censoring the user because i massively respect them as a cryptographer and they design algorithms and theoretically research i could only dream of...
but this isn't exactly a good argument for cryptocurrencies.
if your basis of a future decentralized protocol is based on various different assets that are interchanged, but a major issue for users is that they can't be reversed... and you can only implement this with centralized trust? well, it undermines the core idea.
It doesn't have to be implemented with centralized trust. For example, a merchant could give you a choice of payment arbitrators. Even Bitcoin can do this with multisig and time locks. It's how Lightning works where your Lightning node force closes if the other end misbehaves.
Ah, I get it now.
What I meant is not really escrow but rather having a time lock where merchant can spend it once that's over but there's a way for customer + arbitrator to block that via 2-of-2 multisig where arbitrator then decides if the merchant or customer can spend it.
yeah, no, i get what you mean. there's a lot of different options though, but they all involve either:
- centralized trust
- trust with an arbiter of choice
- predefined rules (some smart contract) probably without a feed of outside data (chainlink probably won't help you here).
Nice thing differentiating it from simply trusting a third party is that you can do multisig approaches like requiring 2-of-3 approval with 3 people trusted to arbitrate it. There are services which implement this for people to implement funds recovery as part of a will, etc.
i mean, this is still a third party, it's just a pretty standard way of doing multisig for the third party: you just have fault tolerance if a certain percent are malicious.
Being able to distribute the trust is extremely useful. It's not necessarily about them being malicious. They could be ordered to seize your funds, etc.
Even for believers in trusting governments/banks with this stuff it's useful to avoid placing all your trust in a single one.