4 ms·

> Parties can prove things without others needing to trust anyone. Yes, but the only things you can prove without trust are events that happen 100% inside the

> Parties can prove things without others needing to trust anyone. Yes, but the only things you can prove without trust are events that happen 100% inside the blockchain. Which, AFAICT the only use for a system that works that way that I've seen is cryptocurrency. The other one people were really hot on was supply chain, but, from what I understand, when you have supply chain trust issues, they don't look like "oh gee someone edited a database record", they look more like https://www.bloomberg.com/news/articles/2021-03-09/trader-buys-36-million-of-copper-and-gets-painted-rocks-instead https://www.bloomberg.com/news/articles/2021-03-09/trader-bu... Which, ironically, that's not a great use case for cryptocurrency either. Traditional escrow services are a better fit. If the trader had bought that tanker full of rocks with BTC, there would have been no way to reverse the transaction after it had been determined to be fraudulent. So, yeah, blockchain is mathematically provable to have certain properties. But, like with any proof, those properties only hold for as long as the proof's assumptions hold. And it's really, really difficult for us laypeople to recognize when the assumptions do or do not hold when they're defined in terms of technical jargon that uses a familiar term like "trust" but assigns it a subtly but critically different definition from the everyday meaning.

tnzm 5y ago
>but, from what I understand, when you have supply chain trust issues, they don't look like "oh gee someone edited a database record", they look more like https://www.bloomberg.com/news/articles/2021-03-09/trader-bu https://www.bloomberg.com/news/articles/2021-03-09/trader-bu... >Normally, in such cases of non-delivery a trading house could make a claim against a cargo’s insurance policy. But Mercuria found that just one out of seven contracts used by the Turkish company to insure the cargo was real. The rest had been forged. Can't forge a smart contract.
- mumblemumble 5y ago
  Fair, though that would be a much more compelling observation if anyone had ever succeeded in implementing a full-fledged insurance policy in the form of a smart contract. I'm curious, for example, how you'd get an errors and omissions clause working in Etherium. Like, how would Ether arbitrate a claim against that policy? Assume, for the sake of argument, that hard AI has not been invented yet.
  - tnzm 5y ago
    You're correct that hard AI would be necessary for the blockchain to perform autonomous arbitration in cases where human discretion is necessary. In practice, this is solved by giving the contract the capability to query an external system (referred to as an oracle), which provides the results of the independent arbitration as an input to the contract.
tinus_hn 5y ago
You’re describing a number of limitations, some of them more valid than others. I don’t think blockchain is worth the hype it got a few years ago, but still, with these limitations, there are use cases. That doesn’t change because you say ‘this use case doesn’t work’.

tnzm 5y ago

>but, from what I understand, when you have supply chain trust issues, they don't look like "oh gee someone edited a database record", they look more like https://www.bloomberg.com/news/articles/2021-03-09/trader-bu https://www.bloomberg.com/news/articles/2021-03-09/trader-bu... >Normally, in such cases of non-delivery a trading house could make a claim against a cargo’s insurance policy. But Mercuria found that just one out of seven contracts used by the Turkish company to insure the cargo was real. The rest had been forged. Can't forge a smart contract.

mumblemumble 5y ago
Fair, though that would be a much more compelling observation if anyone had ever succeeded in implementing a full-fledged insurance policy in the form of a smart contract. I'm curious, for example, how you'd get an errors and omissions clause working in Etherium. Like, how would Ether arbitrate a claim against that policy? Assume, for the sake of argument, that hard AI has not been invented yet.
- tnzm 5y ago
  You're correct that hard AI would be necessary for the blockchain to perform autonomous arbitration in cases where human discretion is necessary. In practice, this is solved by giving the contract the capability to query an external system (referred to as an oracle), which provides the results of the independent arbitration as an input to the contract.

tinus_hn 5y ago

You’re describing a number of limitations, some of them more valid than others. I don’t think blockchain is worth the hype it got a few years ago, but still, with these limitations, there are use cases. That doesn’t change because you say ‘this use case doesn’t work’.