The first Midwest Theory Day was in Dec. 1980, the one coming up is the 52nd. John Rogers at DePaul keeps a list of recent MTDs as well as all past meetings.

One important purpose of the Midwest Theory Day is for people that do theoretical computer science in Midwestern Universities to keep track of what each other is doing. It is a one day conference so that most people come and leave on the same day. Thus, most people attending live within 300 miles of the place where it is meeting. Audience size can be anywhere between 50 and 150.

The people who come are perhaps 1/4th computer science faculty and almost 3/4ths computer science graduate students. The talks are about theoretical computer science in the broad sense of the word. It is pretty much the case that the speakers are either faculty that are normally considered to be theoretical computer scientists or graduate students they know. They can speak on most any topic they like with the understanding that they should pick topics that would be interesting to a theoretical computer scientist. Most talks deal with the heart of the field (things like analysis of algorithms, quantum computing algorithms, etc.) but some range all over the place. The unifying thing is that the topic would be interesting to theoretical computer science faculty pretty much regardless of whether or not you would normally think of it as part of theoretical computer science. Thus, one year there was a nice talk about making 3-D objects out of photo-harden plastic, a topic that did have some theoretical aspects.

We try to select speakers that will give interesting and high quality talks. However, we do not do rigourous refereeing as that term is normally understood. There are no proceedings. We want to have interesting talks, not an alternate means of publishing an idea.