5 ms·
> PoS offers much, much stronger security than PoW Uh, why do you say that? The security of PoS in fact ought to be much LOWER than the security of PoW: The
by eMGm4D0zgUAVXc7 6y ago
> PoS offers much, much stronger security than PoW
Uh, why do you say that?
The security of PoS in fact ought to be much LOWER than the security of PoW:
The goal of requiring proof of work is that you cannot just send multiple versions of the same transaction into different areas of the network to double-spend your money - because you need to commit work for producing a block, and due to consuming energy you can't fake that.
Well, you can compute two (or more) blocks in parallel, but then you'll spend half of your available CPU (or ASIC nowadays) cycles on each block, thus cutting your speed in half. So the non-malicious competitors on the network will produce more blocks meanwhile because they're not splitting their computation power, and thus your fake blocks will get invalidated because they're on the shorter chain.
With PoS on the other hand you can create as many fake blocks as you want and spam them to the network. The only security is the hope that the random network topology arbitrarily results in the double-spending blocks arriving at the targets under attack after the other blocks arrive.
But if you run thousands of nodes on the cloud and thus have better network connectivity than the victims you can make your double-spend blocks arrive first at the victims.
So:
- PoW: Relies on physical limits, you need to have physical hardware and physical energy to conduct an attack.
- PoS: Relies on the network connectivity of the attacker being hopefully worse than the connectivity of the non-malicious network. Who can guarantee that? Nobody.
- delaaxe 6y agoThanks for answering questions I’ve had for a long time about POS. Do you have any links to share on the subject?
- eMGm4D0zgUAVXc7 6y agoGoogle "nothing at stake problem".
- woah 6y agoThis is completely untrue and has been since 2014. The solution is that validators are "slashed" and lose their tokens if they have been found to sign two blocks at the same height.
- ishishkin 6y agoWhat has happened in 2014? Is that a time when brain-washing campaign in crypto-media has been started? "Slashing" is not a solution to this problem.
- woah 6y agoVitalik Buterin came up with a solution for the nothing-at-stake problem: https://blog.ethereum.org/2014/01/15/slasher-a-punitive-proof-of-stake-algorithm/ https://blog.ethereum.org/2014/01/15/slasher-a-punitive-proo... Also, I'm not sure why you and the other commenter are so argumentative about this. There are several PoS networks out there such as Cosmos, Tezos, etc, holding more than a billion dollars. If there was an issue, someone would have hacked them by now.
- ishishkin 6y ago> Vitalik Buterin came up with a solution for the nothing-at-stake problem Is that a link to an article from a peer reviewed academic journal? > There are several PoS networks out there such as Cosmos, Tezos, etc, holding more than a billion dollars. If there was an issue, someone would have hacked them by now. I'll give you a hint: "slashing" is not a thing that sustains security of these networks.
- literallycancer 6y agoWhere did he claim it was a peer reviewed journal?
- ishishkin 6y agoDon't you think this claim should be supported by some reputable secondary source?
- nootropicat 6y agoSecurity deposits of attackers in ethereum are slashed, up to 100%, in case of an attack. The value at stake can potentially grow into hundreds of billions, much more than any PoW chain can hope to accumulate in mining revenues in its entire existence. Last but not least, there's no way to delete PoW attacker's gpus, but hostile stake is always going to be slashed. Asic pow chain can be forked - once - to a gpu pow, but that's it, and after that there's no recourse to sustained attacks. This property virtually guarantees that no attack against PoS with slashing is ever going to happen. https://vitalik.ca/general/2020/11/06/pos2020.html https://vitalik.ca/general/2020/11/06/pos2020.html
- eMGm4D0zgUAVXc7 6y ago> Security deposits of attackers in ethereum are slashed, up to 100%, in case of an attack. What defends against the attacker configuring his nodes to just not relay the blocks which slash his deposits, by having a majority in the network connectivity, and thereby convincing victim nodes that he in fact is the victim of false slashing because the victims will only discover the slash-claims much after the attackers "valid" blocks? Or in other words: Isn't the slashing mechanism also reliant upon mere hope that the network topology randomly happens to be in favor of non-malicious peers?
- woah 6y agoYou seem to assume that the attacker has control over the entire internet.
- eMGm4D0zgUAVXc7 6y agoEthereum has 11 000 nodes currently: https://www.ethernodes.org/ https://www.ethernodes.org/ There are probably many botnets of IoT devices with 10x that many nodes, aren't there? So a single botnet could probably ensure that a target victim has the majority of peer connections to the attacker.
- woah 6y ago
- viraptor 6y agoWhy did you not add "Who can guarantee that? Nobody." to the PoW description as well?