[cryptography] Digital cash in the news...

Ian G iang at iang.org
Sun Jun 12 23:27:54 EDT 2011

On 13/06/11 12:05 PM, James A. Donald wrote:
> On 2011-06-13 9:26 AM, Ian G wrote:
>> However. Unless the laws of financial conservation have been repealed by
>> the design, those who follow have to invest a lot and come out with
>> less...
> Financial conservation does not apply to money.

Right, not to money.  But it applies to value which was what earlier 
poster was talking about.  He had allegedly put small value in and got 
large value out.

In economics terms, although a monetary unit strives to be a store of 
value, sometimes it fails to do that.  Value being distinct from the 
price at which any particular monetary unit trades at.  (Monies tend to 
be better units of exchange, perhaps simply because we can do math these 
days, and they hold for at least instant trades.  And people tend to 
confuse the price with value.  And confuse money with value...)

The laws of economics are like gods.  You don't have to subscribe and 
obey them every day, nor sacrifice your children to them on a yearly 
schedule.  But you have to respect them and be careful not to anger them.

More on point for BitCoin, without a gesture towards the god of store of 
value -- some fundamental relationship to something that humans value -- 
then the unit flaunts itself at will, flying like Icarus as high as it 
feels.  As we can see, it flies skywards as the crowd rushes in (notice 
the upticks in media discussion), and it will collapse just as quickly 
when they rush out again.

> If paper currency
> collapses, and is replaced by gold, those who invest in bitcoin will
> come out with nothing. If paper currency collapses, and is replaced by
> bitcoin, they will come out with immense fortunes.
> The market is at present rating the prospect of the world going to a
> bitcoin standard rather than a gold standard at two chances in a
> million, which seems reasonably conservative.

:)  I don't disagree, opinions on gold, etc, as exercises for the reader.

Back to crypto (soon).

How to do this properly?  Well, the simple way is to elect to denominate 
the unit in some alternate well understood other monetary unit.  So for 
example, the typical thing is to denominate in USD.  You have a field in 
the packets that says "USD" and each 1 is worth a dollar.  You have 
another field for however many of those, say 10.  Simple.

We can therefore see that someone has to make that "worth" mean 
something, so for this we need an "issuer" sometimes known as Ivan. 
It's beyond the scope of a crypto list to discuss this in depth, but 
typically Ivan would deposit $1 for every issued electronic dollar in 
some bank account somewhere.

A more complicated way is to /describe the value/ so for example also 
describe how your deposits weren't to be stolen.  Imagine that you 
overcame the obvious objections to the above by saying you were going to 
bury gold bricks in your backyard on your private island (as mentioned 
in some novels).  The way to do this is a /contract/ which is a defined 
format of promise to deliver, date, consideration, etc.  You write that 
down in boring ascii:

     "I promise to bury one gram of gold
     in my backyard at approx geoloc X,Y, and
     promise to redeem that on presentation
     of one electronic gold gram.  Etc etc."

*Now we get to the crypto*  How do you make that contract work in 
digital form?  Well, you have to /agree/ and you have to be /seen to 
agree/ by your holder-buyer-customer-user.  In detail, every time, 
unwavering, unforgiven.  So:

    * add in some details (elided here).
    * append your public key to the end of the document.
    * sign the document in cleartext using standard digital signature
      (OpenPGP works well, x509 can be easily hacked to do the same).
    * take a message digest of the resultant signed document.
    * stick the MD in every packet as the indicator of which contract.

Then, the act of making the first payment(s) of digital issuance 
includes your MD which includes your PK sig, which then all entangle 
into every subsequent payment.  This process creates an undefeatable 
chain of evidence over your contracted promise, as well as locking down 
a whole host of other details such as stopping anyone inserting false 
payments into the entangled hash sequence.

Hey presto, the cryptographic signature finally comes good as a 
describer of value, and makes payments really work.


PS: google on Ricardian Contract for more.  It's an open concept.  It 
uses a sensible dollop of crypto over a base of classical governance.

More information about the cryptography mailing list