CSE 505 - Project

One of the course requirements is a project on a programming language issue of student's choosing. The project can either just involve reading and discussing papers, or can also have an implementation component. You choose the topic - we'll give some feedback on your initial proposal. Students may work individually or in pairs (somewhat more work will be expected from a pair). Please feel free to use the cse505 mailing list to look for project partners.

Deliverable: 10-15 page paper, and optionally a short presentation in class. Think of the expected work as about 4-5 homeworks worth.

Short project proposal due Nov 19. This should be a short e-mail message to Alan, Vass, and Todd describing the topic, the sources you plan to use, and the format (for example, a paper discussing one language, a comparative languages paper, or an implementation project and accompanying paper). We'll send feedback on this proposal. If you send the proposal sooner we'll send feedback sooner; preliminary inquiries are also welcome.

Paper due Dec 12.

Suggestions

There is great flexibility in choice of topic - pick something that interests you, or that will potentially feed into your graduate research. Below are a few suggestions; I'll add to this list if appropriate.

Literature sources: relevant journals include ACM Transactions on Programming Languages and Systems, Constraints, Theory and Practice of Logic Programming (and formerly Journal of Logic Programming), and others. Relevant conference proceedings include various ACM SIGPLAN conferences, IEEE-sponsored conferences, and standalone conferences such as the International Conference on Logic Programming, and Principles and Practice of Constraint Programming. Also see the 505 Useful Links page.

If you are interested in constraint languages, one topic would be Constraint Handling Rules. (This would be suitable either for a paper alone, or an implementation-based project.) See Thom Fruwirth's CHR page

Another constraint topic, which also includes object-oriented languages and other concepts, is Oz.

Further explore some aspect of functional languages, such as applications of monads, or small domain-specific languages in Haskell, or implementation techniques for pure functional languages such as combinators.

In object-oriented languages, you could investigate a pure OO language, such as Cecil or Squeak; or Java extensions such as Pizza, GJ, and NextGen (we'll talk about Pizza and GJ in class, but there is more that could be done). Or you could investigate other OO languages, such as BETA or Squeak. (See the links on the Other programming language resources page.) Or investigate Python or other clean object-oriented scripting languages.

A local UW CSE language is ZPL, a parallel array-oriented language for scientific computation. (By the way, for years I've been looking for someone to do a project investigating the relationship between constraint languages and ZPL. If you look at the heated metal plate example in the Marriott & Stuckey text (also in ~borning/clpr/progs/laplace on ceylon), and the Jacobi program on the ZPL walk-through page, they are strikingly similar. What I'd like to have someone try is defining a CLP(Array) language, with operators modelled on those in ZPL. There could be a very inefficient but clean prototype implementation in CLP(R).)

Last updated: October 20, 2001.