4 ms·

It only seems to solve it in theory. However in a world of centralized mining where devs have high influence on how forks are treated, the double spend problem

by hudon 3y ago

It only seems to solve it in theory. However in a world of centralized mining where devs have high influence on how forks are treated, the double spend problem is still a problem. This is because what distinguishes forks can be their “validity”, not their proof-of-work, and fork validity is decided by humans and subject to politics and fraud.

AnthonyMouse 3y ago
Value is always decided by humans. The question is, is it decided in a distributed way via general consensus, or in a centralized way that exposes you to the whims of unaccountable bureaucrats? Bitcoin devs can do whatever they want, except that there is no law putting them in charge so if they attempt to commit fraud they lose everyone's respect which was the only thing giving them any power, and they get replaced by someone else. Which they don't want, which keeps pressure on them to behave well.
- hudon 3y ago
  Devs can commit fraud without anyone knowing is my point, actually. Picture a scenario like the 2013 Bitcoin fork but with a twist: a dev introduces a bug that allows a transaction to spend bitcoin without corresponding inputs, and triggers it while in the next block spending Bitcoin to buy a coffee, then after a few more blocks for confirmation, publicly announces the bug and the need to move off of the most-work chain. Miners and devs all agree that mining on top of the most-work chain with the bug doesn't make sense and move all the mining to a minority fork that has the bugfix. Minority fork overtakes the buggy fork. Malicious dev has his Bitcoin back and spends them a second time somewhere else. No one's the wiser. We can argue about the game theory and what exactly would happen if this was attempted today, but the point remains that double spend is possible and the Bitcoin white paper does not address this issue because it assumes Bitcoin code is set in stone. As soon as you consider that the protocol evolves and is subject to human manipulation, the white paper's guarantees all go out the window. This includes the 21 million bitcoin "limit".
  - AnthonyMouse 3y ago
    The white paper assumes that you don't have a malicious device. If someone can modify the code running your device, they don't need to double spend, they can just directly steal your cryptocurrency.
    - hudon 3y ago
      Yes, there are other ways of stealing besides double spends. We're in agreement there. And I think we can both further agree that someone that isn't us modifies the code of the Bitcoin node we run on our devices. Unless you're still running Bitcoin 0.1, but then you're actually not on the "Bitcoin" network as we know it today. There are pressures outside of your control that make it so that you have to update your node, and you have to run new code you probably haven't reviewed or approved yourself. The only way to make Bitcoin work as defined by Satoshi is to have the code be 100% perfect and then not touch it, and that ship sailed a long time ago.
      - AnthonyMouse 3y ago
        What is this supposed to be a detriment relative to? If someone compromises your device then they can also transfer all of the money out of your bank account, use it to buy gold bullion and be on a beach in a non-extradition country before you get your next bank statement.
        hudon 3y ago
        Yes, fraud also exists when you use real currency, and your bank will refund you for fraudulent charges. What's your point?
        AnthonyMouse 3y ago
        "Your bank will refund you for fraudulent charges" doesn't prevent fraud, it puts the cost of fraud insurance into the cost of banking. Which may be significantly less efficient when the user or anyone other than the bank is the lowest cost avoider for preventing fraud. They also won't refund you for fraudulent charges unless you report them quickly enough.
- dukeyukey 3y ago
  Considering the _incredible_ amount of scams and rugpulls in the crypto space, some run multiple times by the same people, that would seem incredibly optimistic.

by hudon 3y ago

AnthonyMouse 3y ago

Value is always decided by humans. The question is, is it decided in a distributed way via general consensus, or in a centralized way that exposes you to the whims of unaccountable bureaucrats? Bitcoin devs can do whatever they want, except that there is no law putting them in charge so if they attempt to commit fraud they lose everyone's respect which was the only thing giving them any power, and they get replaced by someone else. Which they don't want, which keeps pressure on them to behave well.

hudon 3y ago
Devs can commit fraud without anyone knowing is my point, actually. Picture a scenario like the 2013 Bitcoin fork but with a twist: a dev introduces a bug that allows a transaction to spend bitcoin without corresponding inputs, and triggers it while in the next block spending Bitcoin to buy a coffee, then after a few more blocks for confirmation, publicly announces the bug and the need to move off of the most-work chain. Miners and devs all agree that mining on top of the most-work chain with the bug doesn't make sense and move all the mining to a minority fork that has the bugfix. Minority fork overtakes the buggy fork. Malicious dev has his Bitcoin back and spends them a second time somewhere else. No one's the wiser. We can argue about the game theory and what exactly would happen if this was attempted today, but the point remains that double spend is possible and the Bitcoin white paper does not address this issue because it assumes Bitcoin code is set in stone. As soon as you consider that the protocol evolves and is subject to human manipulation, the white paper's guarantees all go out the window. This includes the 21 million bitcoin "limit".
- AnthonyMouse 3y ago
  The white paper assumes that you don't have a malicious device. If someone can modify the code running your device, they don't need to double spend, they can just directly steal your cryptocurrency.
  - hudon 3y ago
    Yes, there are other ways of stealing besides double spends. We're in agreement there. And I think we can both further agree that someone that isn't us modifies the code of the Bitcoin node we run on our devices. Unless you're still running Bitcoin 0.1, but then you're actually not on the "Bitcoin" network as we know it today. There are pressures outside of your control that make it so that you have to update your node, and you have to run new code you probably haven't reviewed or approved yourself. The only way to make Bitcoin work as defined by Satoshi is to have the code be 100% perfect and then not touch it, and that ship sailed a long time ago.
    - AnthonyMouse 3y ago
      What is this supposed to be a detriment relative to? If someone compromises your device then they can also transfer all of the money out of your bank account, use it to buy gold bullion and be on a beach in a non-extradition country before you get your next bank statement.
      - hudon 3y ago
        Yes, fraud also exists when you use real currency, and your bank will refund you for fraudulent charges. What's your point?
        AnthonyMouse 3y ago
        "Your bank will refund you for fraudulent charges" doesn't prevent fraud, it puts the cost of fraud insurance into the cost of banking. Which may be significantly less efficient when the user or anyone other than the bank is the lowest cost avoider for preventing fraud. They also won't refund you for fraudulent charges unless you report them quickly enough.
dukeyukey 3y ago
Considering the _incredible_ amount of scams and rugpulls in the crypto space, some run multiple times by the same people, that would seem incredibly optimistic.