MATLAB Programming Contest Blog

December 6th, 2006

The Generality Prize

Of course we have been following with great interest the discussions regarding probing vs. “legitimate” solving. Some of the problems raised by probing are philosophical and relatively intractable, whereas others are straightforward. As we’ve said before, our main interest is in making the contest entertaining, so we rely on you to tell us what makes the contest fun versus what makes it unpleasant.

Several people have proposed some kind of generality contest. Since we don’t want to rescore all the contest entries as part of a generality mini-contest, we’re going to do the following: Between now and the end of the contest, any entries that have the string “generality” somewhere in their title (e.g. “generality-01a” or “john’s generality”) will be marked for inclusion in a separate smaller contest to run in the next few days (we’ll need some time to catch our breath). These entries will go through the queue today as usual for the regular contest, and then after the final winner is announced we’ll process all the generality entries.

The generality test suite will have an order of magnitude more puzzles, but each puzzle will have basically the same structure as you’ve seen in the current contest. We are removing the time penalty in the score, although entries that take longer than three minutes will still fail. Any ties will be decided by the order in which entries were submitted: earliest entry wins.

Good luck in the last few hours of the contest, and keep the discussions going. We’ve learned a lot during this one and hope to learn more before we’re done.