5 ms·
Stuff like this is why smart contracts won't work for the general public. Nobody wants to live in a world where a bug in a contract lets people steal your money
by twblalock 9y ago
Stuff like this is why smart contracts won't work for the general public. Nobody wants to live in a world where a bug in a contract lets people steal your money and you can't get it back, even when you can prove who did it. We have contract law to deal with this stuff in the real world.
The problem is, if smart contracts are subject to human governance, what advantages do they have over any other kind of contract? How can they be trusted more than anything else that is subject to human governance? Automatic execution of contracts is nice, but if history can be rewritten later, you can't really build a deterministic system on top of them.
- dbasedweeb 9y agoIt’s an expensive, yet ultimately entertaining lesson in why regulations and institutions exist. I suspect that most people knew this was silly going in, but saw there was money to be made. A smaller subset were sufficiently idealistic/ignorant, but quickly learned the lesson. The fun is in the remainder who aren’t just trying to pump money out of rocks, but still really believe in it, the real cryptolibertsrians! It’ll be years before they get over their dreams and crash down to reality, if they ever do.
- AndrewStephens 9y agoJust came in to say exactly this. You can complain about how banks handle transfers all you want, but at least you have a decent shot of getting your money returned if something goes wrong. Sooner or later some court is going to get involved in a case involving smart contracts. I imagine they will ignore what the contract says and go to intent of both parties.
- Paperweight 9y agoAnd by proving here that the blockchain can be forked at will, the whole thing might as well be hosted on an Oracle database at JPMorgan Chase.
- lightedman 9y ago"the whole thing might as well be hosted on an Oracle database" At least the performance difference would be negligibly minimal, given how bad my interactions with various OracleDBs were.
- twblalock 9y agoYeah, and I doubt the judge will be impressed by arguments about immutability, especially if the lead developer of the smart contract system has a history of suspending immutability to help his friends...
- bufferoverflow 9y agoLegal contracts also have mistakes in them, and companies and people get wrecked for millions of dollars. http://www.duhaime.org/LawFun/LawArticle-151/Famous-Attorney-Errors-Lawyer-Bloopers.aspx http://www.duhaime.org/LawFun/LawArticle-151/Famous-Attorney...
- twblalock 9y agoOf course legal contracts can contain mistakes, and even intentional fraud. But when that happens, we don't have to just throw our hands up and say "Well, the contract is immutable, so you're screwed unless you happen to be friends with the guy who runs the blockchain."
- tomcatfish 9y agoThat's actually exactly what happens in a few of the cases mentioned. In fact, in one of them, the courts chose not to fix the error because it would set a dangerous precedent (which is what you are saying would never happen).
- twblalock 9y agoI never said that everyone who goes to court would get what they want, or that the courts would throw out such contracts 100% of the time.
- tigershark 9y agoSome contracts are declared void by the judge because they are against the current legislation. In ethereum the contract is the law (until the people at the top are ok with it, otherwise they will unilaterally change the law)
- marcoperaza 9y agoOne way smart contracts could still prove useful is by changing the position parties are in prior to any lawsuit. Suppose we agree that in two weeks, you are to pay me $1000. Then suppose that some dispute develops between us as to whether the payment is still owed in full, so you refuse to pay. If we merely had a standard contract, I would need to sue you, and the burden of proof would be on me to show that I am entitled to payment of the full sum. If we had a smart contract, the payment would execute automatically, and you would have to sue me, and the burden of proof would be on you to show that you are entitled to recover the sum. So even in a world where a smart contract has the exact same effect on legal duties as a standard contract[1], it could still let parties reallocate counterparty risk. There are of course other mechanisms for reallocating those risks, like escrow services. How smart contracts compare to those seems like a great area for economic and legal research. [1] A doubtful proposition, since parties to a smart contract have manifested some intention to be bound to at least some extent by the mechanics of the smart contract platform. A court will have to consider that as a factor in discerning the true intentions of the parties.
- twblalock 9y agoI'd rather not live in a world where people can take money out of other people's bank accounts without going through due process first. In your example, due process happens at the end of the dispute rather than at the beginning.
- captn3m0 9y agoThis would also mess badly with the economics of law. For a majority of the contracts, the cost will go up as they will need to be scrutinized more because the cost of a mistake is much higher (it is immediate and final).
- shkkmo 9y agoIn that case, don't read too much about how ACH actually works. You are supposed to obtain authorization first, but this is not generally verified at the time of the transaction (some accounts do white list specific accounts for outgoing ACH transfers). People can pull money out of your account with just your account and routing numbers. The only way to initial an ACH chargeback is to attest under penalty of perjury that you did not authorize the transfer. (Not receiving the product / fulfilling the terms of the contract is not a valid chargeback reason for ACH). So if you had indeed authorized the transfer (as maroperaza stipulates) then your only recourse is to sue them in court.
- this_user 9y ago> The problem is, if smart contracts are subject to human governance, what advantages do they have over any other kind of contract? I would argue that they are actually worse, because they are subject to the whims of individuals like Buterin who are effectively accountable to no one. In the real world you have judges, you have juries, you have clearly laid out rules, and there exists a well defined process of how legal issues will be handled.
- seibelj 9y agoI think "smart contract" is a bad name for what it truly is - a program that runs on the blockchain. I tell programmers to think of smart contracts as programs. An Ethereum smart contract (program) may be useful when applied to a problem currently solved with a legal contract. But not necessarily. MakerDAO[0] is a project that achieves a token (Dai) that is termed a "stablecoin" as its value is soft pegged to $1 USD. It was launched in late December and has been very successful thus far. Maker achieves this with a series of smart contracts and operates entirely within the blockchain. MakerDAO is the type of project that is unique to smart contracts and is not possible without Ethereum or similar technology. These are the types of projects you should consider when analyzing the longterm technological capabilities of Ethereum. Imagining a world where paper contracts are replaced with Ethereum smart contracts is a misconception. [0] https://makerdao.com/ https://makerdao.com/
- twblalock 9y agoIt's ironic that a crypto project seeking stability decides to peg its valuation against a fiat currency, i.e. the US dollar.
- profalseidol 9y agoAs always, you'll have to read into the details. MakerDAO doesn't exactly peg it's value to the US dollar. Here's a good interview about how it works - https://www.youtube.com/watch?v=ZggkzPJ5Yt4 https://www.youtube.com/watch?v=ZggkzPJ5Yt4
- ForHackernews 9y agoSo...you can have a program that runs on an extremely slow computer and costs a ton of electricity to execute?
- seibelj 9y ago1) Proof of Stake will end the argument about electricity[0] 2) To criticize Ethereum as "an extremely slow computer" is the laziest, no-effort, no-thought attack you can make [0] https://github.com/ethereum/wiki/wiki/Proof-of-Stake-FAQ https://github.com/ethereum/wiki/wiki/Proof-of-Stake-FAQ
- Dylan16807 9y agoIf you look at it as a way to write a contract, not an assumption about how it will be processed, it's useful. A program can reduce ambiguity by a lot, and reduce the need for legal intervention. A while ago the SEC was looking into having complex securities put a python version of their logic on file, for investors to prod at. That sounds to me like a good use of 'smart contracts'.
- txsh 9y ago> Nobody wants to live in a world where a bug in a contract lets people steal your money and you can't get it back, even when you can prove who did it. If you can prove who stole it, you can go through the courts to gain ownership of their accounts and then transfer the funds back.
- sronever 9y agoDon't forget that Ethereum is a first generation technology. There will be many refinements along the way of any new idea. I think these restores are repulsive but I have to remind myself that their team is always growing and learning from mistakes.
- ckocagil 9y agoThere are a lot of programs that already govern our lives every day. From airplanes to medical devices that can output lethal doses of radiation. We already have established standards for writing and testing safety critical software, like DO-178B. If you want to go even further you use more advanced programming languages that allow you to prove the required properties via its type system. And just in case all of this fails or a case of PEBKAC, you buy insurance.
- Analemma_ 9y agoIs there an insurance company offering to insure against crypto losses? Considering the rampant amounts of both fraud and general incompetence, I assume the premiums are/would be astronomical.
- misrab 9y agosure but it just means there will be a set of standardised smart contracts with a proven track record, and as much as possible other contracts will compose simple ones. It's like saying driverless cars will never exist because the road there will involve some accidents. It sucks, but I don't doubt the future of driverless cars.