Re: Ethereum could afford a 51% attack on Bitcoin, and profit greatly from it
@mjdamgaard what do you think about negative cascade effects for the attackers? I tried to read most of the posts here and I think what you are bringing forward, despite being hypothetical, is worth discussing.
One thing that came to my mind is how likely is it that an ETH whale is not a BTC whale? I know this is mostly speculation and I can't prove it, but I doubt that many of the ETH whales are not owning significant amounts of BTC. But now onto the cascade effect: if a consortium of attackers could take BTC down or cause significant damage to the network value, how can they know that the public wouldn't respond with a major sell-off of ETH as well? The past has shown that BTC worked as a seismograph for the market as a whole. If it turns out that these attacks can lead to major losses in the strongest and most valuable network there is, what is the chance that ETH wouldn't take a major hit and take a dive, going down in insane chaos?
Thank you, and thanks for your post.
You are right that all that would probably be infeasible, especially when you are assuming that the miners have to be anonymous.
But first of all, in order for the Bitcoin devs to respond, they need to find a way to respond. They can't exclude the attacking miners, so it seems that they would have to hard-fork to PoS, or something to that effect. (There is currently a discussion about what steps they could take to mitigate an attack on this thread.)
The also don't have to fear being discovered for legal reasons, it seems, since they are not doing anything legal in the lead-up to the attack, and not in carrying out the attack itself, arguably (see the discussion above about this topic).
Now, they might not want to reveal their intentions to their suppliers, as you point out. That is a good point. But at the end of the day, how can the suppliers really know what their buyers are up to and/or who they sell their ASICs to? Will they really make their customers sign a contract not to participate in a 51% attack, and would that even work?
Also, let me just quickly point out again, that the Ethereum stakeholders can buy/bribe existing mining farms. This proposition has apparently been dismissed so far in this discussion thread, almost as if 'honest' is a predicate that "sticks" to you as a miner, which seems odd to me, especially when the whole concept behind PoW blockchains is that the miners 'behave selfishly.'
I think that an operation of that magnitude wouldn't come to fruition without some very important players in the market noticing and taking the opportunity to stop it. The suppliers aren't only producing ASICs for the sake of mining. But if some of the suppliers (there aren't that many) are approached with an order of that size, I doubt they wouldn't get suspicious. Actually I believe that the suppliers might be well connected with the mining industry. They would ask someone who knows someone who knows someone... If it turns out that there seems to be a group ordering a record breaking, unreal number of ASIC devices, the warning would already be out.
Further, if selfish miners get bribed, what is the chance that all of them would agree to perhaps destroy the entire industry as a whole? Then the big miners are mining pools. What is the legal situation when the allegation of bribery would ever come to the surface and a mining pool operator would be convicted for attacking the network in the worst interest of its users (those who provide the hash power)? It would be obvious if one of those pools attacks the network, but if it happens without the consent of its users, wouldn't there be legal consequences?
I am not sure how it would work, but convincing a pool that operates mining facilities and pools hash power on behalf of its users would probably not agree to getting bribed. But without the pools, an attack that involves bribery wouldn't get the attackers very far.
What if someone from the attacking group blackmails the attackers after they signed contracts with ASIC suppliers? Once the production went on for 12 months, one of the attackers could go rogue and blackmail his own group, threatening to make those plans public.
However I look at it, I don't know how such a huge operation could be pulled off. I know we are discussing theoretical scenarios, but this is really too much theory for me.
One thing that came to my mind is how likely is it that an ETH whale is not a BTC whale? I know this is mostly speculation and I can't prove it, but I doubt that many of the ETH whales are not owning significant amounts of BTC. But now onto the cascade effect: if a consortium of attackers could take BTC down or cause significant damage to the network value, how can they know that the public wouldn't respond with a major sell-off of ETH as well? The past has shown that BTC worked as a seismograph for the market as a whole. If it turns out that these attacks can lead to major losses in the strongest and most valuable network there is, what is the chance that ETH wouldn't take a major hit and take a dive, going down in insane chaos?
...
Thank you, and thanks for your post.
You are right that all that would probably be infeasible, especially when you are assuming that the miners have to be anonymous.
But first of all, in order for the Bitcoin devs to respond, they need to find a way to respond. They can't exclude the attacking miners, so it seems that they would have to hard-fork to PoS, or something to that effect. (There is currently a discussion about what steps they could take to mitigate an attack on this thread.)
The also don't have to fear being discovered for legal reasons, it seems, since they are not doing anything legal in the lead-up to the attack, and not in carrying out the attack itself, arguably (see the discussion above about this topic).
Now, they might not want to reveal their intentions to their suppliers, as you point out. That is a good point. But at the end of the day, how can the suppliers really know what their buyers are up to and/or who they sell their ASICs to? Will they really make their customers sign a contract not to participate in a 51% attack, and would that even work?
Also, let me just quickly point out again, that the Ethereum stakeholders can buy/bribe existing mining farms. This proposition has apparently been dismissed so far in this discussion thread, almost as if 'honest' is a predicate that "sticks" to you as a miner, which seems odd to me, especially when the whole concept behind PoW blockchains is that the miners 'behave selfishly.'
I think that an operation of that magnitude wouldn't come to fruition without some very important players in the market noticing and taking the opportunity to stop it. The suppliers aren't only producing ASICs for the sake of mining. But if some of the suppliers (there aren't that many) are approached with an order of that size, I doubt they wouldn't get suspicious. Actually I believe that the suppliers might be well connected with the mining industry. They would ask someone who knows someone who knows someone... If it turns out that there seems to be a group ordering a record breaking, unreal number of ASIC devices, the warning would already be out.
Further, if selfish miners get bribed, what is the chance that all of them would agree to perhaps destroy the entire industry as a whole? Then the big miners are mining pools. What is the legal situation when the allegation of bribery would ever come to the surface and a mining pool operator would be convicted for attacking the network in the worst interest of its users (those who provide the hash power)? It would be obvious if one of those pools attacks the network, but if it happens without the consent of its users, wouldn't there be legal consequences?
I am not sure how it would work, but convincing a pool that operates mining facilities and pools hash power on behalf of its users would probably not agree to getting bribed. But without the pools, an attack that involves bribery wouldn't get the attackers very far.
What if someone from the attacking group blackmails the attackers after they signed contracts with ASIC suppliers? Once the production went on for 12 months, one of the attackers could go rogue and blackmail his own group, threatening to make those plans public.
However I look at it, I don't know how such a huge operation could be pulled off. I know we are discussing theoretical scenarios, but this is really too much theory for me.