*Thanks for the comments, I already addressed these in various forms on Twitter, reposting here.
> will regularly need to reach out to trusted third parties to ensure he stays on the canonical chain.
Only if he’s dormant for a full unbonding period, which can be on the order of months. And then even in those cases, there are many ways to try to determine the right chain i.e. asking his node’s peers, asking a friend, heuristics (such as economic activity), or my personal favorite, using checkpoints into a PoW chain like Bitcoin.
This again for me is a difference in mindset. The Ethereum & PoS camp in general believes “social consensus” (Vitalik’s misleading term for this is “weak subjectivity”) is an acceptable fail-safe when the protocol fails. I don’t agree.
The goal of blockchains is to minimize trust so that they can scale socially. Increased reliance on social consensus is a step backward. It makes the protocol less robust, easily corruptible & less socially scalable.
>This is already a deal breaker for the majority of PoS protocols, since they are forced to halt operation if the 2/3+ honest majority threshold is not met.
I would rather halt the chain so people can figure out what’s going on, instead of fork and be forced to revert one fork.
Within the context of traditional BFT systems, I agree that the conservative choice is correct. It is the lesser of the two evils.
However, again PoS offers nothing new in this area, since they are very much a subset of traditional BFT systems. Adding obfuscated punishment schemes on top doesn’t make it less so and brings its own set of problems.
In general, the security offered by a “pause-audit-and-choose-fork” mechanism relies on the continued existence & proper functioning of a group of stake holders who:
(i) knows how to audit forks/restart chain every time there is a halt
(ii) can accurately distinguish good/bad actors &
(iii) can always coordinate & come to consensus.
Either of these 3 conditions can fail. For example, coordination between stake holders might not be as simple as you think when the chain is used globally among many distinct groups who are culturally & geographically apart.
The breakthrough that PoW provides over traditional BFT systems is that it allows the fork resolution process to be automatable, doing away with the need to pause & audit, which is a security hole.
> It’s extremely unlikely that all stake holders will participate in the process of staking & validating. Let’s say the rate of participation is 50%, then the attacker only needs 1/6 of the coin supply in order to cause conflicting blocks/checkpoints, instead of 1/3.
Bitcoin only uses 0.27% of the world’s electricity consumption. 51% attack would only require you to control 0.135% of the world’s electricity, not 50% 😮😮😮
That’s not really a valid comparison, since you can use any arbitrary large value in the denominator to make the energy required appear small 🙂
The real comparison is in economic cost-of-attack. In PoW there’s no getting around the fact that you have to spend electricity/money. While in PoS gaining control over 1/12 of the coin supply might not cost you a penny (buying coins is not the only way to get stake).