I participated in some ACM programming contests when I was a student. My university had about half a dozen teams, and we had local contests and participated in

I participated in some ACM programming contests when I was a student. My university had about half a dozen teams, and we had local contests and participated in national and european level competitions. I can see where the author is coming from, but I can't agree with him fully. Programming contests really are no measure of how good a programmer is. At most they are a measure of how good a programmer is at memorizing solutions to "standard" classes of problems and then quickly adjusting them to specific instances of them at the competitions. But I don't think that's why competitions exist at all. I admit, I was never a good programming "athlete". I like solving those kinds of problems, and I like to solve them as quickly as I can, but I also like to immediately go back to them once solved, to "clean up" the code and make sure I can understand my solution when I read it again a few years in the future (which will probably never happen, but nevertheless). The problem is the "as quickly as I can" never was "9 problems in 5 hours", not even 5 problems in 5 hours. And the same applies to my fellow teammates, which I don't see as bad programmers at all. In fact, looking back, it is funny how I seemed to like to pick apart other people's solutions when they happened to solve the problem as far as the automatic judge goes, but then seemed to have some corner case problem where it would probably fail (incomplete solutions). This happened from time to time, and it could also be attributed to me failing to solve the same problem, but incomplete solutions really irked me. :) But the social aspect of the thing was important, probably more important than the actual programming, not only because of the visits to other academic institutions, but also between the teams of the same university.