Quote from: jtimon on February 21, 2011, 01:05:37 PM

My proposal was to set the rules for valid transactions so that any user in the network has to accept the IOU's created as fees to the node even if they don't trust the original issuer of the IOU. Once these IOU's belong to a subnetwork that trust the issuer, those IOU's can be treated as normal and be destroyed by coming back to the issuer. The flaw in this is that if the issuer address becomes isolated, then the IOUs against that address are not redeemable and would stay in circulation forever.

This sounds like a system in which assets, rather than liabilities, are a circulating medium of exchange. By saying that someone else's IOU must be accepted by others you are inverting the normal construction of any debt based system on its head. That can only work if the original issuer of the IOU IS trusted. Again, it comes back to some notion of reserve currency, which is no more really than saying this nodes debt is the best of all the nodes. In fact it is the same as a fiat currency regime in which IOUs written by the government are to be accepted by everyone els,e and of course there are laws to ensure this is the case.

One way to deal with that flaw is to have the fees charged up front. The successful computer of the block is rewarded at the instant the block is solved and broadcast with a new coin. Assuming that block makes it into the longest chain then that chain also incorporates the validation of his coin.

Obviously this rubs against the built-in deflation of bitcoin but it is worth examining the economics here. Essentially one can have a node-incentive provided as an IOU redeemed after the fact in which case all transactors (buyers and sellers) give a small part of their purchasing power to the solving node, or you have the same fee charged in the form of inflation to all transactors. Except, not quite because with the latter only holders of debt are charged, not debtors. That can be balanced by having interest fees on outstanding debt, but that would be a matter for individual debtor-creditors to set for themselves.

Lastly, could it be done alternatively such that each time a coin gets used in ANY transaction it has its value decremented. That is, it is clipped and the clipping discarded. Eventually there would be nothing left of it. This would counter the inflationary impact of new coins. It also has the nice side effect that if you don't spend a coin it doesn't lose any value - it just loses value when it circulates. Essentially it has the same outcome as paying fees to the solver, but without all the troublesome loose ends.

Quote

The IOU's issued here could be denominated in bitcoins. This way it would be easier for the nodes to calculate how much they earn by helping "ripple-coin" compared to what they earn helping bitcoin.
Each new bitcoin block generates 50 new bitcoins and not 1 (anyone correct me if I'm wrong).

Just to be clear, I am suggesting that there are no IOUs for fees, just new bitcoins in lieu of fees. I am suggesting a hybrid which does away with the 21M coin limit and perhaps uses an eroding coin to balance inflation. If 50 coins are made with a new block, that works fine. After 50 uses the coin will have no value left (effectively, it has all gone in fees).

[quopte]
What you're suggesting, I think, is a hybrid system with money based on debt between trusted parties (ripple's IOUs) and fiat (or electronic commodity backed if you wish) money.
I thought about that too, but finally I decided I don't like it.
[/quote]

well you should say why. There is nothing wrong with the fiat system except for its centralization. Having liquid cash that is accepted by all for clearing debts and for performing instant transfers of value for important stuff such as proof-of-work fees without loose ends is advantageous, especially if the way such currency is brought into being is through doing publicly accepted crypto work and the way the currency is retired is also encoded into the system in a decentralized manner.

Quote

What a pity that no one seems to be interested here in the bitcoin community.

Ryan did warn that that might be the case I think!