(this is a Java fragment, I'll provide a working Java program to play with shortly)
creatorFactor = 0.8;
judgeFactor = 1.0 - creatorFactor;
baseClaimTotal = claimCreationCost * creatorFactor;
baseRewardPerCoupon = baseClaimTotal / averageMaxPairsPerClaim;
creatorReward = baseRewardPerCoupon;
sendReward(); // statistically conserved, of course! :-)
The exact final value of creatorFactor is not determined here, so either a voting round or a claim will be needed to determine its value. The examples below are for illustration purposes with the creatorFactor of 0.8 used above.
After the claim is retired, the judge will receive claimCreationCost * judgeFactor as the judgement fee from the bank. This would be fx$10 at current rates.
For example, a claim called DJ300K is proposed in 2009, to be judged
in 2010. Claim creation cost is still fx$50, making
baseClaimTotal = fx$40 (fx$50 * creatorFactor).
The historical average number of max pairs per claim
(averageMaxPairsPerClaim above) is 10,000 at the time DJ300K is
created. Each time a pair is created, the claim creator receives 40 /
10,000 or fx$.0040 per new pair. This would be handled in the
statistical manner referenced below, meaning that if one pair was
created, there would be a 40/100 chance of a $.01 credit being given
to the claim creator (assuming the baseClaimTotal in total payback had
not yet been reached). Note that if it were a 100 pair creation
event, the exact reward of $0.40 would be sent. So, as the first
10,000 pairs are created, the claim creator receives fx$40 (give or
take a few cents due to the statistical nature of the small
transactions and rounding). The claim proceeds to have a total
outstanding pair count of 18,000, which is good to very good
performance. For the final 8,000 shares the claim creator receives
8,000 * baseRewardPerCoupon (again with possible
statistical deviation) or fx$32, and the judge receives nothing.
After the claim is retired, the judge
receives the claimCreationCost * judgeFactor (fx$10) from the bank, for
a total profit of fx$10. The claim creator has received a total of
fx$76.50, for a profit of $26.50.
Now consider another claim, Dow310K proposed for the same period as Dow300K. Since Dow310K was an ill-considered claim (and greatly annoyed many people on fx-discuss) it received little attention in the marketplace and only created 2,000 pairs over its lifetime before being judged. Its hapless creator only received 2,000 * creatorFactor * baseRewardPerCoupon, or fx$8.50 for his troubles, meaning a net loss of fx$41.50. The judge received her usual risk-free fx$10.
(This next paragraph really needs to be discussed!
The claim creator will only receive 1/2 the reward for pairs in which he/she participated in the creating transaction (zero is an option as well, there should be discussion on this - owner created pairs could be left out of the averages).
This means that if a new claim reaches the average historical max number of pairs created (less the effects of claim creator participation;), it will exactly repay the claim creation cost. Claims that create fewer pairs will get less, those that create more pairs will get more. This will tend to balance statistically as long as the ratio of fx player investment to claim count is constant. If this ratio grows, the claim rewards will be somewhat higher. If it falls, they will be lower. Over time, this system will constantly be finding new equilibria.
This option uses the same statistical algorithm described in the option "...must assure that credibills are conserved statistically" (by [email protected]) to conserve fractional fx$.
Rationale: The fx economy needs stimulation. Good claims are the cornerstone of the game, and it makes sense to introduce 'economic growth' as claim creation rewards. This also means that using this system claim creation rewards are not a 'drag' on the fx economy, at which some have expressed concern.
Get some software which simulates this system. You'll need a Java 1.1 (or higher) system to run it.
Terry
