With proof of work, you have to spend your pile of money if you want to influence a vote. Whatever the cost of a 51% "attack" would be, the attacker has to actually spend that money to do it.

With proof of stake, having that money is enough to vote and you don't need to spend or waste it. In fact, you get paid for having that money sitting in the account.

The catch is that people will notice 51% attacks. If you have enough voting money and fudge transactions, all that money becomes worthless pretty quickly. The more realistic issue is that the big voter gets to decide the direction of the protocol, but again sufficiently screwing things up only leads to their own demise.

It was worse with PoW in theory. Just because you have tons of mining equipment doesn't mean you have a stake in the future of the cryptocurrency.

>The catch is that people will notice 51% attacks.

The thing about crypto is that the majority hashrate/stake gets to decide what is an "attack" and what isn't. This behavior isn't a bug, it's a feature and it's what separates crypto from fiat.

With PoW, sooner or later the attacker will run out of money. When they stop paying for the hashrate, their votes all goes away, they don't get extra points for being the top miner at one point in time. With PoS, getting a majority means that your majority will actually increase over time unless someone actively decides to burn the cash necessary to stop you. The longer the community waits/debate over it, the harder it will be to remove the "attacker".

Right, the majority decides what is an "attack" in that it decides which chain to use, and the longest chain wins. But that's not the issue. People (or automated systems) can decide on their own what is an attack. They see transactions being reversed after several confirmations and say "the blockchain is compromised, so I won't use it anymore."

Yes with PoS it's harder to remove the attacker. But the attacker also stands to lose far more money than the attack can possibly be worth.

Recommend you read about slashing staking and zero-knowledge proofs to understand why this is not a problem at least from theoretical point of view

> If you want to give "power to the people", you need some way of estimating independent voting power, that is not tied to resources.

You mean like a democracy?

Hard to implement a democracy without any authority to prove that I'm 1 person with 1 vote and not 100 people with 100 votes.

Except that with blockchain solutions, you generally try to preserve anonymity. (as also elaborated on in other comments in this thread)

You also (almost always) try to preserve anonymity in a democracy. When was the last time you voted in person outside a booth?

> You also (almost always) try to preserve anonymity in a democracy.

You most definitely do not. When you turn up to vote, you get asked for id.

EDIT: your actual vote is anonymous, yes. But your participation in the voting is not anonymous. Blockchain allows for anonymous participation, so to say.

> Blockchain allows for anonymous participation, so to say.

No, different agents are voting. Democracy could be totally anonymous, in principle- get a secret key, generate a valid public one-time key, and your vote is verifiably valid but nobody has any idea who voted, much less who they voted for. If everyone is registered, you don't even have any useful information about who could have voted. Even in reality, your identity is discarded as soon as possible.

Blockchains are inherently public. Validators are NOT anonymously participating- they may be owned anonymously, but the owner isn't the one with the actual right to vote.

It's not just a semantic difference. In a real election, your ballot comes in a voter envelope. The ballot is anonymous inside the envelope, and the envelope is opened blind, but kept until the election is over. If they get two envelopes from the same person, they know it was a fraudulent vote. Once those envelopes are thrown out, there is nothing tying you to an election at all.

On the blockchain you effectively never throw out the voter envelope. Your vote -and what you voted for- can both be tied back to you via ownership of the voting agent. There is zero inherent protection of that relationship. All protection is done externally and the system does not do anything more than making sure it's not actively impossible to conceal your identity.

> Blockchains are inherently public

The vote is public, yes, but the participation is anonymous. I can easily set up two different miners, both controlled by me, but with different addresses. There is not generally an easy way of looking up agent ownership.

And while participation could in theory be anonymous for democracies, as you've described, I'm not aware of any country that actually does that?

It's not very anonymous in the US. You're required by law to identify yourself if possible*. Your party affiliation, registered mailing address, and registered phone number are all publicly visible. Someone knows your votes too, and they try to keep it secret, but I have little faith in that.

* In all 50 states AFAIK you can register with no ID and no home address, but it's illegal to lie.

IDK democracy is the best way to ensure prosperity and wealth. It's basically the idea that the 51% can rob the 49%. Preservation of wealth arguably can be better preserved if taking from the 49% results in their violent frustration of those efforts, and enabling that may require some constitutional republic, benevolent dictatorship/monarchy, or ancap kind of situation.

I'd always rather the greater than 50% coalition have control than the less than 50% coalition. Just look at it from a numbers standpoint. If the vote is to "kill the other team" giving which group power does the least harm? The Majority.

And just because that vote can happen isn't a mark against the system. The three other forms of government you mentioned are all equally capable of that atrocity.

51% killing the 49% in iteration is 1% killing the 99%.

>The three other forms of government you mentioned are all equally capable of that atrocity.

If you ignore the psychological disadvantage of the killers having a monopoly on 'legitimate' violence. For instance, in some anarchistic type scenario there is at least the upper hand that everyone understands the aggression of others is not granted some philisophical legitimacy over their own, and people frustrating these efforts will know their defensive efforts are not philisophically disadvantaged. Moral is an important component of self defense. It is easier to push many people in the cattle car if they think a legitimate 'authority' of the government has spoken.

I like that your anarchy has government, the Ancaps on Reddit would like you.

>your anarchy has government

What are you referring to exactly?

Earlier I mentioned 'constitutional republic, benevolent dictatorship/monarchy, or ancap kind of situation.' Yes some of those include government.

>For instance, in some anarchistic type scenario

I don't see anything 'has government' related in your quote there so now I know you're just full of shit.

> For instance, in some anarchistic type scenario

> ...

> It is easier to push many people in the cattle car if they think a legitimate 'authority' of the government has spoken.

I realize now there's been some massive miscommunication. The statement about government and train cars was not meant to be portrayed "instance" of anarchism. It was supposed to convey a counter scenario that was not so possible under an anarchistic scenario, since obviously government is not anarchism.

I didn't (and don't) think my paragraph conveyed that. But perhaps so it does. Either way it appears I'm unable to effectively communicate to you. I apologize for that. I do not think I am an able communicator in conveying my ideas to you, so I will just stop here and apologize for the trouble.

Not so fast. In both networks, the ultimate control is held by those who run end-user nodes. These people create demand for the asset, and creating demand is ultimately the way to control the asset.

An end-user node is like a device, which verifies that bank notes are genuine or that gold coins are pure. If everyone can independently reject bad money and accept good money, the good money will have more demand and therefore be more valuable.

Even if someone is very rich or has resources to mine blocks, they can't dictate demand for other entities. They must abide by the rules which are enforced by end-user nodes.

If there are entities which have power to dictate demand, and the network can't defend itself under pressure, means that the network is not properly decentralized.

Also, the decision between "good money" and "bad money" should probably be a game theoretic focal point, rather than something set by a foundation and large staking entities.

> These people create demand for the asset

I don't think I really understood what you meant by this... Nor did I really follow your overall point here, if I'm to be honest...

> means that the network is not properly decentralized

ummmm..... this just sounds like a convenient excuse in case of failure.

- "Somebody did a 51% attack!" - "Oh, well it's not the system's fault, it just means that the implementation was bad"

Actually 51% attacks are a theoretical feature of these systems, before you even get to the implementation details.

in a proof of stake system doing a 51% attack is suicidal, it's kind of the whole point

> An end-user node is like a device, which verifies that bank notes are genuine or that gold coins are pure. If everyone can independently reject bad money and accept good money, the good money will have more demand and therefore be more valuable.

The key word there "independently". Proof of resources is one way to tell if an individual node is actually independent.

Proof of resources gives an individual who has more invested in the network more weight in deciding which "bank nodes are genuine". Those individual's votes in theory deserve more weight because proof of resources assumes those individuals are harder to influence (or bribe) and have a stronger incentive to keep the network functioning because they have more at stake. These systems are trying to ensure that the nodes that have the power to accept or reject "money" only accept "good money" and always reject "bad money", and they're using demonstration of resources as a proxy of trustworthiness and independence.

Whether someone ties a bunch of money into specialized ASIC chip that's really only good for mining bitcoin, or they tie a bunch of money into staking eth is an implementation detail.

Miners need not only to spend money on hardware, but also the upkeep of earning money by mining is expensive (electricity, or the cost of electricity generation).

Thus, Proof of Stake let the rich grow their richness for doing nothing. In Proof of Work they need to continue working to make their money grow. Not to mention that they removed the miners from the equation, proving that in ETH system the poor and the people don't have power at all, whereas in BTC no one ever had any such power.

Like with ballots tied to people’s real identity in some kind of “polling station”?

If you're willing to admit a central authority to authenticate who can and cannot vote, cryptocurrency becomes easy.

Really bitcoin is fundamentally trying to solve the problem that its impossible to tell who is a "real" person on the internet. If you dont have that problem everything is trivial.

So what you're saying is, bitcoin was always useless in sufficiently advanced society, like mine, where I routinely verify my identity online through my bank?

Interestingly, the original bitcoin white-paper had a motivating example that basically amounted to side-stepping banks, because banks do not provide fully non-reversible transactions. In essence, bitcoin was a technological solution to a problem which would have been solved more elegantly by changing financial laws, and requiring banks to offer non-reversible transactions. (the problem with that is obviously that it could be used for fraud, but then you get into the whole debate about freedom vs protection)

> In essence, bitcoin was a technological solution to a problem which would have been solved more elegantly by changing financial laws, and requiring banks to offer non-reversible transactions.

Ahh, this may explain why I don't see how cryptocurrency is a good thing. I consider the fact that banks do not allow fully non-reversible transactions as a very strongly desirable feature, not a bug to be done away with.

And I think this is a very valid viewpoint. Clearly, many people think the same, otherwise we'd already have non-reversible transactions in the conventional banking system.

I mean, yes. If you read the bitcoin paper it very clearly talks about what problems it is trying to solve, so this shouldn't come as a surprise.

Whether or not it is useless really depends how much you value independence from a central authority (i was originally going to say anonoyminity, but if you only care about anonoyminity, digicash is a much better solution than bitcoin)

Indeed, for more nuance here's a full explanation of the problem being dealt with by various Proof of Whatever algorithms: https://en.wikipedia.org/wiki/Sybil_attack

They always rely on scarcity of something: scarcity of processing power, scarcity of being on the ledger of the blockchain and owning an amount of currency, scarcity of hard disk space, etc.

I help run the Ultra blockchain.

We have a concept called Proof of Authority where we basically went out and picked competent teams that have a vested interest in ensuring the blockchain is successful. They run the infrastructure in a fully decentralized manner, but we work as a group. There are 7 "pools" and we will eventually expand to 11. It's not decentralized as commonly understood, but way better (imo) than just putting whoever is richest in charge.

Which is what the Worldcoin project tries to do with their retina scans. The idea isn't bad in principle.

On the other hand they want the same amount of their virtual currency for themselves as 2 billion people.

There isn't any cryptographic basis to human retinas, though, to prove a scam is "real"; so, at the end of the day, this is just a centralized actor in the form of a hardware manufacturer that can forge as many retina scans as they want, with the only limit preventing any of us from doing the same being whatever DRM they can try to pile on their device.

On that note, is there anything we could use as a cryptographic basis? I'm guessing DNA would qualify. We could make everyone's addresses be derived from a content-address of our DNA.

You can I believe forge DNA from scratch with enough effort or just pick some up that we perpetually shed everywhere.

DNA is among the easier biometrics to great large quantities of without the victim ever knowing it happened.

Handing over a copy of my biometrics to a VC-funded-and-ran private company is about the last thing I'd want to do in my life, just below taking said eyeballs out.

Not personally into crypto much, but wouldn’t proof-of-identity for those who want to vote be enough? Those who simply want to use the system (to send/receive) need not disclose themselves, but those participating in the ledger ought to have to unmask themselves so that when something goes horribly wrong there’s a person to point at/sue/whathaveyou…

Yeah exact. Proof of trust, web of trust,... datalisp.is