On 12/27/13 8:19 PM, S. Dale Morrey wrote:
Well Levi you would be quite correct if the intent were to actually seek a collision across 256 bits of space. That is not what I'm going for here. In my mind detecting a collision would be evidence of a flawed implementation of the hashing algorithm which is what my experiment is actually seeking to uncover. So I'm checking the specific implementation to rule out a flaw, not nessecarily the algorithm.
If that's the goal, then wouldn't the best way to test it be to hash a given string with your implementation, as well as with a known good implementation, and see if they match? Your implementation could be completely wrong, but still not yield any collisions in your problem space.
Steve /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
