Barry Boehm, founder and Director of USC’s Software Engineering Center and TRW Professor of Computer Science won this year's IEEE CS Harlan D. Mills Practical Visionary award. We spoke to Barry about this award and his career.

Q: Could you tell us, without embarrassment, how and when you first began to be involved in the Computer Science field?

I was a math major through my sophomore year in college, and wasn't quite sure what math majors did for a living when they graduated. In the summer of 1955 I had a summer internship at General Dynamics, and when I showed up the first day, they had a board with possible summer assignments in thermodynamics, propulsion, and also in something called the "digital computer lab". That's where I ended up.

Q: How about how and when you got involved with software engineering?

Even at the beginning I wanted to know why I made mistakes in coding and in getting the user's needs right. But actually, I tried to get out of the computer field for about five years after that. I went to RAND and was involved in studying why a command and control system had failed. We concluded that it was the software, and that renewed my interest in software and the difficulties in constructing it.

Q: What are the biggest changes, for better and worse, since you first began software engineering research?

I think that we've been good at being able to make paradigm shifts to keep up with the changing technology. In my years, computers have gone from batch, sequential, and single-user to interactive, concurrent, and multi-user. Software engineering has been able to re-invent itself to keep up with delivering software that supports these changes. For the worse, the field has grown faster than the ability of people to learn old lessons. There is too much re-invention of old material and relearning of old lessons. If you look at current research papers, there are rarely any citations older than five years. We need to be better at establishing a corporate memory.

Q: So, what has stayed the same, for better and worse?

We've been able to keep this adaptation paradigm and react to changes in computing technology over the many years. At the same time, the "trailing edge" has seen too many people unwilling to make that change, holding on to the past for too long.

Q: Can you give us an anecdote about Harlan Mills that you recall?

Harlan had a tremendous scientific curiosity and a belief in people's better nature. This showed in his goal to build defect-free software. He didn't make excuses about being human, and the people working with him often did create apparently defect-free software.

Q: What advice would you give to all the "youngsters" who are just starting out in software engineering research?

Spend some time in industry working on real software development projects. You need to get your hands dirty and learn not in your mind but in your heart and gut the problems that real software engineering faces. This will help you understand what research ideas you have that might be most applicable in practice.

Q: What does your crystal ball show will be hot in five or ten years?

Software engineering for systems that coordinate multiple people and agents, with multiple stakeholders and decision makers. Software-based decision agents present an interesting systems engineering challenge.

Q: Where do you see the most troublesome or long-lasting gaps in software engineering?

The integration of software and systems engineering. I think there was an over-adoption of separation of concerns, which hampered people coming back together.

Q: Would you list your career accomplishments that you are most proud of?

Well, COCOMO, the Spiral process model, and Win-Win are the big three.

Q: What goals do you still want to accomplish?

We're working on Mbase, a model-based engineering and architecting paradigm for system design. We've been refining it by using it in the classroom, and it is currently packaged as a way to educate.

I heartily recommend a stroll on the river path. It's a beautiful diversion.