[Grustaf]

I mean I'm not a smart contract evangelist, I'm just replying to your point that they are risky, that you need to understand them before using them. You do, that's why most people shouldn't.

As to use cases, you only proposed two things, coffee and real estate. Those examples are both silly, for different reasons. I think in general it very seldom would make sense to use smart contracts for real assets.

The only real use cases I can think of are financial derivatives with crypto assets as underlying, and various decision making mechanisms that govern processes that are controlled by code. As soon as you need to cross into the physical world, it's hard to see how it would work. Then you need a centralised institution that people trust, which mostly defeats the purpose.

[dmitriid]

> Then you need a centralised institution that people trust, which mostly defeats the purpose.

However, even with "purely digital assets" you will end up needing to trust people and exactly for the reason I outlined: unless you're able to read and debug code written in esoteric programming languages, you trust the creator of the contract not to screw you over. And even creators of contracts themselves can't find mistakes in their own code and have all their money drained or locked [1]

[1] Just this week, https://web3isgoinggreat.com/?id=akudreams-earns-34-million-...

"The contract suffered from several flaws, however. The first allowed an exploiter to stop all refunds and withdrawals from the contract... not so lucky with the second issue. A bug in the code failed to account for users minting multiple NFTs in a single transaction... the team can never withdraw the 11,539 ETH ($34 million) earned from the NFT sales—it is stuck there forever"

[Grustaf]

We've been over this, I don't think anyone is claiming that it's impossible to write buggy smart contracts.

But it's not the case that you need to trust the creator, since you can verify it yourself. If you don't understand the code, maybe you can find someone you trust that does. Or just trust that _someone_ hopefully finds any flaws. It's a much better situation than having to blindly trust its correctness.

I wouldn't personally use a smart contract for anything of significant value, but they're not useless just because the transactions are irreversible.

[dmitriid]

> But it's not the case that you need to trust the creator, since you can verify it yourself.

Literally my very first message, and also the one you're replying to: "Ah yes, it's up to the user to review and debug code written in an esoteric programming language to make sure that it works... When even creators of said contracts can't find bugs in their code and fall prey to mistakes "

> Or just trust that _someone_ hopefully finds any flaws. It's a much better situation than having to blindly trust its correctness.

So, at best it's not better than what exists today (trust thrid parties), and at normal (that is not even at worst) is worse than whatever we have today.

I fail to see how this is "not the flaw of technology it's just one way in which it differs from what we are used to"

[Grustaf]

You do remember that my response to your first message was "yes exactly"? Clearly I agree with your assessment. Well minus the snark.

No, at worst it's like today, at best you can verify it yourself.

But as I've said in almost every message, I'm not a proponent, I haven't seen any convincing use cases so far. And I don't even think it's very relevant to compare it to "what we have today", unless you are discussing replacing everything with smart contracts. Yes, some people do that, but I don't, and it's not an argument against the technology that some overzelous true believers think crypto will replace banks.