Hacker News Reader

Comment by jimmaswell | Hacker News Reader

> Smart contracts are a hilariously bad idea

How exactly?

I didn't write the comment... but I would probably second this. I suppose in general smart contracts "replace" escrow companies. There are escrows for tiny things like buying random stuff in Alibaba. There are escrows for large things - like buying a house/property.

I would say that in general it seems weird to handle either of those cases with a smart contract. For example. Let's say I am on Alibaba as a seller. The escrow unlock is probably based on arrival of an item - which is tied to freight companies or possibly a fedex tracking number. This "seems" like it's totally amenable to a smart contract. But where things break is fraud. Let's say an alibaba seller starts shipping empty boxes so that the funds are released.

Now one could say that a smart contract as well as an escrow company would make the same mistake. But it's still not the same. If that seller repeatedly sends empty boxes - it will quickly catch up to them with an escrow company. At some point the escrow company will stop believing the company and start trusting the pictures of empty boxes arriving. Not so with a smart contract. It would mean un-ending fraud. Now you could back things up by saying there could be a company that the smart contract is tied to, not specifically fedex tracking. But then what is that? It's an escrow company.

For houses - if anyone has been involved in a real estate transaction I would say you are probably familiar with the notion that there are 15 days to attempt getting your loan. Some contracts can be 45 or even 90 days (or years for complex business/government escrows). There are also situations where the seller decides to not cause grief because of some other issues or mistakes. Again a smart contract is not going to understand this shit. It's very peoply.

Again, not the OP on this, but just some perspectives. Maybe someone else has a great use case on people betting on weather outcomes - but what great societal use is that?

Providing a few examples of places where smart contracts do not replace existing solutions does not mean smart contracts are "a hilariously bad idea." Someone more creative than you will come up with a good application for them. Your lack of imagination is not a good enough argument to call something useless...

Maybe so, but “someone will come up with a good application for them” sure makes them sound like a solution in search of a problem.

What is wrong with that? There are lots of new technologies that don't solve some specific problem. They were just created in the name of science or curiosity or even just for fun. Real world , "useful" applications may not come for decades.

Nothing is wrong with that, but if it's true, then that doesn't do much to refute the notion that there aren't any good uses for smart contracts.

If the best argument in favor of them is "well, they're not really good for much right now, but someday somebody might figure out a good use for them," then it's hard to see why people are so excited.

(Though, maybe it's not so hard to see why people are excited if the folks who are excited are all crypto fans who want to build hype to drive the prices up....)

Two quick examples of the use of Bitcoin smart contracts:

1. https://suredbits.com/discreet-log-contracts-part-1-what-is-...

2. https://atomic.finance

I didn't say I'm the authority of what is useless in the entire world though. I added my perspective and agreed - hence the "just some perspective".

> For houses - if anyone has been involved in a real estate transaction I would say you are probably familiar with the notion that there are 15 days to attempt getting your loan.

For land transactions in the US (not houses yet), check out Fabrica (disclaimer: former co-founder). We/they have built something incredibly hard to plug into the existing legal structure in the US, but essentially you can transact on land properties in minutes, as opposed to days.

[0] https://www.fabrica.land/

https://en.wikipedia.org/wiki/The_DAO_(organization) serves as an illustrative example.

> In June 2016, users exploited a vulnerability in The DAO code to enable them to siphon off one-third of The DAO's funds to a subsidiary account. On 20 July 2016 01:20:40 PM +UTC at Block 1920000, the Ethereum community decided to hard-fork the Ethereum blockchain to restore virtually all funds to the original contract.[11] This was controversial, and led to a fork in Ethereum, where the original unforked blockchain was maintained as Ethereum Classic, thus breaking Ethereum into two separate active blockchains, each with its own cryptocurrency.

Is this what you're referring to? The vulnerability has been patched, so I'm not sure what your argument here is. Are you saying we should never use code because code can have flaws?

I'm saying "smart contracts" nearly immediately wound up breaking the "code is law" principle it claimed to be built on. It didn't survive its first big challenge.

I think the more important lesson from the DAO hack was the fork itself, and that by organizing it, Vitalik and friends saved their own wallets (after hilariously asking exchanges to stop trading). If the people holding the reigns can get what they want by applying the right amount of influence, it's not any better than what we already have, and is probably not a great vehicle for realizing the utility of self-executing financial or legal agreements.

So one hack and we write off a whole set of revolutionary ideas and projects? Have you looked into all the other big projects that are providing useful services to people? Maker, compound, aave, uniswap comes to mind.

For me, the most illustrative aspect of the Dao was the hard fork that happened afterwards. For all the talk of smart contracts, the mess was eventually fixed by a small cadre of miners and Ethereum developers who agreed to a hard fork.

But that's just it: It's not a hack. The DAO was functioning exactly as intended.

Decentralized exchanges and smart wallets are useful applications of smart contracts. A smart wallet could include multisig functionality, estate management functionality, etc. For example, a friend has a wallet that will automatically be released to heirs one year after his attorney releases it. Of course he is able to cancel the transfer at any time, if he's alive. Is that a perfect system? I dont know. But it does sound useful.

It's definitely true that many ridiculous and farfetched bad applications of smart contracts have been attempted or proposed. None of these have any bearing on the utility of smart contracts.

I think there are some edge cases where you might want them, but for most situations they seem like they make things worse?

When you do have a contract that a computer can execute, the value of having that contract executed on a machine owned by a 3rd party seems pretty low. At the same time, contracts frequently have undefined behavior and, when they do, courts step in to work out what to do. The option to stop or roll back a contract is usually a feature, not a bug.

Reply on news.ycombinator.com