I attended some of the final presentations of an undergrad class on Game Programming today with a friend. We went in expecting something more like a poster session, where people are arrayed around a room showing their work off to a few people who managed to crowd around them. The poster session is ideal for brief browsing, because you can skip anything you’re not interested in. Instead, it was a series of power point presentations followed by an on-screen demo.
This was the first time I’ve been around a group of undergrads outside of public areas at CMU. It was interesting in several ways. One, I had forgotten how bad undergrad presentation skills are (shameless generalization alert!). Grad skills aren’t a whole lot better probably on average, but I think grads have a lot more practice, so some of the silly problems appear less often. The second thing that struck me was that there are a lot of engines out there for making cool games. I’ve never taken a class on game programming, but I did have a class on computer graphics. We focused on OpenGL exclusively. The third thing that got me was that I was surprised by the quality of the work. I am probably being unfair to my undergrad institution, but I really don’t see a class like that producing there what I saw today. Of course, when I was there, there wasn’t a class on game programming at all.
This got me thinking again on what constitutes quality in a computer science program. Is it just that CMU has a ginormous CS program and so can offer classes on game programming, whereas the University of South Carolina can’t? Is it instead that CMU can attract students who are smarter (according to the standardized ways society measures these things) and are more dedicated? Presumably the smarter and more dedicated students make it easier for professors to teach advanced material, since they care and move through it more quickly.
If that is the case, then there is essentially no difference between USC and CMU in terms of education potential. I’ll define education potential as the amount of knowledge that a school’s professors can impart to students over four years. To be clear, I will consider professors to be knowledge-imparting machines: that is, they each transfer knowledge to students at some rate, which depends on the professor. This is normalized by four years of schooling, so all things being equal, two schools can be compared whose departments are of different sizes.
However, a larger school is still more advantaged because having more professors, the knowledge per professor ratio is lower. The higher the knowledge per professor ratio is, the more a single professor has to know to impart the same amount of knowledge to a student. Students are knowledge consuming machines. They are able to consume knowledge at some rate, so the total amount of knowledge that can be transferred is bounded both by the amount the professor can send and the amount the student can receive. According to my theory, students are the bottleneck here. Assuming that student consumption rates are normally distributed, if a professor teaches at the average rate of consumption for the class, half of the students will be left behind, while the other half will be consuming less than they are capable of. Therefore, the average consumption rate of students is the deciding factor in the quality of education between two schools with education potentials that exceed the consumption potentials of their students.
If a school has an education potential that does not exceed the consumption potential of their students, this is a sign of a bad department. This might occur when the level of teaching ability is particularly low. I doubt it occurs very often because the students are smarter than the professors. Another cause is a bad curriculum or a lack of diversity in the distribution of professors’ skillsets (e.g. a school may have a strong databases group but nonexistent theory group).
Unfortunately, this leaves me no closer to an answer. It also may be utter nonsense. As a parting note, I wish some of the games I saw today would be made available online. There were at least two of the six I saw that I’d consider to be worth playing.
Update
In any education system, there is a certain amount of education potential that is lost: that is, radiated to the environment. This is basically the difference between the education potential and the cumulative consumption rate of the students. Making lectures available online decreases this loss, since the environment consists of anyone with enough time to learn, rather than just the walls of the classroom, who as a rule, refuse to learn.
The Mendicant Bug 