entropy

[Bruno Wolff III]

On Sun, Jan 10, 2010 at 03:47:56 -0800,
  Don Quixote de la Mancha <quixote at dulcineatech.com> wrote:
> 
> If our Monte Carlo had any kind of non-randomness in it, it would have
> been very difficult for us to tell.  It would have caused a systematic
> error in the calculated acceptance, which would have caused a
> systematic error in our published measurements.

The output of /dev/random is supposed to be cryptographically strong. So
even if the there was some input bias, it shouldn't be visible on the
output.

If there is some problem with the output (say because of a bug), you are likely
to have the problem regardless of whether or not good random input was
provided.