On: Google and UT’s 2011 AI Challenge: Ants

[muuh-gnu]

Interestingly, the leading bot language this year seem to be Java, while last years winner, Lisp, isn't among the current top 100 entries any more.

[nafta → muuh-gnu]

Perhaps, that stems from the fact that Java's starter package implementation was the earliest to be developed and is still today the most mature (besides Python's anyway).

[ggfa → nafta]

I'm sure that's a huge component.

[cperciva → muuh-gnu]

In order to succeed on this year's contest, you need good path-finding, which you won't be able to do easily in lisp within the time limit.

[wes-exp → cperciva]

Why is lisp worse at path-finding?

[cperciva → wes-exp]

Lisp isn't worse at path-finding; it's slower at path-finding, simply because it's slower at everything.

Lisp has its place; performance-critical code isn't it.

[ohyes → cperciva]

This is incorrect.

Lisp is not slower at everything.

When properly tweaked it can be pretty close to java in speed.