RE: Seeking Consensus on Consensus - DPOS or Delegated Proof of Stake and the Two Generals' Problem

You are viewing a single comment's thread from:

Seeking Consensus on Consensus - DPOS or Delegated Proof of Stake and the Two Generals' Problem

in eos •  7 years ago  (edited)

Interesting history and proposal...
Anyway I have some disagreement, or at least some possible dissent.

The way miners are described is not rich enough.

Today miners are working in pool, farmers pool to follow your metaphor.
Unlike medieval world, they select on internet their depute, their pool, able to change at will.
If they don't like the policy of the pool, about farming, about wheat stock management, they can change of pool, or create a new pool like some created a workers union.
PoS maye make the choice harder, as there will be locked stake conserved for long time.
Your idea of vote is interesting, and if it is incentivized, it will be like paying the people to vote.
Best would be to create a mechanism to enforce variety in voters, and deter scaling.
Contrary to you I suspect PoS will increase (cultural, national,organizational) centralisation following the same problem finance encountered, leading to anti-trust laws, and to subprime crisis, creating huge systemic risk, bigger than the "Chinese Miners Fear"...

This market of pool is a bit like the democracy of landlord you describe, but not with asset, but with mining power.
Of course there is big farmers, exploiting big farms, but it is much harder to build 1Mn$ of farm, than to buy 1Mn$ of ETH.

Beside that there is a currency risk in killing miners.
the value of a currency relative to others currencies is related to the exchange with others, with the prices, the import.

Miners pay GPU and energy in fiat currency, and their behaviors is pegging their income to what the GPU and energy cost.
If there is no activity liked to a fiat currency, the price will purely float as a speculative value.
Cryptocurrencies are not so far from the market of art, and the luxury expected by gallery managers, and artists, set a price, like the miners set the price. it can grow higher, but prevent lowest prices.

Remove the miners paying ETH mining capacity in dollars, and there will be no need for ETH to have any minimum value.

I suspect that the miners investing in capacity , forgetting to anticipate their collective work to increase difficulty, create an increase in crypto currency value, as to compensate the difficulty increase they need increase in asset value, and freeze their crypto assets until they can cashin some benefit.

The tend may look weak, but since there is no other relation to reality, miners may just be breaking the symmetry of speculators behaviors, who just enjoy self-realizing predictions.
Miners have no symmetry in their prediction, they predict a benefit, and refuse to sell until then?

If PoW is a bit like farming, PoS is a bit like loaning, with the common fact that participant share the same income.
PoW power is paid in fiat currency, so mined crypto is pegged to real world that way.
PoS power is paid in the same currency of the income, so there is no pegging to any reality.
Only if the currency is pegged to another reality asset can the currency have a stable or growing value.

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
Sort Order:  

Dissent :-)

I think in the big picture it is fine to try DPoS against PoW because the market will tell us which is right. In the small picture we don't want to waste so much value in running that experiment so we really want to get it as right as possible. But even then, there is a potential benefit: if we put up the first governed & open chain, there will be enough people who want that over an ungoverned open chain that we'll attract an audience.

The big thing about using DPoS is not that it changes the nature of the interactions but it returns the mining tax back to the community. We don't have to pay for mining! That's 4% bounty right there, or 11% if you're into Ethereum. That's a powerful amount of money, and it is what drives the ability to deliver free transactions as well as pay for development and other things. However it's done, that's the big transfer we are looking to effect, and the game theory aspects of mining have to be near perfect to fight against that.