For example, no one has solved the mining centralization dilemma yet.
Very sorry for the off-topic question, but there have been talks in Darkcoin forums about making all mining go through masternodes, i.e. every masternode would be a p2pool node. Those would be the only blocks that would be accepted. In your opinion, would this decrease the mining centralization and the 51% risk? Currently there are ~1000 masternodes, and the goal as I understand it is to get that number to 2000-3000 eventually.
What is the economic incentive for running a masternode?
Currently 20% of the block reward goes to masternodes, and the percentage is going up 5% every month until equilibrium for target number of masternodes is found. Eventually probably 50% of the block reward goes to masternodes and 50% to the miners (just my guesstimate). And will there be other services that generate value for the masternode owners, is not clear yet what ideas people can come up with.
GHash.io charges no fees to miners because it is subsidized by an ASIC miner which mines on it, that is why it consistently gets near or over 50% of the network hashrate.
If you charge for something, the fiat overlords can always use transfer pricing and debt to take control of that resource.
As I told rpietila in his thread, I am in the unique position of having both technical and macro economics expertise. Smooth is also very astute, as you can see.
With the proposition there would be no 3rd party pools, or incentive for people to go through GHash.io or any other mining proxy, because they'd have to mine on masternodes anyway which enforce (or don't enforce) the fee. There would probably be no use for a fee anyway because the percentage of the block reward is issued to masternodes already by the protocol. Do you think this has any merit at all compared to 2-3 pools having > 50% combined?
Miners who can use a pool which doesn't take some of their block reward, will have an economic advantage, thus masternodes will become unfunded and be centralized (offered for free) by whomever can gain from stealing the anonymity (or what ever value they can extract from centralizing masternodes).
Blocks that don't issue masternode reward are rejected, so why would people start mining on a pool whose all blocks become orphaned?
Masternodes can refund this to miners. There is an extra incentive because masternodes have all the anonymity information.
You can't stop the rich and the government from aggregating resources. They always find a way.
The only way to stop them long-term is to make the resources incompatible with aggregation, e.g. the government would have a very difficult time owning all the human actions we do as they are inherently diverse.
There's still the problem for the attacker to get people mining to his (master)node, because until he actually has 51% of the hash, all his blocks will be rejected. And the masternodes ban "misbehaving" masternodes already, if they detect that one masternode is generating invalid blocks they would ban it from the network. The banning is not voted by hashrate, but by the 1000 DRK stake each masternode holds. Seems like an attacker would need 51% of the masternodes as well.
It's not perfect, and it's unclear whether a perfect system is even possible (maybe you already have thought of one, and I hope you have, but it might be far into the future). In the meanwhile, the question was, "would this decrease the mining centralization and the 51% risk?".