/biz/ - Business & Finance

>>20526738
Why not? We are arguing semantics at this point. It's different needs for different purposes. Consider something like fiverr. Yes, you could easily program in simple conflict resolution like "if customer isn't happy then reject payment". However, the problem is that as you explore more complex cases, code is only as good as who programs it.

So let's say I send out something I was commissioned by. I argue it isn't a turd and I worked real hard on it, but the buyer says that it isn't what he asked for. A smart contract system may not capture the fact that the buyer is a turd burglar. So, the platform defers to kleros. Maybe these things could be captured in code, but not all platforms have the need or ability to get down into the bricks for something when the cheaper solution may be to leave it to human input that can smooth out interference. So, it goes to a Kleros court, and it finds in favor of the seller as the buyer was indeed being a turd burglar.

This is the general principle of kleros and why it can have a future together with chainlink. Writing bulletproof contracts costs time and money. It serves as a catchall for anything you don't ask code to do for one reason or another.
>But that's not a smart contract!
Based on what? There isn't exactly something like a turing test for smart contracts unless I'm mistaken (and I love to be mistaken). And even if it isn't a smart contract, it's still a practical way of conflict resolution that I think can have a place in a truly decentralized finance way.

Hell, kleros X Monolith could be good. Let kleros courts decide if a chargeback seems valid.
Sorry for Reddit spacing I'm mobile posting which is much worse