4 ms·

The points in this post are all legit, but I think it misses a couple things. The blockchains are all controlled by some kind of plutocracy (people with wealth

by pfraze 5y ago

The points in this post are all legit, but I think it misses a couple things. The blockchains are all controlled by some kind of plutocracy (people with wealth) as the gist describes. However there is a relatively complex dynamic between the miners/executors, the protocol developers, and the community. A blockchain protocol is a shared contract between them and the power to change that contract requires some kind of consensus between multiple parties. At one point, Bitcoin was using signals from miners to show adoption of a protocol change, which AFAICT amounted to a weighted vote, weighted by mining power. I would argue that makes the miners similar to a legislature, and makes the protocol similar to a legal system. These are the rules for social computing networks, after all. Whether any of this is really more meaningful than the alternatives, I can't say. You could argue that PoW and PoS tie wealth too directly to the "legislature" membership and we ought to use a democratic vote to decide the miners/executors. Or we should just appoint 3 or 5 different legal entities to run the system, and put each of them under some kind of standards body, etc etc etc. It's hard to say. All of these approaches involve complex tradeoffs. My broader point is that the original framing of blockchains (that they're not controlled by anybody) was wrong, but the reality is still relatively interesting.

kevingadd 5y ago
I think the way mining power ended up centralized kind of undermines the legislature comparison, though in some countries like the US the legislature is certainly manipulated enough to probably function in a similar way.
beambot 5y ago
> You could argue that PoW and PoS tie wealth too directly to the "legislature" membership and we ought to use a democratic vote to decide the miners/executors. Hence the advent of the DAO for governance alongside the actual utility tokens.
- Animats 5y ago
  PoW and PoS tie wealth too directly to the "legislature" membership There is that. But nobody seems to be able to make an alternative work. Etherium postponed proof of stake yet again. What was the excuse this time? Amusing potential alternatives: - Proof of Space. Nodes have to be some minimum physical distance apart, like a few kilometers, so you can't have "mines". This might be feasible via some low-orbit satellite network you can use to establish approximate location. - Proof of Hate. Nodes are run by organizations that hate each other, so they won't collude.
godelski 5y ago
I think we should also be clear that the proof systems do distribute control differently. POW requires large investments and knowledge (to create a mining infrastructure). POS is a lot easier for people to get involved. Even though the wealthy can just invest more in POS systems there is still a more level playing field as they do not need the logistics of building mining factories and all that. Any wealthy person can simply invest. This dynamic creates the potential for a more distributed network than the POS system, although it still is biased towards the wealthy class. I'd say that this is still more fair (far from being fully fair) than Chia's system where you need hardware and technical knowhow. You're essentially making the space more competitive. Then you have PiCoin's "Proof of click" where you're just proving you can click a button every 24 hours. Wealthy still get control as they can get people to be the "sub-miners" (earning team/pyramid scheme) that normal people can't achieve. Though this is fairly easy to emulate. This still allows the wealthy to just get wealthier. I think there is still a long way to go to determine the most fair system, but just because there is some consolidation doesn't mean that the systems are equally exploitable by a wealthy class. Ether2 is more decentralized than BitCoin, although there is a fair amount of consolidation. I doubt we will end up with a fully decentralized system (by any proof system) but the question is rather how do we make a proof system that has sufficient competition that it becomes difficult for a single entity to take control? POS is better than POW in this respect and we should note that. But we do need something better than POS. The article does not make these distinctions and does undermine the questions within the cryptocurrency communities. It is a difficult problem to solve but we do need to be clear that the different proof systems aren't equally centralized. There's probably no way to prevent the wealthy from continuing to get wealthier from these systems, but the question is if a blockchain can become sufficiently decentralized that meaningful collusion will be near impossible to make. We are seeing steps in that direction.

by pfraze 5y ago

kevingadd 5y ago

I think the way mining power ended up centralized kind of undermines the legislature comparison, though in some countries like the US the legislature is certainly manipulated enough to probably function in a similar way.

beambot 5y ago

> You could argue that PoW and PoS tie wealth too directly to the "legislature" membership and we ought to use a democratic vote to decide the miners/executors. Hence the advent of the DAO for governance alongside the actual utility tokens.

Animats 5y ago
PoW and PoS tie wealth too directly to the "legislature" membership There is that. But nobody seems to be able to make an alternative work. Etherium postponed proof of stake yet again. What was the excuse this time? Amusing potential alternatives: - Proof of Space. Nodes have to be some minimum physical distance apart, like a few kilometers, so you can't have "mines". This might be feasible via some low-orbit satellite network you can use to establish approximate location. - Proof of Hate. Nodes are run by organizations that hate each other, so they won't collude.

godelski 5y ago

I think we should also be clear that the proof systems do distribute control differently. POW requires large investments and knowledge (to create a mining infrastructure). POS is a lot easier for people to get involved. Even though the wealthy can just invest more in POS systems there is still a more level playing field as they do not need the logistics of building mining factories and all that. Any wealthy person can simply invest. This dynamic creates the potential for a more distributed network than the POS system, although it still is biased towards the wealthy class. I'd say that this is still more fair (far from being fully fair) than Chia's system where you need hardware and technical knowhow. You're essentially making the space more competitive. Then you have PiCoin's "Proof of click" where you're just proving you can click a button every 24 hours. Wealthy still get control as they can get people to be the "sub-miners" (earning team/pyramid scheme) that normal people can't achieve. Though this is fairly easy to emulate. This still allows the wealthy to just get wealthier. I think there is still a long way to go to determine the most fair system, but just because there is some consolidation doesn't mean that the systems are equally exploitable by a wealthy class. Ether2 is more decentralized than BitCoin, although there is a fair amount of consolidation. I doubt we will end up with a fully decentralized system (by any proof system) but the question is rather how do we make a proof system that has sufficient competition that it becomes difficult for a single entity to take control? POS is better than POW in this respect and we should note that. But we do need something better than POS. The article does not make these distinctions and does undermine the questions within the cryptocurrency communities. It is a difficult problem to solve but we do need to be clear that the different proof systems aren't equally centralized. There's probably no way to prevent the wealthy from continuing to get wealthier from these systems, but the question is if a blockchain can become sufficiently decentralized that meaningful collusion will be near impossible to make. We are seeing steps in that direction.