|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
Subject: Re: [Re: cycle in hash chain ?]
From: Ben Laurie (ben
algroup.co.uk)Date: Fri Apr 28 2000 - 11:56:05 CDT
- Next message: Joseph Ashwood: "Re: PRIMAIRY KEY Authorisation DBase/ registration utility?"
- Previous message: R.M.: "Re: [Re: cycle in hash chain ?]"
- In reply to: R.M.: "Re: [Re: cycle in hash chain ?]"
- Reply: Ben Laurie: "Re: [Re: cycle in hash chain ?]"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
"R.M." wrote:
>
> It seems clear by intuition that any function whose input is from an infinite
> field and whose result is from a finite field must eventually generate
> duplicate values.
You don't need intuition for that, its obvious. However, the interesting
question is whether it cycles. I can trivially prove it need not,
however, the really interesting question is whether a one way function
should generally fail to cycle. Intuition says to me: "yes".
> In a process like
> > S1 = E_k(S0 ^ n),
> for any given SO it has to cycle when n >= SO.
Proof?
Cheers,
Ben.
-- http://www.apache-ssl.org/ben.html
- Next message: Joseph Ashwood: "Re: PRIMAIRY KEY Authorisation DBase/ registration utility?"
- Previous message: R.M.: "Re: [Re: cycle in hash chain ?]"
- In reply to: R.M.: "Re: [Re: cycle in hash chain ?]"
- Reply: Ben Laurie: "Re: [Re: cycle in hash chain ?]"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]