On 21/12/10 10:52, Ian G wrote: > On 19/12/10 3:58 AM, Stephane Bortzmeyer wrote: > >> (I even suspect it is impossible to find a pure P2P secure solution, >> where all peers would stay equal. It is just a personal suspicion not, >> it seems, a scientifically proven result.) > > Hmmm, a challenge :) How about: > > 1. Security only means something in the context of people. > 2. All people are different. > 3. Therefore, all security systems will be non-symmetric.
Right, but it doesn't follow that non-symmetric systems have to be non-P2P. You can have a pure P2P system in which different people come to different judgements about the security properties (Cheng and Friedman's Sybilproof reputation mechanisms, for example). How about this, as a riff on Zooko's triangle: you can have equality or consensus, but not both. > 4. And a pure secure (technical) solution is impossible, including any > technical notions of symmetry. QED! > > The mistake would be thinking that a technical P2P system can deliver a > security system. It can't really, what it can do is deliver a set of > technical characteristics which might form part of a security system as > per its users' demands. True... but just as true if you remove the word "P2P". ;-) Cheers, Michael _______________________________________________ p2p-hackers mailing list [email protected] http://lists.zooko.com/mailman/listinfo/p2p-hackers
