On Friday, April 25, 2003, at 03:51 PM, Tim May wrote:
Regarding "digital notes circulating in the wild without server contact," you need to look at some of the articles here (Cypherpunks) from around 1994-97 on "money changing."
Cf. articles by Doug Barnes, Ian Goldberg, myself, and others. Accessible via Google.
Basically, there is no reason why intermediaries will not develop who agree to take in digital money and issue new digital money, for a fee.
The operation of making change is just this.
I should have also made clear that the digital notes do not circulate around and around, without server contact (redemption). For one thing, such circulation would a) expose the digital numbers to copying by intermediaries and b) defeat the idea of untraceability. And if the note were not redeemable, the "stuckee" would have no recourse unless the identity of the links were known (and measures could be taken, blah blah). It is best to think of a digital note as a _relationship_ between a and b: aRb. An arrow. A transfer relationship. Alice sends to Bob a digital money token. What he does with it is another transaction (canonically, he sends it to a bank, and, if it is redeemed, he is satisfied. He may also be a bank, or the digital money token may be a form of money he recognizes, as with a remailer token, a stamp). It is best not to think of there being any intermediate steps. That is, any two nodes linked by an arrow have no other nodes between them. The token does not get passed from hand to hand to hand, no matter how complex a series of transactions is. (To do so invites copying, which leads to the double spending problems so often discussed.) (Digression: Even actual folding money works this way, basically. Alice transfers money to Bob who transfers it to Charles, and so on. Of course, with digital money the same token is not transmitted this way. Each stage effectively reissues the money (or Bob "redeems" the money at a bank, which is a special, terminal case).) "Money" is a loaded term, conjuring up various and often-contradictory images of paper notes, bullion, coins, IOUs, personal checks, cashier's checks, warehouse receipts, bearer bonds, drugs, artwork, wire transfers, SWIFT transfers, etc. The relationship R for money is something which needs to be discussed at more length: there may be forms of R for small value or coin-like uses, for medium value banknote-like uses, or even for high value bearer bond-like or bank-like uses. Just as there are many forms of non-digital money, for various uses and with various levels of security and authentication, so too must one expect various kinds of digital money. First class objects are critically important here, but not in a "one size fits all" sense. (Not sure if this is clear or not...as I said, much more needs to be said.) One of the interesting properties of the relationship R is that it involves _belief_. This is really what money is all about. The fact that one's belief that a $20 banknote with the right Andrew Jackson portrait on it is "real money" is only an expression of one's belief that the odds of it not being accepted by some other party, or by the U.S. Treasury, is close to nil. In areas where banknotes are more commonly forged, and thus not accepted, such a belief would be naive. And so on for various other forms of money. Even digital money. Which gets us into reputations and ping systems (with blinding, an issuer can decide to "burn" (renege on) particular users, which makes repudiation difficult and not something banks which wish to stay in business will lightly do). Properties of these graphs (or, in certain interesting cases, lattices) are crucial to understanding digital money. --Tim May "The Constitution is a radical document...it is the job of the government to rein in people's rights." --President William J. Clinton