There is a much easier way for the government to take over Bitcoin than 51% attack and selfish mining. Simply tax the coin and per footnote [2] all coin is tainted regardless if you used Tor or not, so in effect regulate every pool and exchange. Regulation can accomplish much of the same goals as takeover. Then later the large corporations beholden to the government buyout the pools and exchanges, and then the drift towards cartels and corporate-fascism.
Don't you think that P2P mining pool can solve the problem of regulation or big corporations takeover?
Definitely not since it is so easy for the centralized pools to have their miners compute shares for a competing P2Ppool but withhold block solutions from the P2Ppool (and submit them for themselves) thus parasiting the P2Ppools of revenue while not impacting the revenue of the centralized pool. So centralized pools which wanted to eliminate P2Ppools could do so (in theory).
I will quote myself where I answered this in the past.
https://bitcointalk.org/index.php?topic=339902.msg3715641#msg3715641Why don't you just use P2Pool? Is there any reason?
I was waiting for bytemaster to answer because I wanted to know his thoughts. Seems to me that you have no way to stop the Share Withholding Attack since it is decentralized. And every peer has to run more of a full client if I am not mistake. And there is a lot more overhead I believe. And perhaps also much less resistance against denial-of-service flooding. Frankly I didn't analyze for long enough to be very sure of my initial intuition which is to stay away from it.
I know it is generally impossible to enforce reputation on a 100% decentralized system. So I am intuitively skeptical of P2Pool.
https://en.bitcoin.it/wiki/P2Pool#Payout_logicA miner with the aim to harm others could withhold the block, thereby preventing anybody from getting paid.
Incorrect. The miner can still be paid for the shares he generated which were not a block solution, because another peer might generate the block solution after. Thus the share withholding miner can parasite income off the pool while robbing the pool of his major contribution to the revenue. So this brings revenue to the miner, while parasiting on the pool. If enough shares are from attackers, this destroys the economics of the honest miners in the pool and probably destroys the pool.
So if you owned a centralized pool, it would probably be in your interest to attack all the P2Pools.
Whereas centralized pools could (in a redesigned block chain) implement Meni Rosenfeld's oblivious shares to thwart the share withholding attack.
So as I wrote upthread, P2Pool is not a sustainable or reliable solution.
After making those posts, I remember I eventually studied P2Ppool in more detail and don't remember changing my opinion. I think I posted about it again with additional details, but I can't find the latter post.
Let me add the oblivious shares fix I mentioned in the OP can not be applied to a pool which is decentralized. Many readers may not understand well this very technical issue. Try reading the link I provided in the OP to Meni Rosenfeld's white paper. Perhaps at some point in the future I can take some time to write a layman's explanation of this issue.