5 ms·

THat's not how private blockchains work at all. IF you have a private blockchain (usually called an "app chain") people don't "buy hashing power" on it. The en

THat's not how private blockchains work at all. IF you have a private blockchain (usually called an "app chain") people don't "buy hashing power" on it. The entity running the network has 100% of the validators on the network. "Hashing power" doesn't apply because it's not bicoin so would be POS, but the equivalent is owning the validators, which if it's an app chain there isn't (ever afaik) a process by which some other org can buy the ability to run a validator. The controlling entity just decides that.

Retric 3y ago
I was doing the mathematics thing of skipping over the most trivial case. Set N = 0 dollars of hash power and it becomes really obvious that the person running the network can just arbitrarily create any blockchain they want at any time for ~zero cost. It therefore clearly isn’t an “immutable system of record” because they can change it. But let’s flip it around what specific benefit are you proposing and how exactly does cryptography enable it differently than a Bank.
- junofan 3y ago
  You can checkpoint the canonical fork on a public PoW chain like Bitcoin every so often. Banks carry insurance.
  - Retric 3y ago
    What’s your argument for >>centralized<< blockchains here? Bitcoin and other public PoW blockchains are decentralized. > Banks carry insurance. And???
- seanhunter 3y ago
  Firstly I don't disagree that private chains are generally pointless, so I'm not going to post these supposed specific benefits. I was saying your "hash power" based thought experiment is not valid. Also the "most trivial case" you're proposing here is a strawman. You demonstrate that your trivial case wouldn't work and everyone who does this knows that it wouldn't work so that's not how it works. Genarally these chains submit a transaction proof to some base chain so they can't go back later and rewrite history unless they could also rewrite history on that base chain. Here's a source if you want to read up about how these things actually work https://wiki.polygon.technology/docs/cdk/ https://wiki.polygon.technology/docs/cdk/
  - Retric 3y ago
    I am not disagreeing that in the wider context it comes off as a strawman. Which is why I was asking for specific benefits. However, I was specifically responded to someone who seems to think blockchain = immutable system of record and I was pointing out that private blockchains don’t provide that because the mechanism public PoW chains uses to get immutability isn’t applicable. I realize there are a lot more than just 1 way people are proposing private blockchains and most of them involve publishing something, but as long as we can agree the blockchain bit is pointless then discussing them is a waste of time.
    - nightski 3y ago
      It's immutable in the sense that you only need to check the node you are interested in to verify it's history hasn't been tampered with. That can be valuable. I'm well aware that someone could create an alternative chain, but it will be different and easily noticeable.
      - Retric 3y ago
        You can’t actually validate a node without knowing the prior state and thus if each account contains enough tokens to preform the relevant transactions. So, it would be noticeable if every prior node and all its contents were published and people are validating the integrity of those prior nodes. Which is true of literally any public ledger. The specifics of what get published can make things more convenient as there’s more locations you can send hashes of logs than the full logs, but publishing hashes alone doesn’t do anything without those logs also being public.

Retric 3y ago

I was doing the mathematics thing of skipping over the most trivial case. Set N = 0 dollars of hash power and it becomes really obvious that the person running the network can just arbitrarily create any blockchain they want at any time for ~zero cost. It therefore clearly isn’t an “immutable system of record” because they can change it. But let’s flip it around what specific benefit are you proposing and how exactly does cryptography enable it differently than a Bank.

junofan 3y ago
You can checkpoint the canonical fork on a public PoW chain like Bitcoin every so often. Banks carry insurance.
- Retric 3y ago
  What’s your argument for >>centralized<< blockchains here? Bitcoin and other public PoW blockchains are decentralized. > Banks carry insurance. And???
seanhunter 3y ago
Firstly I don't disagree that private chains are generally pointless, so I'm not going to post these supposed specific benefits. I was saying your "hash power" based thought experiment is not valid. Also the "most trivial case" you're proposing here is a strawman. You demonstrate that your trivial case wouldn't work and everyone who does this knows that it wouldn't work so that's not how it works. Genarally these chains submit a transaction proof to some base chain so they can't go back later and rewrite history unless they could also rewrite history on that base chain. Here's a source if you want to read up about how these things actually work https://wiki.polygon.technology/docs/cdk/ https://wiki.polygon.technology/docs/cdk/
- Retric 3y ago
  I am not disagreeing that in the wider context it comes off as a strawman. Which is why I was asking for specific benefits. However, I was specifically responded to someone who seems to think blockchain = immutable system of record and I was pointing out that private blockchains don’t provide that because the mechanism public PoW chains uses to get immutability isn’t applicable. I realize there are a lot more than just 1 way people are proposing private blockchains and most of them involve publishing something, but as long as we can agree the blockchain bit is pointless then discussing them is a waste of time.
  - nightski 3y ago
    It's immutable in the sense that you only need to check the node you are interested in to verify it's history hasn't been tampered with. That can be valuable. I'm well aware that someone could create an alternative chain, but it will be different and easily noticeable.
    - Retric 3y ago
      You can’t actually validate a node without knowing the prior state and thus if each account contains enough tokens to preform the relevant transactions. So, it would be noticeable if every prior node and all its contents were published and people are validating the integrity of those prior nodes. Which is true of literally any public ledger. The specifics of what get published can make things more convenient as there’s more locations you can send hashes of logs than the full logs, but publishing hashes alone doesn’t do anything without those logs also being public.