From: gtoal@an-teallach.com (Graham Toal)
Date: Sat, 26 Mar 94 11:57:07 PST
To: cypherpunks@toad.com
Subject: Re: Whoa, now... (was Re: Digital Cash)
Message-ID: <199403262043.UAA03969@an-teallach.com>
MIME-Version: 1.0
Content-Type: text/plain


:From: "James G. Speth" <december@end.end.com>

:In fact, he began this thread by citing someone who was talking about using
:an obscured algorithm to prevent digital cash double spending.  (ie. If you
:can't get to the algorithm, you can't cheat the system.)  His comments were
:on the dangers of relying on this.

:That's the point.  Mikolaj was _never_ referring to cryptographic security.
:He was pointing out how security through obscuring algorithms can never be
:considered reliable.

Oh, well if that's what he meant, we agree completely.  I thought he was
saying his el33t hackerdoodz buddies could break the crypto part of ecash.

By the way, the reason I've never discussed the ecash threads on this
group is because it was obvious right from the start that double-spending
makes the schemes unworkable, and that only a central reference authority
could patch the system to make it work, which (in my seldom humble opinion)
entirely negates the point of these schemes.

My suspicion is that anonymous ecash can only be made to work if giving
the tokens to someone else is a destructive operation - the way core
memory (I'm probably one of the few people on this group who've had to
worry about stuff like this :-) ) used to be erased when read.

Except that these days, the destructive read would have to be something
secured by the laws of physics, like say a quantum state being trashed
by virtue of being observed.  (The way secure quantum comms works - if
we had something akin to a quantum delay line in a card, maybe that
would work.)

However, the technology to do that sort of stuff is probably centuries off.

G