On Thu, Jul 19, 2012 at 6:09 AM, Chas. <[email protected]> wrote:
> There isn't any way any of this will work. Keep in mind the only goal is to make it more difficult for greedy users (potentially with armies of Sybils) from abusing the network. If I'm doing better than a random selection, I think I'm succeeding. I'm not looking for some unpwnable bulletproof system that will prevent you from ever interacting with malicious Sybils. I just want to make a best effort to try to filter good peers from bad. Bad peers may still be acting in good faith, but perhaps the network topology is unfavorable. > So long as there are unverified endpoints (by some third party) there is > no way any "trustworthy" system can be built up! What is to prevent me > from being both a sybil, and under a second identity being just some > "random" other party. Once we - as in me as a sybil generator and me as a > disguised second party - transfer a file and sign off on it, I become > slightly more trustworthy. So I do this enough times, with randomly > generated "second parties" and soon I look really "hot and trustworthy"! You now have: one "good faith participant" (who's actually a liar), and one sybil, with a shared history of lies. However, collaborative filtering finds similarity patterns in large groups. That entire history of lies will be filtered out as mere noise in the data set (in the event we have a sufficiently large number of "good faith" peers) To confuse the algorithm you would either need an extremely small pool of peers (e.g. you're still being bootstrapped into the network and you don't know enough to tell good peers from bad), or a large number of lying-but-otherwise-good-faith participants to convince a given peer to interact with a Sybil. -- Tony Arcieri
_______________________________________________ p2p-hackers mailing list [email protected] http://lists.zooko.com/mailman/listinfo/p2p-hackers
