Quote from: justusranvier on July 05, 2014, 11:27:15 AM

Quote from: kodtycoon on July 05, 2014, 10:38:22 AM

that mechanism is far less than optimal for many reasons and excludes those who don't have mining gear and favours those with higher powered mining gear. It's no different to "rich getting richer" in nxt. In that respect, pow is elitist

This is the problem with proof of stake: it was invented by people who have no idea what problem mining is supposed to solve and have some agenda other than solving that problem.

Mining is not about allocating the issuance of new coins. The fact that they are tied together in Bitcoin is a temporary coincidence. Mining is about solving the problem of distributed consensus - how do a bunch of independent nodes spread all over the planet agree on a precise ordering of transactions when every node must operate with an incomplete view of the network and anybody might be trying to cheat?

This problem has nothing to do with elitism or notions of fairness or populism. Overlaying those agendas into the solution is a great way to not solve the problem.

As nodes on the network continually work to establish a consistent of narrative of what has happened in the netwowk based on their own incomplete knowledge, there will be times where two nodes disagree. Mining is nothing more than a signalling mechanism which provides an objective basis for choosing which version of history to treat as correct, whenever a conflict occurs such that more than one alternative version exist.

The design criteria for what makes a good mining algorithm comes from signalling theory:

Quote

Quote

Two individuals have access to different information.

They could both gain if they could honestly share this information.

However, their interests do not coincide entirely, and so each has an incentive to deceive the other.

How can honest communication be ensured?

How can honest communication be ensured despite conflicting interests between a signaller and a signal receiver?

Economists and biologists independently proposed that the costs associated with producing signals can provide a solution to this problem. Loosely paraphrased, the solution typically takes the following form.

Quote

Suppose that signals are costly, and that for one reason or another, lies cost more than honest signals.

If telling the truth is cheap enough and telling a lie is costly enough, it may be worthwhile to communicate honestly but not to lie.

There's a reason that when Wei Dai proposed b-money in 1998, he didn't even bother to explain why calculations in a proof of work system, "must be easy to determine how much computing effort it took to solve the problem and the solution must otherwise have no value, either practical or intellectual." He assumed this statement would be so obviously true that no explanation was needed. Apparently this is no longer the case.

The signal sent by proof of work is the amount of opportunity cost the miner has paid in order to produce the block. The fact that mining calculations are completely useless outside the signalling system itself is what makes lies more expensive than telling the truth, thus satisfying the conditions for honest signalling. The opportunity cost the miner pays to produce a block only represents a profitable trade for the miner if the network accepts their block. So when it comes to a node in the network choosing between two valid blocks, choosing to accept the block with the higher PoW means choosing the block which produced by the miner who has the most at stake in terms of opportunity cost paid.

Note that if the miner has to use specialized hardware for which there is no possible use other than mining, the signal is even better than performing otherwise-useless calculations on general purpose hardware. Higher opportunity costs = more reliable signal.

Proof of work is a proof of stake system, the only one that actually works.

PoS coins use the number of coins held as the basis for their signalling system. Since coins have an exchange rate, they obviously do not fulfill the criteria of having no value, either practical or intellectual. Thus PoS is not an viable mechanism for honest signalling.

Thank you for answering that roll your eyes assertion. I clearly did not have the energy to do so. You did.

Great explanation.