All this "unconsensus" is weird to me given that PoW was created to fix just that. I don't understand how can any other group of people decide what should happen other than the miners. After all, anybody can be a miner. Anything other than that just doesn't make it decentralized anymore.
If you trust the developers, exchanges or even users to make decisions, then why not just make a BitcoinSQL where the servers are controlled by these groups?
Mining specifically allows for this not to happen. one-CPU-one-vote as per Satoshi's paper. No matter the rules of the protocol, the chain with most work is the one that most people agreed upon. This seems to me the only true democratic solution and I don't understand how anything else is possible.
With regards to fees being to high and miners actually liking that, that's bullshit, because miners (which are also users!) care about the health of the entire system. If something like SegWit will bring many more users, that's a win for them.
Let's not forget that anybody can be a miner! Miners aren't just these chinese groups of people. It's the only true democratic way of reaching consensus - anything else is really not a way to reach trustless consensus in my opinion.
Even if you repeat a lie a thousand times, it's not going to become true :) Bitcoin's mining algorithm is so simple that mining industry has specialized too much, and now the best technology for it is ASICs. You cannot mine (or rather, it's nowhere near profitable) if you don't have access to this kind of hardware.
And of course, claiming that anybody can have access to this hardware is very very debatable. The best indicator of how this is not accessible by everybody is how centralized this industry is nowadays (just look at the correlation between ASIC manufacturers and owners of mining pools).
Why is it false? I didn't say "mining is profitable". I just said anybody can do it, and that's the key. Just like anybody can seed torrents without any incentive, or just like Tor nodes exist without any incentive.
IMVHO, ASIC hardware provides such a huge efficiency boost that even if everybody in this planet ran a mining raspberry pi, the ASIC manufacturers would maybe still own more than 51% of the hashrate.
I think they mean it is true in the same sense that "anyone can be president of the US" is true, e.g. true in theory, but in practice, it requires significant wealth which most do not have. I would have worded it differently than calling it a "lie" though (since it technically isn't).
But since "everyone being a miner" doesn't solve anything (even if everybody decides to donate their electricity without getting profits), since the ASIC manufacturers would still control the majority of the network, then any argument based on that analysis is moot already. (Yeah sorry I should not have said "lie" but just "infeasible" and "useless".)
> And of course, claiming that anybody can have access to this hardware is very very debatable. The best indicator of how this is not accessible by everybody is how centralized this industry is nowadays (just look at the correlation between ASIC manufacturers and owners of mining pools).
Even if you repeat a lie a thousand times, it's not going to become true :)
If there was no centralization, people would still be profitable doing solo mining. Most of the pools these days are pools because they own huge mining datacenters already (not because they have a lot of miner-users).
That's not how we should see decentralization, "lacking a central point of control" is a better definition. What you're arguing for is the mining to be distributed.
That's like saying Thomas W. Farley controls 21 trillion dollars worth of companies.
Pools do not control hash power, they merely organize it. That's not to say they don't play a major role in the politics of Bitcoin clients and rules, but to say they 'completely control' Bitcoin is wholly incorrect. These sorts of hyperbolic statements greatly degrade the quality of discourse on centralization.
Thomas W. Farley is so far from the same thing, I can only assume your being intentionally disingenuous. It's much closer to talking about the CEO of hedge fund, and they frankly do have massive amounts of control.
However, in this case the Chinese government can take over these data centers on a whim. You can argue it's a different story when it's spread across a million basements, but highly centralized hashing power is easy to either take over or take offline.
Centralization is not a bool variable: true|false. There are degrees to it. Of course I'm not claiming that bitcoin mining is centralized (100%), but it's definitely not very decentralized.
Pools are not miners. Pools can only make decisions that the miners will tolerate, and historically miners have recognized the dangers even of pool centralization.
That's the four biggest pools (to get a consistent > 50% it might not even be enough) who need to collude. That's not very centralized. Of course the situation could be better, but it's still in a good place imo. It's better than it has been historically for example when a single pool (GHash.IO) approached 50% by itself.
> I don't understand how can any other group of people decide what should happen other than the miners.
You are missing one of the key economic components of Bitcoin. Mining costs a lot of money. Nobody mines for free or at a loss, because 'at a loss' means to the tune of hundreds of millions of dollars.
Miners unquestionably mine the chain that has the highest block reward. Historically, hashrate has always been a function of $hardware_efficiency * $block_reward. If the block reward goes down, hashrate goes down (except where hardware efficiency is increasing fast enough to compensate). This is a lot more visible in the altcoin world, where miners can easily jump from coin to coin, and do as the coins fluctuate in value.
So what determines the price of the coin? Well, supply and demand. If people don't like your fork, there will be low demand, and your fork will have a low coin price. This will result in low hashrate, because miners aren't going to be willing (or even capable) of mining your fork at a loss.
Consensus ends up being fundamentally driven by the economics, and the economics follow the userbase.
I don't think any of that is a contradiction to the main claim that 'miners control the network'.
The very fact that miners are beholden to the economics is the key innovation in Bitcoin. It's the magic of the whitepaper.
> The incentive may help encourage nodes to stay honest. If a greedy attacker is able to
assemble more CPU power than all the honest nodes, he would have to choose between using it
to defraud people by stealing back his payments, or using it to generate new coins. He ought to
find it more profitable to play by the rules, such rules that favour him with more new coins than
everyone else combined, than to undermine the system and the validity of his own wealth.
It is precisely because miners have a vested interest in the economics of Bitcoin that they control the network. That's the magic behind 'Nakamoto consensus'.
> They vote with their CPU power, expressing their acceptance of
valid blocks by working on extending them and rejecting invalid blocks by refusing to work on
them. Any needed rules and incentives can be enforced with this consensus mechanism.
The problem with your understanding is two-fold: first, "one CPU, one vote" sounds very egalitarian, but in practice it just means people buy vote share and therefore control.
The second is the idea that democracy is a perfect consensus system. That's only true if nobody agrees to fork.
No social contract is written in stone (despite millennia of efforts to do so).
PoW does not "fix" consensus. It orders transactions. That is the only thing it can do, but it is an important thing as the well defined ordering of transactions globally is the only thing that can prevent double spending. Miners are paid to do this.
How the software functions is the subject of the old-fashioned open source project. This is also the weak spot. If you want to increase the number of available Bitcoins, your first step is to convince all developers on the project.
Understanding a UASF requires understanding where hashrate comes from. Bitcoin is very clever in that hashrate costs money, and not just a little money, it costs a lot of money.
Which chain has more work is approximately saying 'which chain destroyed the most value in electricity'. The proof of work powering bitcoin costs hundreds of millions of dollars per year.
Nobody burns hundreds of millions per year for free. They do it because you receive Bitcoins as payment, and then you can go sell those Bitcoins to pay your electricity and hardware bills.
That means you have to be able to find someone willing to buy them. A UASF is a bunch of users saying 'we'll never buy your coins unless you follow our fork'. If only a few users say this, it's no big deal, and the miners can decide to listen to them, or they can decide not to bother.
But if 2/3rds of the userbase all agree to enforce the UASF, the miners suddenly are going to have a lot of trouble selling their non-UASF coins. Enough trouble that they probably can't pay their electricity bills. Even worse for the miners, the UASF chain has a high block reward, so they know that if they do mine the UASF chain, they actually will be able to pay their bills. And if a competing miner does it, that competing miner will have much higher margins, higher profits, and will be able to out-invest you when it comes to buying more hashrate.
Scariest of all, if the UASF chain ends up with more work than the non-UASF chain, the non-UASF chain gets completely obliterated, and its full transaction history is reversed, and all the blocks you mined as a miner are destroyed. So even if the UASF doesn't seem to have majority support, it's really bad for you as a miner and as a user if it ever eventually does gain majority support. So when the UASF has a lot of traction, really the safest move is to join the UASF, because at least then your coins aren't at risk of being eliminated entirely.
> So even if the UASF doesn't seem to have majority support, it's really bad for you as a miner and as a user if it ever eventually does gain majority support. So when the UASF has a lot of traction, really the safest move is to join the UASF, because at least then your coins aren't at risk of being eliminated entirely.
Yes, but that is only if UASF gets a significant portion of the miners to switch.
This is also why Bitmain has publicly announced the UAHF, to hard fork away from the original chain, to avoid this possible scenario.
I don't have much time right now, because I am at work, but maybe later I either post more or a better link.
To be very short:
Bitcoin has not been thought to give miners all the power. In the contrary, their main function is timestamping the transactions! And not much more. Full nodes (or hopefully in the not too distant future advanced-spv-nodes) that should be run by many people (at least the conomic important ones, like exchanges, ...) do also verify the correctness and reject miners if they lie or don't behave in the economically-majority intended way! Miners are there so that we know what we (full nodes) should verify (and not blindly believe). Without miners we would not know which blocks to verify (there would be near infinite opinions/blocks). Miners make the choice what to verify just smaller, it doesn't make it automatically the truth.
Here a comment that shows that it was written from day 1 like this in the whitepaper:
(not the best link, but ok for now. And of course economic majority is not a clear thing, but that doesn't mean that miner alone have the power. Beside miner there are the exchanges, the users, the api-providing companies, ... often grouped in 4 categories if you look at videos from Andreas Antonopoulos)
If you trust the developers, exchanges or even users to make decisions, then why not just make a BitcoinSQL where the servers are controlled by these groups?
Mining specifically allows for this not to happen. one-CPU-one-vote as per Satoshi's paper. No matter the rules of the protocol, the chain with most work is the one that most people agreed upon. This seems to me the only true democratic solution and I don't understand how anything else is possible.
With regards to fees being to high and miners actually liking that, that's bullshit, because miners (which are also users!) care about the health of the entire system. If something like SegWit will bring many more users, that's a win for them.
Let's not forget that anybody can be a miner! Miners aren't just these chinese groups of people. It's the only true democratic way of reaching consensus - anything else is really not a way to reach trustless consensus in my opinion.