Hi,

I'm currently working on another library and can't tackle this one right now.  Also I don't have much knowledge in generating random numbers.

And while writing a little sketch I thought it could be nice to have a library with a very simple API that generates random numbers from different distributions from the uniform (random() from the core.  Some that I found in the Wikipedia:

+ Discrete distributions:
- normal
- poisson
- binomial
- negative binomial distribution
- Maxwell-Boltzmann

+ Continuous distributions:
- normal
- gamma
- student's t
- exponential

It would also be nice to choose the parameters.

Last week I attended a research seminar which talked a bit on how to compute some generation of random number using Markov Models, and it sounded a bit advanced.
But I think that a simpler implementation would do the job, specially if it's fast, even if losing precision.

Is there anybody out there who likes the idea Any opinions

thanks
ricard