The Indiana REU
I've worked at the Indiana University at Bloomington REU the past two summers. Both times it was a very rewarding experience. The program involves working directly with a professor on some specific area of math. Most professors work with only one student, although some are paired with two. The possibility of students working together in this case has been suggested, but it has never worked out because of differences in background, ability, personality, etc. Unlike some REU's where topics are chosen specifically with the idea of summer research in mind, IU projects tend to be professor's pet projects. This has the disadvantage that the background is rarely self-contained and students can't learn everything that could be relevant before starting to work and that professors don't always have a clear idea what they want to accomplish. On the other hand, this also means the topics are more likely to be useful for later things like senior theses, and it allows greater flexibility both in terms of the range of topics and what the students can work on within the topic. The main problem some people have is that they never really find a question or questions that they are trying to answer and consequently don't come up with much. In particular, some people seem to spend a lot of time writing computer programs to simulate things which don't lead anywhere. This is more something to keep in mind if you go to IU or a similar program than a reason to avoid the program. The professors are all very nice, but they vary in how good they are at helping students find productive things to do. You are welcome to ask me about specific professors if you like. In particular, I strongly recommend against the Population Dynamics project if it is offered again. Projects aren't assigned until after students have committed to going to the program, but they weigh background heavily in assigning projects, so Harvard students should usually get what they want. Research involves meeting with the professor periodically and otherwise working independently when and where you want. There are introductory talks by the professors at the beginning of the program, final talks by the students at the end, and four or five talks on random subjects by faculty members through the program which no one quite has the courage to blow off, but students spend the vast majority of the time on their own. Nothing is scheduled before 1 in the afternoon and generally lunch closing at 2 is the only real constraint on the hours one can keep. Students are housed in a grad student dorm with small rooms that have been compared to cells, but it is air-conditioned and has a computer room and a nice lounge. The dorm is about a twenty minute walk from the math building. The students vary a lot in math background but are generally nice people. IU Bloomington has a well-known music department and puts on an opera and a number of other good classical music performances each summer, but otherwise Bloomington is an unremarkable college town.
--Dean Hildebrandt