RE: Analysis of 51% Attack by Nation State - Vulnerability of PoW vs PoS
I just saw this but on the PoW angle you have missed a major gap, a nation state could simply make or use fictional accounts to buy their own ASICs/GPUs. If they had already set up their own mining pool (under another fake identity) to gain a proportion of global hash rate then they only need enough ASICs/GPUs to bridge the gap to 51%. This is cheaper and more secretive than trying to buy up a PoS stake, which cannot be done without the knowledge of the market and become exponentially harder the more is accumulated.
Another consideration is a government can attack multiple coins with ASICs/GPUs but cannot do the same with PoS as once the accack is know all the funds will be lost.
PoW is even weaker if you consider just 3 entities run a lot of PoW pools, simply closing these 3 down even temporarily, by hack or conventional means would significantly reduce hash rate on a number of coins allowing the government owned hash rate to get to 51% before the miners re-organise to new pools.
Thanks @scaletrix for the considered response. In my respectful view your suggestions misunderstand where the real power in the Bitcoin network lies.
The idea that someone can execute a 51% attack on Bitcoin using other people's hashing power by controlling a pool ignores the fact that miners will leave that pool long before it gets anywhere near 50%. This has been borne out in practice.
The same is true re your idea of hacking the pools. The hashrate will just move to other pools.
Miners, not pools, control the Bitcoin network. Pools doing anything dodgy or getting near 51% quickly have an exodus of miners to other pools. Bitcoin miners keep very close watch on what is going on in the network and have a massive incentive to stop activity that is adverse to Bitcoin - they have a huge sunk investment in Bitcoing ASIC miners.
While one can execute rented hash power 51% attacks on minor PoW networks (eg Verge) this is only because they are small and use the same hashing algorithm as other much larger networks. Miners don't care if some minor coin is wrecked by a 51% attack but they DO care if a major coin (that is their main livelihood) is attacked. Also the hash rate required to execute the attack is so small miners mightn't notice.
Even trying to buy enough GPUs or ASICs to "bridge the gap", say 10% of hashing power would be a huge, time consuming and very detectable undertaking.
While you are correct that the funds used in a 51% attack on a PoS network would likely be lost, you assume this matters to a government that can just create more funds out of thin air.
My idea for a government would be to build a secret mining farm, that perhaps could net them 10% hash rate, but to leave that offline. They then make a pool that gives them 15% of other users hash rate. Then they simultanoeusly turn on their farm and attack the current largest 3 pools; remember just preventing the pools accessing the blockchain for a few hours would be enough to cripple the network and cause a sell off. In terms of miners moving hashing power, if the top 3 pools were intermittently able to access the block-chain, where would they move to? How would they know which pools were reliable at any moment?
In any case even if we dont agree, at least it gave me an opportunity to give you an upvote for a carefuly considered post :)