Aye, the "Monte Carlo" method for computing pi. Works fairly well. However, I recall reading about a closed-form solution a year or two ago. It only worked in binary (or another power of two, like hex) but that's good enough.

The point of the "Monte Carlo" algorithm is to throw dart at a 2x2 square, and of those darts that hit the square, count the number that are within a circle of radius 1. The area of the circle is pi, the area of the square is 4.

We must assume that "rand" returns a value between -1 and 1. Between 0 and 1 works too (I leave that as an exercise for the reader).