Slashdot Mirror
Linux Journal Interview With Brian Kernighan
pndiku writes "Linux Journal has an interesting interview with Brian Kernighan where he talks about AWK, AMPL and how he had nothing to do with the creation of C."
← Back to Stories (view on slashdot.org)
The AMPL link points to nothing; it should be http://www.ampl.com.
Floating face-down in a river of regret...and thoughts of you...
In order to reduce the slashdot effect and avoid ads, presented for your viewing pleasure: Linux Journal: Could you tell us a little about yourself?
Brian Kernighan: I was born in Toronto and went to the University of Toronto as an undergraduate, in a course [of study] called Engineering Physics. It was basically a lot of science, math and engineering for kids who were good at math and thought they might be engineers but didn't know what kind. It was a tough course, and about two thirds of the people eventually dropped out, but I managed to survive and learned a lot (a small amount of which I still remember).
I didn't really know much about computers--this was in 1960-64--and there was only one big computer at Toronto, an IBM 7090, plus a small 1620 in the electrical engineering department. I saw my first computer, an IBM 650, after my second year, and I learned a bit of Fortran when I was in my third year. I spent a summer writing Cobol for a big oil company (honest) after that. It was enough to get me hooked on programming, though I sure didn't know what I was doing and was a terrible programmer.
At Toronto, I also did a senior thesis (a literature survey, really) on artificial intelligence, which was showing all kinds of promise in 1964. So I decided to go to graduate school, without really knowing what that was all about. But it was easier than looking for a job. I wound up at Princeton because they made a better financial offer than any other school. I had a good friend, Al Aho, who was already there; he had been one year ahead of me at Toronto, in the same course.
Princeton didn't have a CS department at that time, only a group of good young people in electrical engineering, but I enjoyed it and had a good time for several years before settling down to work on a thesis. It was a very nice place to be a graduate student.
LJ: How did your life become connected with computers?
BK: I think the real turning point was the summer of 1966, where through good luck I got a job at Project MAC at MIT, [working] for Fernando Corbato. This was a fantastic experience: I was using CTSS, which was the first general purpose time sharing system and is still one of the nicest to use. It was infinitely more productive than the punch cards I was used to up to that point. I learned to program in MAD and wrote programs to help collect information for the Multics machine; the first GE 645 arrived that summer. It was a wonderful place to live and work, with great people (like Corby, who is still alive and active). It was definitely one of the best times of my life.
The next summer, probably because of the MIT experience, I got a job at Bell Labs in the Computing Science Research Center. This time I learned assembly language properly and met a bunch of the people who I had heard of while at MIT (they also were working on Multics). Another great summer.
I went back to Bell Labs the summer after that. This time, I got lucky and worked with Shen Lin, a great mathematician and problem solver. Shen was interested in hard combinatorial optimization problems, such as the Traveling Salesman Problem. I had been working in a casual way on what came to be called the graph partitioning problem for my Princeton thesis. Shen had an idea of how to attack the general case, and I made the algorithm work in a Fortran program. It became the core of my thesis, along with some other special cases. Anyway, I had such a great time at the Labs those two summers that when I finished my thesis early in 1969, I didn't even look for another job--I just went to the Labs. I was lucky to be in the group that did UNIX and C and all of the great things that came with them; that started just after I arrived. In many ways it was the best computer science research group anywhere, part of a large and productive research organization, and it had an enormous influence on the world. I stayed there until 2000, 30 wonderful years with an amazing group of people.
LJ: What is your work these day
int foo, *bar; creates an integer named foo, and a pointer to an int named bar. Right? Or am I wrong?
:-8
Wrong.
foo is an int.
*bar is an int.