12.3.1. The nature of money, of banking and finance, is a topic that
suffuses most discussions of digital cash. Hardly surprising.
But also an area that is even more detailed than is crypto.
And endless confusion of terms, semantic quibblings on the
list, and so on. I won't be devoting much space to trying to
explain economics, banking, and the deep nature or money.
12.3.2. There are of course many forms of cash or money today (these
terms are not equivalent...)
+ coins, bills (presumed to be difficult to forge)
- "ontological conservation laws"--the money can't be in
two places at once, can't be double spent
- this is only partly true, and forgery technology is
making it all moot
- bearer bonds and other "immediately cashable" instruments
- diamonds, gold, works of art, etc. ("portable wealth")
12.3.3. Many forms of digital money. Just as there are dozens of
major forms of instruments, so too will there be many forms
of digital money. Niches will be filled.
12.3.4. The deep nature of money is unclear to me. There are days
when I think it's just a giant con game, with value in money
only because others will accept it. Other days when I think
it's somewhat tied to "real things" like gold and silver. And
other days when I'm just unconcerned (so long as I have it,
and it works).
12.3.5. The digital cash discussions get similarly confused by the
various ideas about money. Digital cash is not necessarily a
form of _currency_, but is instead a transfer mechanism. More
like a "digital check," in fact (though it may give rise to
new currencies, or to wider use of some existing
currency...at some point, it may become indistinguishable
from a currency).
12.3.6. I advise that people not worry overly much about the true and
deep nature of money, and instead think about digital cash as
a transfer protocol for some underlyng form of money, which
might be gold coins, or Swiss francs, or chickens, or even
giant stone wheels.
12.3.7. Principle vs. Properties of Money
- Physical coins, as money, have certain basic properties:
difficult to counterfeit, pointless to counterfeit if made
of gold or silver, fungibility, immediate settling (no need
to clear with a distant bank, no delays, etc.),
untraceability, etc.
- Digital cash, in various flavors, has dramatically
different properties, e.g., it may require clearing, any
single digtital note is infinitely copyable, it may allow
traceability, etc. A complicated mix of properties.
+ But why is physical money (specie) the way it is? What
properties account for this? What are the core principles
that imply these properties?
- hardware (specie like gold) vs. software (bits, readily
copyable)
- immediale, local clearing, because of rational faith that
the money will clear
- limits on rate of transfer of physical money set by size,
weight of money, whereas "wire fraud" and variants can
drain an account in seconds
- My notion is that we spend too much time thinking about the
_principles_ (such as locality, transitivity, etc.) and
expect to then _derive_ the properties. Maybe we need to
instead focus on the _objects_, the sets of protocol-
derived things, and examine their emergent properties. (I
have my own thinking along these lines, involving "protocol
ecologies" in which agents bang against each other, a la
Doug Lenat's old "Eurisko" system, and thus discover
weaknesses, points of strength, and even are genetically
programmed to add new methods which increase security.
This, as you can guess, is a longterm, speculative
project.)
12.3.8. "Can a "digital coin" be made?"
- The answer appears to be "no"
+ Software is infinitely copyable, which means a software
representation of digital money could be replicated many
times
- this is not to say it could be _spent_ many times,
depending on the clearing process...but then this is not
a "coin" in the sense we mean
- Software is trivially replicable, unlike gold or silver
coins, or even paper currency. If and when paper currency
becomes trivially replicable (and color copiers have almost
gotten there), expect changes in the nature of cash.
(Speculation: cash will be replaced by smart cards,
probably not of the anonymous sort we favor.)
+ bits can always be duplicated (unless tied to hardware, as
with TRMs), so must look elsewhere
+ could tie the bits to a specific location, so that
duplication would be obvious or useless
- the idea is vaguely that an agent could be placed in
some location...duplications would be both detectable
and irrelevant (same bits, same behavior, unmodifiable
because of digital signature)
- (this is formally similar to the idea of an active agent
that is unforgeable, in the sense that the agent or coin is
"standalone")
12.3.9. "What is the 'granularity' of digital cash?"
+ fine granularity, e.g., sub-cent amounts
- useful for many online transactions
- inside computers
- add-on fees by interemediaries
- very small purchases
+ medium granularity
- a few cents, up to a dollar (for example)
- also useful for many small purchases
- close equivalent to "loose change" or small bills, and
probably useful for the same purposes
- tolls, fees, etc.
- This is roughly the level many DigiCash protocols are
aimed at
+ large granularity
- multiple dollars
- more like a "conventional" online transaction
-
- the transaction costs are crucial; online vs. offline
clearing
- Digital Silk Road is a proposal by Dean Tribble and Norm
Hardy to reduce transaction costs
12.3.10. Debate about money and finance gets complicated
- legal terms, specific accounting jargon, etc.
- I won't venture into this thicket here. It's a specialty
unto itself, with several dozen major types of instruments
and derivatives. And of course with big doses of the law.
By Tim May, see README
HTML by Jonathan Rochkind