From: Jim Choate <ravage@ssz.com>
Date: Sat, 13 Sep 1997 21:45:38 +0800
To: cypherpunks@ssz.com (Cypherpunks Distributed Remailer)
Subject: Re: Monetary Economics (fwd) [amusement]
Message-ID: <199709131351.IAA25585@einstein.ssz.com>
MIME-Version: 1.0
Content-Type: text



Forwarded message:
> From owner-traveller@Phaser.ShowCase.MPGN.COM Sat Sep 13 07:01:15 1997
> To: traveller@MPGN.COM
> X-Original-Article-From: Scott Ellsworth <Scott_Ellsworth@alumni.hmc.edu>
> Subject: Re: Monetary Economics
> From: shadow@krypton.rain.com (Leonard Erickson)
> Message-ID: <970913.000519.9C4.rnr.w165w@krypton.rain.com>
> Date: Sat, 13 Sep 1997 00:05:19 PST
> In-Reply-To: <3.0.3.32.19970912130653.007c83c0@mail.deltanet.com>
> Organization: Shadownet
> X-Newsreader: rnr v2.20
> Sender: owner-traveller@Phaser.ShowCase.MPGN.COM
> Reply-To: traveller@MPGN.COM
> 
> In mail you write:
> 
> > For this to work, the entity doing the credit management needs to have the
> > highest technology available, which I assume the Imperium has for a very
> > long time.  The techniques are likely known to those who want to find them
> > out, because someone would have defected.  Thus, you need a scheme that is
> > robust, even knowing the algorithm and some of the keys.
> 
> Old rule of cryptology. No code/cipher is secure unless it can provide
> protection against someone who knows the algorithm, but not the key.
> This is because it is *inevitable* that the algorithm will either leak,
> or worse, be guessed. 
> 
> If a key is known, that key is blown. But again, knowledge of one key
> should not compromise others. If it does, the code system is a piece of
> junk.
> 
> Theoretically, one time pads will *always* be secure. The only trouble
> is key distribution (well, key generation is tedious, but that's
> relatively minor).
> 
> Public key ciphers are subject to mathematical advances. Not
> *computational*, because if you have faster computers, then you can use
> bigger keys. But if someone comes up with a new mathematical technique
> that greatly decreases the work required to carry the mathematical
> operation that "secures" the cipher, then you are toast. For example, a
> whole bunch of public key schemes went down the tubes a few years back
> when somebody came up with a better solution to what's known as the
> "knapsack problem". The remaining schemes are based on the difficulty
> of factoring large numbers. So if a breakthrough is made in algorithms
> for factoring large numbers, they go away.
> 
> But for game purposes, we can assume that factoring is inherently hard
> (or that some new trick is found). So public key ciphers would be
> usable, though the keys may be a small book's worth of digits (stored
> in a tiny bit of storage). 
> 
> Given the fact of there being higher tech cultures, one time pads will
> be used for military and diplomatic stuff as well as anything that you
> are afraid might be trouble if someone with higher tech shows up.
> 
> But for generic business purposes, you just use the highest TL
> available for the public key ciphering gear. That way, it's unlikely
> that anybody can crack things soon. Sure, if a TL 16 race is found,
> they can crack your ciphers in weeks instead of years, but they aren't
> likely to *bother* with most commercial stuff.
> 
> -- 
> Leonard Erickson (aka Shadow)
>  shadow@krypton.rain.com        <--preferred
> leonard@qiclab.scn.rain.com     <--last resort
>