4 ms·

Banks don't require pseudonymity. If you remove pseudonymity from the list of requirements you no longer need mining, you can just run any consensus protocol to

• scott_karana 11y ago
  That's a great point. Whoever these R3 people are, I'm sure they'll rake in some good cash irrespective of whatever benefits are provided... ;)
• natrius 11y ago
  Sure, but if you're reaching consensus on a cryptographic chain of transactions, that's exactly what a blockchain is. Mining is not an integral feature.
  • danbruc 11y ago
    Mining is the core feature that allows Bitcoin to work without knowing all the parties. Reaching consensus and cryptographically protecting information is not a very hard problem if you know the parties.
    • natrius 11y ago
      These banks clearly disagree with you. It's not cryptographic protection of data. It's the cryptographic audit trail that's useful. No one can break the rules of the system except for validators, and everyone instantly knows when a validator breaks the rules so they can stop trusting them.
      • danbruc 11y ago
        I meant integrity protection, i.e. ensuring that no one can unilaterally modify the information afterwards. Hash chains and trees are at least 35 years old.
        • eru 11y ago
          Ie if you know all players, you could store your transactions in git instead of the blockchain.
          • joelthelion 11y ago
            There's a difference between knowing and trusting...
            • eru 11y ago
              You'd still sign your transactions. You only need to trust the players not to DoS or double spend---and these things are easy to audit for.
• icebraining 11y ago
  Yes, but Bitcoin is a full system already designed for this exact purpose. The only real issue is the waste of proof-of-work, but that's proportional to the hashing power given to it; if the chain is private, a single server per bank would suffice.
  • repomies691 11y ago
    What is the point of doing any mining at all, if it is unnecessary? Adds just extra complexity to the system. These bankers are idiots.
  • IkmoIkmo 11y ago
    None of what you said makes any sense. No bitcoin isn't designed for private ledgers, it's the exact opposite, an innovation to make global, public, decentralised/distributed (you can argue the semantics a bit here), trustless system without a dependency on any particular set of gate keepers, while keeping alive some semblance of privacy/pseudonymity. This private banking database requires none of that and is the opposite of what bitcoin was designed for. And the gigantic waste of proof-of-work is completely unnecessary, and forcing the use of it regardless by keeping a 'single server per bank' is insane, it's literally a waste of money. Not only that but it doesn't even achieve anything you want as it creates a false sense of security, after all what stops a bank from simply having more than 1 server and blowing all the proof of work from anyone else out of the water and dictating the entire database? Well guess what, the exact same thing that obviates the need for a proof of work system in the first place. Bitcoin is the dumbest possible technology outside of the context for which it was designed, in which it appears to be the only viable technology and works incredibly well. Running a private blockchain with a set bunch of gate keepers is simply ridiculous.
    • icebraining 11y ago
      I agree. My point is: Bitcoin is a well-tested shared ledger, with plenty of eyeballs trying to exploit and secure it. Does it make technical sense to run Bitcoin for a private ledger? No, it does not, you're absolutely right. But does it make business sense to waste a few CPU cycles mining, compared to wasting probably hundreds of thousands of dollars developing custom software for this purpose, which will still be less tested and probably more insecure? I'd say it does.
• CydeWeys 11y ago
  The key advantage that the blockchain provides is distributed consensus-finding. It's the best known solution to the Byzantine generals problem to date. Pseudonymity is besides the point.
  • zaphar 11y ago
    Paxos and Raft are arguably the best known solutions to the Byzantine Generals problem not bitcoin. They provide stronger guarantees Bitcoin which allows forks. What Bitcoin provides that those algorithms do not specify is a cryptographically verified audit trail. However as some have mentioned here if you don't need the pseudonymity as these banks do not then you could add cryptographic auditing to one of those protocols. Bitcoin's proof of work is not a requirement in that case.
    • DavidSJ 11y ago
      I think it would be more accurate to say that what Bitcoin provides that Paxos and Raft do not is trust-free consensus. The members of the Paxos parliament all trust that the others are properly following the protocol, but they still have to deal with all of the pathologies of an unreliable network. Bitcoin has to deal with unreliable networks and dishonest nodes who spend money multiple times. It handles that, at the cost of huge computational overhead.
      • zaphar 11y ago
        Actually since Paxos is meant to solve the byzantine generals problem they don't necessarily trust that the others are properly following the protocol. But the algorithm guarantees that as long as no more than N others are dishonest you can still reach consensus. Bitcoin does the same where N's value differs but with far more wasteful computation and less strong guarantees.
        • DavidSJ 11y ago
          Perhaps you're thinking of Byzantine Paxos (not regular Paxos)? I'll have to read up on that to determine its relevance to the double-spend problem. One concern I'd have is how parliament membership is determined (e.g. what about Sybil attacks?).
• kang 11y ago
  Not exactly. You still need to solve the origin of money problem. Refer problems with proof-of-stake coins.