[NCLUG] My attempts using /dev/random...

[Jim Hutchinson]

On Mon, Apr 14, 2008 at 1:09 AM, Sean Reifschneider <jafo at tummy.com> wrote:

> >Unfortunately, the harsh truth is that both /dev/random and /dev/urandom
> >are flawed.  They're PRNGs -- they do not provide *true* random numbers.
>
> /dev/random on Linux is *NOT* simply a PRNG.  Which is why it took nearly
> a
> day to gather enough entropy for 1.6MB worth of /dev/random output.
> See the comments at the top of the /dev/random driver source for
> information about how it collects entropy, or it's approximation of it.
>

This conversation is extremely interesting. I think it's great that a 6th
grade project has encouraged so much discussion. I took Leina to the state
science fair at CSU on Saturday, and it's pretty obvious that in order to
get anywhere you have to have a solid project with a real application. That
is something she never really went into beyond saying pRNGs are needed for
encryption. It's also interesting to learn that Ruby and Python are
basically the same in terms of the RNG module they use (or so I understood
from previous emails). Leina wants to continue her project and improve on it
for next year. I think some of the questions she will need to address are 1)
what is wrong with bad (or just less good) pRNGs and 2) why is that
important. There are probably more as well. I copied Leina on this. Please
share if you have any thoughts.

Also, for anyone working on these little experiments, I think that using
sums of dice as a model may not be the best test of randomness. I don't know
what would be good, but there is nice, 107 page paper on the
random.orgwebsite that purports to tests their RNG (
http://random.org/analysis/). This stuff is over my head but maybe some of
you will enjoy/understand it.

Cheers
-- 
Jim (Ubuntu geek extraordinaire)
----
Please avoid sending me Word or PowerPoint attachments.
See http://www.gnu.org/philosophy/no-word-attachments.html