News for hackers. Stuff that matters.
Re:The Strange Story of Dual_EC_DRBG – suspected NSA backdoo... by marshray on Friday July 05, 2013 @11:23PM (#5997808)
This is one of my favorite examples showing the need for 'Nothing up my sleeve' numbers. https://en.wikipedia.org/wiki/Nothing_up_my_sleeve_number

I believe NIST learned their lesson, and it was applied in the design of SHA-3: http://crypto.stackexchange.com/questions/6444/why-are-the-c...

Re:The Strange Story of Dual_EC_DRBG – suspected NSA backdoo... by nightcracker on Friday July 05, 2013 @11:38PM (#5997847)
Nothing up my sleeve numbers are very hard to create for asymmetric crypto. Asymmetric crypto is filled with mathematical relationships and it's almost impossible to prove that a certain set of parameters have no "hidden" properties.
Re:The Strange Story of Dual_EC_DRBG – suspected NSA backdoo... by bajsejohannes on Saturday July 06, 2013 @12:44AM (#5997976)
What do you mean? The (asymmetric) RSA algorithm has no numbers up its sleeve. All the numbers involved are created by the participants themselves.

These "numbers up ones sleeve" would only be an issue if they are part of the algorithm itself, like the S-Boxes in the (symmetric) AES algorithm.

Re:The Strange Story of Dual_EC_DRBG – suspected NSA backdoo... by marshray on Friday July 05, 2013 @11:48PM (#5997867)
OK, so why in the world would you use asymmetric crypto in a DRBG? It makes no sense, unless of course you wanted a system with a trapdoor hoping no one would notice.
Re:The Strange Story of Dual_EC_DRBG – suspected NSA backdoo... by pbsd on Friday July 05, 2013 @11:51PM (#5997880)
In this case, it could have been done. Dual_EC_DRBG hinges on the discrete log of two points being unknown; P and Q could have been both generated verifiably at random.
Re:The Strange Story of Dual_EC_DRBG – suspected NSA backdoo... by marshray on Saturday July 06, 2013 @12:29AM (#5997949)
Perhaps something like "The smallest prime number not less than the base-2 expansion of pi*2^(n - 2)".