There are actual reasonably-decentralized systems out there and this isn't one of them.
Can you name any such “reasonably-decentralized” proof-of-stake system or are you only thinking of proof-of-work variants?
For some of the past discussion of why extant distributed ledger designs aren’t decentralized, c.f. also my comments in Theymos’ thread as posted by @Traxo, @theheolyn, @Hyperme.sh and @CRED.me in the linked thread.
One of the design goals for making a proof-of-stake system decentralized is to make the profit for stakers from decentralization competitive with the profit for them from centralized extraction. To this end, one of the first problems that no proof-of-stake (nor proof-of-work) system has solved is a decentralized, competitive free-market for transaction fees. You may remember I presciently identified transaction fees in 2013 as the Achilles heel of extant blockchains protocols. Voting is not a decentralization paradigm.[1][2] I believe I had (in 2016) solved this design problem and other design flaws of proof-of-stake.
[1] Some Iron Laws of Political Economics
[2] Nothing is Cheaper than Proof of Work: Money and Politics (archived)
At best this has promise of future decentralization.
Unlikely when the premise of voting to reward from a collectivized debasement pool has a math and logic which is unavoidably centralization. Any changes they have made to the reward algorithm since I wrote that linked blog are irrelevant. I can still show how whales can take more than their proportional share of the rewards.
P.S. Neither Ned nor the new lady running the daily operations have the necessary skills, knowledge, and intellect to achieve the game theory and economics models that decentralizing a consensus ledger requires. Heck even Satoshi wasn’t able to design a system that could remain decentralized as minted block reward declines as a percentage of miner’s block revenue.
Well given Satoshi was likely a group of researchers with a collective IQ much higher than yours and mine combined, I say it’s quite likely that it was designed to fall to oligarchy (i.e. centralized) miner control at the end game (NWO, 666, etc).
And Monero doesn’t solve the problem with certitude given its relatively minuscule tail reward, for many different reasons, one of which includes a huge carrot incentive to centralize the mining so as to deanonymize Monero’s honeypot. I do think Monero is a honeypot, as is Tor. Remember way back I was arguing to you (and other Monero faithful) that Tor and I2P are honeypots and you guys shook me off like a bad case of fleas. And I ended up vindicated.
Underlying problem is that proof-of-work mining isn’t a consistently profitable endeavor (listen to Richard’s video) which means it must be monopolized. Mining is high-risk similar to farming, the farms get aggregated because the smaller farmers eventually go bust (due to bad weather, blight, etc). This is why Bitmain was for example was toppling the applecart (and ended up apparently destroying themselves).
Also it’s not correct to claim with certitude that a hypothetical future centralization of Monero’s mining will not deanonymize current and past transactions because:
- We don’t know when the mining has become centralized, might have already occurred in the past.
- And/or even unmasking future Ring signature transactions can aid in combinatorial unmasking of past Ring signatures wherein the future transactions share Ring signature inputs with the past ones.
Along with numerous other logic errors I caught him making, @dinofelis isn’t cognizant that theoretically a 51% attack can deanonymize Monero’s ring signatures and Grin’s MimbleWimble because the miner can pay himself transaction fees and spam the mixes with his own transactions thus unmasking who is in the mix other than himself:
As to the problem of censor resistance, I think that not decentralization, but anonymity is the answer.In Monero for instance, even if mining were done by one single entity, it couldn't censor any transaction, because there's no way to know which one to censor. In Monero, only the payer and the payee know the transaction. You can't sensor "an address".
Although the Zcash-like zk-snarks/starks hypothetically can be structured to eliminate the vulnerability, there’s nothing stopping @dinofelis’s hypothetical future world government from requiring an auditing key exclusively for the said world government which the world government can then use to enforce said censorship.
EDIT: C.f. also my comments about proof-of-stake in the context of Ethereum’s failed scaling research.
I'm not aware of any (confidently) reasonably decentralized proof of stake systems, however the non-delegated varieties are at least somewhat more decentralized since they lack a by-design centralization feature (for better or worse).
To claim DPoS is decentralized when one stakeholder is known to control 35% (i.e. far more than 19% vote received by the highest voted witness) of the of the voting stake is quite absurd. In other cases it might be unproven or unprovable but is at least less absurd on its face.