[1309] in cryptography@c2.net mail archive
Re: Using satellite feeds to get shared pseudorandom numbers
daemon@ATHENA.MIT.EDU (Bill Stewart)
Tue Aug 5 15:01:26 1997
Date: Mon, 04 Aug 1997 22:49:54 -0700
To: "Perry's crypto list" <cryptography@c2.net>
From: Bill Stewart <stewarts@ix.netcom.com>
Cc: John Kelsey <kelsey@plnet.net>
In-Reply-To: <MAPI.Id.0016.00656c73657920204542324330303033@MAPI.to.RFC8
22>
At 03:34 PM 8/3/97 CDT, John Kelsey wrote:
>I'm sure this has been invented before, but I haven't seen it
>published anywhere, and I think it has particular relevance with
>respect to various potential laws against some kinds of services
>(particularly involving various ecash and gambling systems).
...
>a. There's no surprise left in the pseudorandom stream. That
>is, they have to do this protocol each time they want to have
>some unpredictable numbers.
Beacons have been discussed somewhat in the literature.
The classic sources are things like state lottery numbers or
horse race results. Since the stock market is on line,
typically with 15-minute delays, there's a moderately large amount
of "random" data available; it can be manipulated, but not very precisely,
and it generally costs more to do than the amount at stake in the
transaction using the data :-) If you're just using closing prices,
you can use the pseudo-random generator to pick a bunch of them,
or just use a pre-arranged set with enough expected entropy;
increasing the number of stocks involved also reduces the opportunity
for market manipulation. Alternatively, if you're using an
on-line source, you can pick a pre-arranged starting time and
stock to start with, allowing some time-slop.
# Thanks; Bill
# Bill Stewart, +1-415-442-2215 stewarts@ix.netcom.com
# You can get PGP outside the US at ftp.ox.ac.uk/pub/crypto/pgp
# (If this is a mailing list or news, please Cc: me on replies. Thanks.)
