> "Abdulaziz Ghuloum" <[EMAIL PROTECTED]> said:
> Doesn't /dev/random produce pseudo-random numbers also? What makes
> reading from it better than using perl's rand or C's rand?
This is getting a bit of topic.
/dev/random is seeded with supposedly random events. In Linux I believe it
takes the time between keyboard and mouse events and builds up an entropy
base. The better the entropy base, the more random the bits from /dev/random
are.
Pseudo random number generators like rand() do not truly generate random
numbers. If you run them long enough, they will repeat the generated values.
For most application of random number this does not pose a huge problem.
However, in the cryptography world a source of random numbers is essential to
insure the robustness of most cryptographic algorithms. If you know the
random seed used to generate cryptographic keys is not truly random, then it
is far easier to break data enciphered using the key.
I believe with the right seeding that /dev/random is close to being a source
of random numbers.
--
Smoot Carl-Mitchell
Consultant
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
