Hacker News Reader

Comment by GTP | Hacker News Reader

The problem that I see here, is that with the "code is law" approach if there's a bug in the smart contract that is supposed to protect you then you're screwed without recourse.

In a traditional system, if there is a bug in your bank's software that makes you loose money, you can ask the bank to give you the money back. If they refuse, you can take them to court. If instead the smart contract has a bug, there's no one you can appeal to to get your money back. So I think that having humans-in-the-loop is still important, at least for big transactions.

As others pointed out, this problems are much smaller for small transaction so maybe there could be some use there. But I have to think better about this use case.

View on HN

"you can ask the bank to give you the money back" "you can take them to court"

Which like the article argues, is not something that can be solved by any ledger, distributed or not. Even if banks reverses records, they don't change the data which is already on the database/record, instead they just issue a new transaction from the backend or withdrawing it from a legit reserve that they own. In any case, no reversing was done.

What if instead of taking the bank to the court, we have something in place for it that runs based on certain "if statements" or "switch case". Of course, is not as easy as it sounds, we need conditions that are robust enough and not vulnerable to abusers. It's a very complicated problem to solve.

> What if instead of taking the bank to the court, we have something in place for it that runs based on certain "if statements" or "switch case"

As I said in my comment, the problem is that if you have a bug in the statements that should protect you, then there is no recourse for your lost money.

No computer programs are perfect, there will be bugs and exploits.

The question then comes to whether we want to deal with unpredictable human "bug" or consistent computer bug. For the later, we can engineer our way to a minimum risk model, where it become punishing to exploit a bug.

We are doing the same thing with the current system, it's just that different banks and companies are responsible for different systems. There are bugs and exploits in systems of our finacial institution and we rely on human to mitigate the risk.

Will computer be better than human or can we build a hybrid system? That's the question we need to answer.

Reply on news.ycombinator.com