On Sun, 13 Oct 2002 18:08:56 +0200, Paul Gillingwater <paulat_private> said: > First, there should be a timestamp server, which upon request, will generate a > strong hash using its private key of the current time. This can then be > incorporated into a data set's signature. I don't have my copy handy, but I do believe this is all covered in the usual detail in Schneier's "Applied Cryptography". For bonus points, analyze what happens to your protocol if the system clocks on the client and timestamp server are skewed (note particularly that Very Odd Things can happen if the client asks for one timestamp, then moves its system clock backwards and asks for another timestamps....) -- Valdis Kletnieks Computer Systems Senior Engineer Virginia Tech
This archive was generated by hypermail 2b30 : Mon Oct 14 2002 - 11:01:44 PDT