[erlang-questions] Speedy unsort:shuffle/1,2 ?

[Richard O'Keefe]

On May 30, 2010, at 7:53 PM, Kenji Rikitake wrote:

> In the message <5DD3CDF1-6148-46FC-8496-A3010B97BD30@REDACTED>
> dated Fri, May 28, 2010 at 10:56:45AM +1200,
> Richard O'Keefe <ok@REDACTED> writes:
> [...]
>> It might be a good idea to move the core random number generator
>> in random: into C (at present random:uniform() is quite slow) and
>> perhaps even replace it by a better RNG.
> [...]
>
> For a testing purpose, the RNG doesn't really have to be
> cryptographycally safe; it only has to have a uniform distribution.

Where did "cryptographically safe" come from?
By better, I meant
  - faster
  - having better statistical properties

>
> Anybody has implemented a NIF or port version of Mersenne Twister or  
> SFMT?
> http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/SFMT/index.html

and those are excellent examples of what I meant.