at Monday, December 16, 2002 5:19 PM, R. A. Hettinga <[EMAIL PROTECTED]> was seen to say: > As I've said here before... > > At 6:51 PM +0530 on 12/16/02, Udhay Shankar N wrote: > > >> Peppercoin is > > ...Ron Rivest's random-settlement "lottery" payment protocol. > > Essentially, you write 100000 checks for $100.00, and redeem one of > them, yielding an "expected" payment of a tenth of a penny. > > You need very strong is-a-person digital signature credentialling, > just like checks. It's quite compatible with PayPal, etc., and so I > expect that that's part of their exit strategy. > > If they could get plugged into the ACH/ATM network, it might work > there as well, so you could also sell it to banks, if they're buying. > > Cheers, > RAH Indeed so, yes - however, statistical payments only work if there are a lot of them - when you dip below the single unit (100,000 hits that provide a peppercorn) then you have a statistical chance of either getting a full $100 or nothing at all - when your outlay is definitely due every month regardless of if you get paid or not. This would encourage aggregation - pooling peppercorns amongst several sites and splitting the income in proportion to the peppercorns collected - but given enough aggregation, that is no different than the frequent current setup where you gain "membership" of a multi-provider site and the fees (minus a handling fee of course) are split in proportion to hits amongst the content providers. There is also a fairly high processing overhead - for every hundred bucks (on average) you get, you have to verify 100,000 digital signatures and present each one for fullfillment (possibly repeatedly unless the validator knows up front which one will be redeemable and which 99,999 will not) Honestly I can't imagine the pain of this being worth paying a patent usage fee for.
--------------------------------------------------------------------- The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
