- Prepare an insightful catalog of Busy-Beaver candidates: programs that seem to produce the largest output possible for a given size. Use whatever computing system(s) that you like, but Combinators and their close relative, lambda calculus, are particularly interesting.
- Read the other key works of René Descarte (besides the Discours), and find the passages that connect to computation and/or formal systems. The work on geometry may be particularly interesting.
- Read key works of the empiricists, such as John Locke, George Berkeley, David Hume, and find discussions of the knowledge present in the human mind at conception or birth. Look in particular for phrases, such as tabula rasa, or white paper, that indicate zero knowledge. If we understand knowledge to be information, in the sense of information theory, then there is no objective way to determine a zero state. Look for evidence supporting or attacking the interpretation of knowledge by modern information theory.
- Present the history of Hilbert's program to derive all of mathematics from a single system that is proved to be consistent. Cover at least the period from the key paper by Hilbert to the realization that Gödel's theorem proves that the program is infeasible. Analyze the history in some interesting way. For example, take Aristotle's definition of tragedy, and argue whether Hilbert played the part of a tragic hero in regard to his program.
- Trace the influence of the success of axiomatic geometry on the development of rationalism in European philosophy. Which widely held beliefs about geometry from before 1900 are still believed today, and which are now repudiated?
- Install some interesting artificial life software, and present experiments with it.
- Install core wars software. Produce a catalog of interesting competitive programs, either of your own invention or from the work of others. Indicate which programs beat which others. Look for peculiarities, such as cycles analogous to the rock/paper/scissors game.
- Report on the progress of the Tierra experiment. If possible, install the Beagle software, so that we can monitor Tierra. You may even find a way to participate in the project. Look for insights regarding the sorts of programming languages and the sorts of error mechanisms that do and don't lead to interesting types of evolution.
The results of all projects should be presented online, except for
materials that truly cannot be digitized. Papers may be presented in
HTML or LaTeX. Unless someone objects, I intend to archive projects
and make them publicly available along with other course materials.
Maintained by Michael J. O'Donnell, email:
![[]](http://www.cs.uchicago.edu/images/misc/lightningbolt2.gif)
odonnell@cs.uiowa.edu
27 February 1997