Prof. Gustavo Alonso

Diese Seite existiert nur auf Englisch!

Laboratory Course on Distributed Systems

Information and Communication Systems Research Group

Why should you take a lab course on distributed systems?
Prof. Gustavo Alonso answers this question (and a few others) in a candid interview with Ruth Bürkli.

June 17, 2003

What is your main scientific interest?

I'm mainly interested in the architecture of large distributed information systems.

Can you give us the highlights of one of the courses you’ll be teaching next semester?

I'd like to talk about the laboratory course in the major on distributed systems The laboratory is a special project course. Students have the opportunity to work on a software development project in a realistic setting comparable to what they will have to face once they're going into industry. They must complete these projects in groups. We set a number of concrete goals but do not explain in detail how to get there. The students must decide how to reach these goals. We expect them to manage their group and to assign tasks to the different people in the team, and we expect them to be able to explain how they are organized. If something goes wrong, they have to find ways around it, change the design, always providing solid explanations for their decisions. The presentation of the results, the degree of achievements in terms of the stated goals, and the ability to manage the project, all play a role in the final grade. We have very interesting projects where students must develop a wide variety of skills - working with robots, designing peer-to-peer systems, confronting the constraints of small devices, facing environments with a variety of networks, and so on.

Why should a student be excited about the opportunity to take this course?

What the course does is to combine both the technical and practical aspects of computer science. Students must face a realistic problem. There is not a single solution to the problems we pose; students must look at the problem and decide how to go about it. The students must also face all the constraints of realistic system design. They have to work in teams, manage a team, and present and defend their design decisions. They have to justify what they're doing and explain the results obtained. We are not going as far as asking them to market the results, but this is a bit the idea! The experience of working in a team on a realistic project is a crucial part of this course. In addition, the laboratory offers the opportunity to work in depth with tools and technology that we cannot easily use in normal lectures due to the large number of students.

How big are the groups?

There are 4 to 5 students in a group.

Who is your target audience?

Students at a master level, that is 7th semester or above.

As a student, what do I get from the course?

We want to confront students with real problems, not just homework problems. Because once students graduate and face industrial life, they won't get a well designed homework assignment to work on either. They will get a set of requirements and a description that sometimes is not even correct, and they will have to come up with something that works. The course goes in this direction: Students should realise that they're not just doing homework, but designing a solution to a concrete engineering problem.

Why should anyone study computer science?

Nowadays, computer science is probably the most relevant of all engineering disciplines. Everybody is using computers. There is not a single branch of science today that is not based on computer work:computational chemistry, computational physics, computational biology. This doesn't mean of course that scientists in chemistry or physics are all computer scientists, but it points out the importance of computer science. Not only the industrial world relies on computers, but also the world of science. In this respect, studying computer science is a great opportunity to make a contribution to society.

Everyday life relies on computer science as well. Just see all the digital cameras and other toys around you! They are all based on computers. So studying computer science is an opportunity to contribute to the quality of everyday life by developing technologies and systems that are useful and needed.

So you think it is still exciting to be in computer science?

Certainly! After the internet bubble burst, it has become even more exciting. Now we can focus again on more fundamental aspects of computer science rather than on those dictated by the stock market!

I think people are starting to realize what the true potential of computer science is and what you can actually do with computers. So this is a very good time to study computer science. You can actually do research without the background noise and commercial pressure of the past years.

Is computer science creative?

Computer science is very creative! It's the science of the artificial. To a very large extent we are not limited by laws of nature like in physics or many other traditional disciplines where there are very clear boundaries to what you can do. In computer science, the limit is your imagination! The more creative you are, the further you are going to get.

What's the importance of mathematics in computer science? Do you have to be a good mathematician?

A computer scientist is not a mathematician. Computer science is a very general area. But some people work on topics very close to mathematics; on the other hand, applied mathematics is very close to computer science. But computer science is a mature discipline. It has its own body of knowledge. Mathematics is part of it, but it's by no means the dominant part.

Why aren't there more female students?

Well, I think that there are different factors. For me it is a novelty that women would not want to study computer science. In Spain (where I come from) computer science used to be a women's engineering discipline, the majority of the students were women. I think that there is a very strong cultural factor to this. Still, if you compare the department of computer science with other ETH engineering departments, we are even slightly ahead. But certainly we are not as far as other branches of science like for example biology. And this is something we'd like to change.

What do you think of ETH as a research center?

ETH has a significant advantage over many other universities in the world: The emphasis on long-term research. Unlike other universities or other research centers, the pressure to work on small immediate problems is not there. This allows not only faculty but also graduate students to concentrate on valuable and long-term contributions. People can work on long term projects that take several years before they come to completion. Research at ETH is not driven by fashion, but by finding solutions to fundamental problems.

Why did you become a computer scientist?

Originally I studied telecommunications. It was only when I did my PhD that I switched to computer science. One of the reasons was that I found there is much more room for imaginative and creative work in computer science. There are so many more possibilities for creating new realities.

Do you find it interesting to work with students?

Of course I find it interesting. That's why I am a professor, otherwise I would be a in a research lab or in industry. Working with students is interesting mainly because of the creative aspect. You are constantly surrounded by people who are eager to learn and be creative. It's a constant exchange of ideas, sometimes an intellectual fight to see who comes up with the cleverest and best solution to a problem. I find it very stimulating and exciting when students can't wait to discuss their latest results or developments.

So: Be creative and enjoy your time at ETH!

Mr. Alonso, we thank you for the interview and wish you good luck for the upcoming laboratory course.