• 0 Posts
  • 805 Comments
Joined 2 years ago
cake
Cake day: October 6th, 2023

help-circle













  • If each vote is a block in the chain them it is definitely recountable.

    That’s not what it means. With this system, you can’t independently verify what happened. You can only also look at the blockchain and see that some hash has registered some vote.

    But you don’t know if that is actually true. You can’t see if pushing the red button makes the red vote come out. You can manually count if you want, but the original billet doesn’t exist, only a processed form of it on the blockchain.

    If I get a reciept of my vote’s hash in the chain, I can confirm it’s being included.

    Only THAT it’s included. Otherwise you have something linking you to your vote, which is bad


  • That’s not the first-owner problem. I’ll try to explain in more detail. The problem arises when you’re using the blockchain as a “reciept”. You can only ever trace the ownership of the reciept, not the item it represents, without a trusted party.

    • Say we made a blockchain that determines the ownership of all fluffy hats in the world. It starts at june 1st 2026. Lets just assume there’s a trivial way to perfectly describe fluffy hats that we can put in a token. Or hell, pretend it’s super complex, that changes nothing.

    • You bought a fluffy hat in 2002, and made one for yourself in 2008. You own both, wearing one to bed when you go to sleep on may 31st, 2026.

    • At 1 second past midnight, june 1st 2026, I make two tokens, one for each of your hats.

    • I am now officially the first owner of those hats. You are suddenly a thief holding my property, even though it never left your head.

    That’s the first owner problem. Without a trusted source, there is no way to ensure the first owner in a blockchain is actually the owner under the current legal definition (as in, you made the hat from homespun wool, it’s on your head right now). It gets even worse though, because I can even make tokens for nonexistent fluffy hats that haven’t been made. As soon as someone makes it, i’m already the owner.

    The ONLY application for a blockchain with a trustless system is if the entire property is directly on the blockchain, and that doesn’t work.


  • I have a PhD, I know how research works. But the new for a new data distribution system isn’t caused by a technological advancement, it’s driven by social systems.

    The usecases for blockchain technology are wellknown, and they are basicaly the antithesis of the technology itself. You need a massive breach of trusted institutions, while maintaining free, open and fast electronic communication throughout the network. You also need to solve the first-owner problem without introducing a trusted party, which is a huge issue in almost every practical application.

    So yeah, it can be an unknown unknown, but it’s one in the same way that research into moonratpoison has potential applications. Maybe something vastly unlikely happens, and then it’ll be a good thing we have the tech.




  • OpenCerts does require institutions to be trusted partners, so this isn’t permissionless. Why not just make the normal database searchable if it relies on trusting someone?

    It’s not that blockchain doesn’t have any applications. It’s just that it only solves problems in ways that are easier and better solved by other implementations. It’s not that are no practical applications, it’s that there aren’t even any realistic theoretical applications.