of eleven new classrooms. It’ll be a really nice place. Oh, and I brag about this to other department heads at other universities at every chance I get. Stanford has a Gates building… Harvard has one too, and Cornell is building. They’re really nice things that the Gates Foundation has done, and I think ours will be one of the nicest, if not the nicest.

What classes have you been teaching recently, and out of all of the classes you have taught, which is your favorite?

When I was a vice provost, I didn’t teach. In my first year as department head I didn’t teach either, but I really miss it. I’m planning to co-teach CS 411 (the compiler design course). It’s the course that I’ve taught quite a few times. I really miss teaching. It’s really painful. One of the things I was asked to do as department head was to improve our external visibility and reputation. That means I’ve been traveling a lot, maybe two or three times a month. And so for teaching, that’s a complication because that would mean I’m missing a lot of class. I’m thinking of doing a graduate program course where it’s ok to miss quite a few classes. But, the real fun is in teaching undergraduate courses, and that’s what I really miss. It’s just really hard if you have to miss two or three lectures a month, so my plan is to co-teach with Frank Pfenning. Hopefully, that’ll make it easier. When I’m not in town, Frank could do the lectures.

Besides education, what do you think students should get out of their college experience?

Well, it’s different for undergraduates and graduates. For undergraduates, it’s about becoming a complete adult in society – not just a grown up who can make a living in society, but an activist in society, working to make our world a better place.  It’s important to be fairly social and create a network of friends and colleagues and to become comfortable communicating. Because more than ever in almost any other important job that’ll you do, it’s important to make the other people around you better. You want to be the person who walks into the room, and suddenly, everyone else in the room is a better person. In terms of education, it’s important to not do just courses, but projects and have a one on one relationship with the faculty.

For graduate students, we have a very simple goal – to become the world’s expert for some important problem in CS. And that’s a tall order… That’s why it takes five or six years to complete the program. We really expect and demand a lot from our graduate students. Just like with undergraduates, it has become important for researchers to become good communicators. Our PhD program has changed to give graduates more training on how to explain their ideas more clearly, present better, and write better.

You seem to have a lot of research interests.  Is there a common theme between these topics? And what’s your favorite research area? 

One of the main themes is how we can write the best software. Most of my research has been on programming language ideas that allow you to prove that your software will do what you want it to do. Some of the research has been on taking certain kinds of programming problems and thinking about the best way to express the solutions to those problems. Overall, my main interest has been on how can we produce software effectively.

How has your research evolved during this decade?

I would say that from 1996 to 2004, I was very focused on this idea of how you prove that your programs are safe, that they won’t do anything bad to you. In other words, if someone gives you a program, how do you prove that it won’t crash your computer, or erase your hard drive, or steal your information and send it to me?

Lately, I’ve expanded to some other problems mainly because of my graduate students’ interests, one of which is shape analysis, trying to analyze how programs make use of memory. I also got hooked onto the Claytronics Project (led by Intel and Carnegie Mellon) [link:]. I’ve been developing a language with two of my students for programming claytronics. We invented a language called Meld, where we program these millions of things to take whatever shape you want. It’s an interesting problem – how do you write a piece of code for a million robots? So programming language design has always been a big interest of mine. Meld is a declarative language – designed to write what you want instead of how to do it. It uses ideas from logic programming. It’s a very high level language. By the way, I asked the Gates center architect to give us a digital model so our claytronics can model it.

Several years ago you were an expert witness in the Microsoft vs. Sun case, did you have to be careful not to offend either side?

I was an expert witness for Microsoft. There was a newspaper article about it, and my name got printed in it. I received around eighty pieces of mail from strangers – complete strangers – some of them very angry that I would do such a thing. I gave a completely truthful testimony in court. It had to do with the nature of Microsoft’s Java technology. I have very good friendships with people in both Sun and Microsoft. Some of them have been really great supporters about education and research here. Sun Foundation has provided really great support for Alice. Microsoft Research is providing funding for the computational thinking center. I have a lot of great friends.

As the CS department head, how do you interact with companies in the computer science industry?

I interact with companies quite a lot. There are a number of companies that have always been strong supporters of this department. I have really good relationships with those companies. A lot of those companies are suffering right now due to the economy. We can’t help them financially, but we can help them by giving them as much access as possible to talent that’s graduating from here. So we’re trying very hard to make recruiting easier for the companies that we work with the most. I would say that a lot of the companies I interact with are my friends. I have a lot of fun with them. I interact with them not only on university business but on a lot of government advising, too. Many of the colleagues from Google, Intel, Microsoft, and lots of other smaller companies are often times involved with the same government advising committee with me. We often work very closely. Right now there are a lot of interactions with our Senators and Congressmen because our government is trying to understand how to stimulate the economy. With our friends in the industry, we’ve been doing a lot of work, writing, talking, and trying to give advice.

So how did you become interested in computer science?

My dad, who’s a physicist, told me in high school that there’s this new thing called the computer that looks like it might be important. He was a professor, and he told me that his university was having a summer camp on computers, that maybe I’d like to try it. I went to the summer camp and liked it. Though I liked it, my dad was a physicist and my mom a chemist, so I grew up with this idea that if you weren’t doing hard science, you weren’t doing anything important. When I went to college, I started as a physics major. I finally switched to CS my junior year when I decided that it was still respectable to be a computer scientist.

Do you think you have influenced your child’s perspective on computer science?

I think my son thinks that music and the arts are more important than computer science. My son is very good at math and science, but he takes computers for granted. He’s been using Alice for a couple of years now though, and he really likes that a lot.

How old is your son right now? Are you going to convince him to come to Carnegie Mellon?

He’s only 11 years old. Isn’t it an axiom that whatever your parents do is uncool? I think my son might end up being a computer scientist, but right now he’s at an age where what his dad does is uncool. I’m not yet his hero; I’m still an uncool person. [Laughs] He already told me that he wants to come to Carnegie Mellon. He’s very interested in Carnegie Mellon. I don’t know if that’s because he knows Carnegie Mellon has great CS or if Carnegie Mellon has great arts. Maybe both.

Has your son experienced education in other countries?

As a professor, you often times have a chance to spend a semester or whole year at another university. I thought about taking my son to a school in France. I think it would be a great thing to do. It’s not something I can do very easily right now, but maybe later.

We read your CSDiary blog. How did you come up with the idea for a blog for CSD and how has it evolved over the years?

I started over a year ago, and I wanted a way to communicate to the faculty and students often. I didn’t like mass emailing. Thus I started the blog and put it behind some authentication so only Carnegie Mellon computer scientists could read it. Then I realized RSS readers couldn’t deal with authentication, so I opened the blog to the public. Not many people were reading it, and my writing wasn’t very good. But then I wrote an article about possible collaborations with a university in Saudi Arabia, and that ended up being controversial. Other universities making deals with Saudi Arabia said Saudi Arabia was treating women badly. Los Angeles Times noticed my article, and all of a sudden, a lot of non-Carnegie Mellon people were reading it. So it evolved to being read more by outsiders than insiders. When I write in the blog, I put myself in the mindset that I’m writing for all the alumni and friends regarding how they can all keep up with Carnegie Mellon and what’s going on here.

I would say that up until the start of this winter, I didn’t really understand how to write for a blog. In academia, I have to think really carefully how to write. I revise and edit over and over again. Because of that, most of my blog articles came out overly rehearsed and stilted. It’s only been the last three or four months that I’ve broken away from that and started blurting out the first thing that came to mind. I made a decision not to revise or edit anything as I write for the blog. I think I’ve gotten better because now it’s more spontaneous and maybe even a little more honest. But I’m still learning. One of the benefits is that I’m spending less time on the blog now. I just type in whatever comes to my mind.

You talked about President Obama’s inauguration speech, particularly his statement that he “will restore science to its rightful place,” in your blog. As a computer scientist, how do you see the future of computer science? As the department head, how do you think the future of the CS department will fit into that?

The best way the government can demonstrate this is by restoring funding trends to the levels they were before Bush took office. Restoring funding for basic research and science before the year 2000 would be a good first step. Even while Bush was president, he and Congress passed the law called the America COMPETES Act, which authorizes the government to double the funding for basic research over the next ten years. But they never increased research funding. What Obama could do is actually start increasing funding and get to the level that’s consistent with the law that was passed. Research funding has just been too low. It’s gone down in the past eight years.

The second thing Obama could do is to listen seriously to the advice that researchers and scientists give on the most important issues facing the country – energy and climate change, healthcare, and how the public can interact with the government. As leading professors and scientists, we haven’t been asked to give advice or opinions. We’re hoping the Obama administration will. And Obama’s already gotten into a good start. He hired 2 professors into his cabinet, the Science Advisor, John Holdren and the Director of the Department of Energy, Steve Chu. He’s already showing something that President Bush didn’t.

So one of the things that has always been important about the CS department here is that we have a really broad view of what computer science is. We were the first to recognize that robotics, language translation, and HCI are all part of CS. We have new directions to go into. That ability has been more important than ever. Continuing to broaden the idea as to what is CS is very important. There are big directions in the social sciences and energy that are emerging now and becoming more important.

Having said that though, there are core computer science problems that are really big, like the multi-core problem. Another problem is called DISC, short for Data Intensive Scalable Computing. We’ll be trying hard to strengthen the core computer science areas, as there are these huge new problems hitting us right now. Some people call it the end of Moore’s law. So how do we keep getting more and more computing power with Moore’s Law ending? Hopefully we won’t lose the ability to generate more and new frontiers of CS. While we’ve been leaders in research and have been blazing a trail for the rest of the world to follow in the sub-disciplines of computer science, we haven’t done as much for education. We have the best CS education program in the world, but we aren’t as demonstrative or influential as other universities have been as to here’s how you do it. I’m hoping we’ll have ideas to develop new kinds of courses and teaching methods.

I guess the last thing is more human. People here are really busy, and everyone has a lot of pressure. Because of that, everyone’s having a lot of success. I wouldn’t mind people also having some more obvious fun. Some good things have happened, like Women@SCS, the Dean’s List pizza party, and SCS Day. I’m hoping we’ll be doing more of those fun things.

Dr. Randy Pausch had said that he believed brick walls existed for a reason. We noticed that you had wanted to come to Carnegie Mellon for undergraduate and graduate school - but were rejected. Finally on your third try, you got in as faculty. Have there been other instances where a brick wall kept you trying and trying?

It’s very hard for me to answer because I have setbacks just like anyone all the time. Grant proposals, a paper doesn’t get accepted, maybe one of my students doesn’t pass an important course. There are always setbacks. But I have a hard time remembering those setbacks. I think the answer is yes to this question, but I can’t really remember. It’s good to have a short memory of the bad things.

How do you deal with stress?

When dealing with stress, I remind myself to laugh, smile, and have fun. I’m a relatively competitive person, so I believe the most successful people not only accomplish the most but do it with the most happiness and laughter. And I think that helps with the stress. Also, I’ve come to realize that I can’t do everything and finish everything, and that it’s ok. My wife and son are a great source of calm – and my dog as well.

What kind of dog do you have?

I have a Shetland sheepdog – they look like miniature collies. She’s a very good dog.

When you have free time, what family activities do you like to do?

I live in Frick Park, so every morning with my wife and dog, I take a long walk to the park. It’s about two or three miles every morning. The time is very calm, healthy and happy because the dog is happy [laughs]. It’s also a time I’m with my wife – just with her and no one else in the morning. My son is very musical and plays the violin. I play the piano, so we play together quite often. He’s gotten too good for me, so I’m not able to sight-read the accompaniment anymore; it takes practice now. Often, I’m not a very good accompaniment because of that.

I mentioned before that I travel a lot. I try to take them with me as much as possible. So for example, last year when I had to go to Taipei and Tokyo, I took them with me. It was a great trip. We really had fun. Then I went to Washington around the time of inauguration, and I took my wife and son with me. That was a big adventure. We travel all over the world together.

In your last interview on the SCS website, you mentioned someplace tropical as your dream vacation. Have you gone since then?

[Laughs] No, I still haven’t gone. I’ve never been anyplace tropical. I think someone needs to kidnap me and take me to the Caribbean. Hawaii would be great too.