7 ms·

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 happe

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.

imaginenore 9y ago
> I don't understand how can any other group of people decide what should happen other than the miners. How about Bitcoin owners instead? Your vote is proportional to how much you have.
numbsafari 9y ago
> one-CPU-one-vote "He who has the chips, makes the rules." 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).
knocte 9y ago
> anybody can be a miner That is false, nowadays. > Let's not forget that anybody can be a miner! 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).
- sktrdie 9y ago
  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.
  - knocte 9y ago
    Expanded a bit my answer. 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.
    - flashdance 9y ago
      Not maybe, definitely. A first-gen pi can mine at 200KH/s [1], and a 3rd gen pi is roughly 10x faster in cpu tests [2]. At 2 MH/s per pi, times 7 billion people, you get a paltry 14 petahashes/s. That's a hair over 0.2% of the network... You'd need three trillion raspberry pis to get half the network. Good luck with that! [1] https://en.bitcoin.it/wiki/Non-specialized_hardware_comparison https://en.bitcoin.it/wiki/Non-specialized_hardware_comparis... [2] http://lifehacker.com/how-the-raspberry-pi-3-benchmarks-against-older-models-1762417275 http://lifehacker.com/how-the-raspberry-pi-3-benchmarks-agai...
      - knocte 9y ago
        Thanks so much for that analysis! My mind was not that irrational then when making that guess :)
  - upvotinglurker 9y ago
    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).
joeblau 9y ago
One of the things that was confusing to me is the U in UASF. It's not a user actived soft fork, its certain miners vs other miners.
- Taek 9y ago
  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.
  - lawn 9y ago
    > 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.
imaginenore 9y ago
> I don't understand how can any other group of people decide what should happen other than the miners. How about Bitcoin owners instead? Your vote is proportional to how much you have.
- wyldfire 9y ago
  That's a proof-of-stake! There are coins that use it. The only that comes to mind is peercoin (first to use PoS IIRC).
  - qznc 9y ago
    Ethereum (Vitalik) wants to switch to PoS.
nfin 9y ago
Hi! 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: https://www.reddit.com/r/Bitcoin/comments/6hefd8/comment/diy4ryc https://www.reddit.com/r/Bitcoin/comments/6hefd8/comment/diy... (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)
xorcist 9y ago
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.
Taek 9y ago
> 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.
- Obi_Juan_Kenobi 9y ago
  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.

imaginenore 9y ago

> I don't understand how can any other group of people decide what should happen other than the miners. How about Bitcoin owners instead? Your vote is proportional to how much you have.

numbsafari 9y ago

> one-CPU-one-vote "He who has the chips, makes the rules." 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).

knocte 9y ago

> anybody can be a miner That is false, nowadays. > Let's not forget that anybody can be a miner! 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).

sktrdie 9y ago
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.
- knocte 9y ago
  Expanded a bit my answer. 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.
  - flashdance 9y ago
    Not maybe, definitely. A first-gen pi can mine at 200KH/s [1], and a 3rd gen pi is roughly 10x faster in cpu tests [2]. At 2 MH/s per pi, times 7 billion people, you get a paltry 14 petahashes/s. That's a hair over 0.2% of the network... You'd need three trillion raspberry pis to get half the network. Good luck with that! [1] https://en.bitcoin.it/wiki/Non-specialized_hardware_comparison https://en.bitcoin.it/wiki/Non-specialized_hardware_comparis... [2] http://lifehacker.com/how-the-raspberry-pi-3-benchmarks-against-older-models-1762417275 http://lifehacker.com/how-the-raspberry-pi-3-benchmarks-agai...
    - knocte 9y ago
      Thanks so much for that analysis! My mind was not that irrational then when making that guess :)
- upvotinglurker 9y ago
  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).

joeblau 9y ago

One of the things that was confusing to me is the U in UASF. It's not a user actived soft fork, its certain miners vs other miners.

Taek 9y ago
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.
- lawn 9y ago
  > 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.

imaginenore 9y ago

> I don't understand how can any other group of people decide what should happen other than the miners. How about Bitcoin owners instead? Your vote is proportional to how much you have.

wyldfire 9y ago
That's a proof-of-stake! There are coins that use it. The only that comes to mind is peercoin (first to use PoS IIRC).
- qznc 9y ago
  Ethereum (Vitalik) wants to switch to PoS.

nfin 9y ago

Hi! 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: https://www.reddit.com/r/Bitcoin/comments/6hefd8/comment/diy4ryc https://www.reddit.com/r/Bitcoin/comments/6hefd8/comment/diy... (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)

xorcist 9y ago

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.

Taek 9y ago

> 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.

Obi_Juan_Kenobi 9y ago
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.