Topic: Proof of Stake - page 3. (Read 16439 times)

Right, seems like I still haven't got the hang of your system.

Starting p at 0 and letting it gradually grow to the target value seems like it could work, but there's still a nontrivial bootstrapping problem. I guess that a faucet to get a small stake to start could mitigate it. p should remain very small until there's a somewhat decent exchange.

There is a form of PoS that could work if it is possible. What if you could choose the mining pool you want to authorize your transaction through the P2SH script?

We would also need a demurrage tax for the fees earned by the pools. While the pools can auction their service competetively, they will still lose a percentage of their fee profits back into minable blocks. The tax would keep mining desirable, the auction would keep fees low, and the stake would be controlled by the people spending their Bitcoin. It would be less attractive for a monopolist to make a lot of transactions and pay the demurrage tax.

This wouldn't technically be a PoS, but would allow consumers to choose to avoid pools that threaten to reorganize the blockchain. I'm just not sure how to implement and verify a mandatory transaction tax. I suppose it would be visible in the blockchain if it was paid.

An offset introduces decreasing returns to scale (DRS).  With DRS, output per unit time increases when you subdivide accounts. Fixing hashing power, mining with 100 accounts with 0.01 bitcoins in each has a higher output rate than mining in 1 account with 1 bitcoin. Nevertheless, an offset works too. It just needs to be phased out completely after the initial distrubition period is over because DRS allows gaming of the system.

A better solution is to add an offset. Instead of giving stakeholders an improvement factor of s^p (s is the stake and p is the influence level of stake on mining), make it a factor of (a+s)^p where a is some constant. So early on, when nobody has a large stake, mining will be proportional to hashrate without much consideration of stake. Also later, even someone with low or no stake can still mine somewhat.

First off, a more immediate problem than adoption is that no one with appropriate skills has volunteered to modify the bitcoin code to make a proof-of-stake altchain or minority fork possible. The modifications should be very simple, but I cannot do them because I am an economist with no programming skills.
 
Secondly, your idea works and sounds good. Let's consider the starting result where p starts out at 0.01 and increases by some small fixed amount or percentage with every block until it reaches some long-term target.

The initial block finder would have mining power at block 2 equal to:
50^0.01/(50^0.01+9*0.000001^0.01)=11.7% of the network. The other 9 miners would each control 9.8% of the network.

Thus we no longer have a winner take all lottery. I agree that this is probably a better system than a winner take all lottery (though I still find the winner take all situation amusing). Moreover, it should be extremely simple to code this.

An instant monopoly?  Faith and begorrah!   


What about setting the PoS weight (your "p" value?) very very low at first, and growing it over time?  Couldn't you make it behave like the current system's growth of the BTC money supply, where it approaches some arbitrary value (80%) as the system matures?  Would this not address the initial distribution problem in the hypothetcal start-up chain?  Might growing p over time offer a solution to a different problem you're facing with implementing PoS in the current system? (adoption)

Now I realize I'm coming from a position of ignorance on these matters, so I'm not advocating anything...just asking questions.   

The lucky miner's percentage of the network is approaching 100%.
 
I'll calculate it for you:

Long-run hashing power of individual holding 50 BTC and x units of hashing power
(50)^0.8*(x)^0.2= 22.87x^0.2

Long-run hashing power of individual holding 0 BTC and x units of hashing power. In this case the individual has hashing power of 100-satoshis or 0.000001 BTC-confirmations.

1.58489319*10^-5*x^0.2

Adding up these 9 miners and rounding, they have
9* 1.58489319*10^-5*x^0.2 = 1.4*10^-4*x^0.2

The share of the lucky miner is thus
22.87/ (22.87+1.4*10^-4) = 99.9993878%

If you don't like this extreme lottery style distribution arrangement, then you could set it up with some constraints on the use of recently minted coins. Essentially you would require that newly generated coins have y confirmations on them before they can be sent or used to mine blocks. This would give the other 9 guys a time span of y blocks to establish a stake too. The initial lucky miner still has an advantage, but it is much, much smaller.

I think that the y confirmation limit on use of newly minted coins is definitely a better arrangement, but I am not proposing stuff like this now because I want to focus on the core issues (which are long-run issues not initial distribution issues). I don't think initial distribution matters much.

I also am curious about what would happen in the extreme lottery scenario.

The guy who has 99.9999% of hashing power has a strong incentive to sell off or give away almost all of his coins until he has say 10 or 20%. His 99.9999% of hashing power is not going to be worth anything if the other speculators drop out of the game. Instead, he would want to ponzi up the currency. First giving away coins to one group of speculators (most likely giving away more than half at this stage), charging the next group a very very low price, the third group a very low price, the fourth group a low price, ... until he has a small amount left like 10%.  He will have to be willing to use some of his initial revenue to buy back coins if price fails to rise. If the downward price pressure is really strong, then he should let the bubble pop and sell on the way down.

Initially, the speculators perceive that the value of the currency has steadily risen, so they want to invest in the bubble. Sale to new investors gives the initial miner his profit. The early stage miners who bought in low profit too. I'm not sure how long the ponzi would continue until the bubble collapsed. However after the bubble collapses, ownership in the currency would be quite widely distributed.

Question...let's say 10 miners initially participate in this start-up, they all have identical hardware, they all start mining at the exact same instant, and PoS is weighted at 80%. When the first block reward is collected by one of the ten, what has happened to his percentage of the network?

(yes, I'm computationally challenged)   

If indeed "advantages associated with control of computing resources" is decreased with PoS, then oligopolist conspirators using botnets would certainly be able to create a very large shell game. They would be thoroughly decentralized.

Yes, PoS would reduce the attractiveness of bot mining in at least one way:

1) [primary] PoS would simply decrease the advantages associated with control of computing resources such as botnets.
2) [marginal] Bots would need to communicate with an account holding the bitcoin in order to hash effectively. This might make bots slightly easier to detect/trace, but the owners of victimized computers probably wouldn't know how to do this, so I doubt this has any practical relevance.

Yes, this is what I mean by the Ponzi-like properties.

Yes, each iteration corresponds to a block mined. The probability you draw is the probability of you being the miner of that block and (all else equal) it doubles when you double your hashing power.
I think you are right that it is okay as is, but i will still play around to see if anything changes if I convert from [one iteration = one block] to [one iteration = one minute]. It should yield the same answers.

The start-up of such a chain could be interesting. First person has no stake. Each new person has no stake unless they buy it from an earlier person...

-MarkM-

You are also right. The length of hashing rounds should be a Poisson random variable, whereas I made them deterministic. However, this should be coded differently from what you suggest. If you write it as conf(2)=conf(1)+cX, then this implies that the miner randomly misses the opportunity to participate in mining for a sequence of blocks. Instead, it should be that the miner can get multiple hashing opportunities before accumulating another confirmation. I will fix it when I get the chance.