[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Poisson numbers for random keyspace assignment

                  [Great statistical summary deleted]

TC> * For opportunistic attacks on keys in challenges, the odds are 95% that a
TC> key will be found with only twice the total effort (or time) using a
TC> totally random method of picking up keyspace to search.

  The odds can be improved somewhat by scaling the granularity of the
sweep to the size of the sweep. (Align larger chunks on large-chunk
boundaries, eliminating the chance of overlap with other large chunks.)

TC> * This is probably good enough. (And if one only wants to be 90% sure of
TC> finding the key, even less effort is needed.)

  The best advantage of the random method is that it allows people to
participate completely anonymously, as there is nothing to report save
the Eureka!, and that can be done through a remailer anyway. When the
challenge is solved, everyone can stop cracking.

       It is one thing to work on an academic exercise, but the =real=
     test is how well the resources could actually be marshalled for a
     =real= attack. The requirement of leaving an audit trail to
     participate reduces the supply of volunteers. This approach can not
     be challenged as unrealistic.

  The whole thing can be managed via Imail, which I must point out to
you Totally Connected People, is the least-common denominator for
participation here. You have =lots= of people out here who would like to
join the Cypherpunks Brute Squad (We gotta get T-shirts for this one!)
but don't have WWW access.

 * John was a complete D**k, until Lorena got through with him.
 * [email protected] *