home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.crypt
- Path: sparky!uunet!usc!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!mcdchg!chinet!schneier
- From: schneier@chinet.chi.il.us (Bruce Schneier)
- Subject: Re: Telephonic Poker?
- Message-ID: <C0C73F.1v9@chinet.chi.il.us>
- Organization: Chinet - Public Access UNIX
- References: <1993Jan3.185628.2976@zip.eecs.umich.edu>
- Date: Mon, 4 Jan 1993 15:59:39 GMT
- Lines: 33
-
- In article <1993Jan3.185628.2976@zip.eecs.umich.edu> positron@quip.eecs.umich.edu (Jonathan Haas) writes:
- >I was reading a paper that explained the RSA algorithm
- >in layman's terms, and among other things it said, "An
- >interesting side benefit of this algorithm is that it
- >makes it possible for two people who do not trust each
- >other to play poker by telephone." I've thought about
- >it, and I can't for the life of me figure out HOW. The
- >paper did not elaborate. Can someone explain to me the
- >algorithm with which two mutually distrusting players can
- >play poker by phone (or, more simply, have a fair coin
- >toss)?
-
- It's actually fairly simple. I take 52 messages, each equalling a card
- of the deck, and encrypt them each with my public key and then send them
- all to you. You can't decrypt any of the cards, so you choose five at
- random, encrypt them with your public key, and send them to me. I can't
- read any of them; I decrypt them with my private key and send them back
- to you. You then decrypt them with your private key, and that's your hand.
- Additionally, you choose five more cards at random and send them back to me.
- I decrypt them with my private key, and that's my hand.
-
- This protocol assumes that the encryption/operation is communitive, which
- RSA is. Given that property, even a secret-key algorithm will do.
-
- There are also a bunch of other things you have to worry about. A new
- public key/private key pair must be generated by each player for each hand.
- Each of the 52 messages should include a different random string, to avoid
- various attacks against the protocol. And finally, it is possible to "mark"
- a card to the tune of one bit of information, using quadratic residues.
- While there are two-player poker protocols that prohibit that one-bit marking,
- I know of no three-or-more-player protocols that do so.
-
- Bruce
-