Smart contracts are all the rage in the blockchain world these days. They are the key feature that has pushed Ethereum to prominence. They’re everywhere!
There are just a few little problems. They’re not smart. They’re not contracts. They’re rife with security issues. And they violate the core principles that are supposed to make blockchain wonderful. Other than that, they’re great!
There is a huge amount of rhetoric and propaganda about what Smart Contracts are supposed to be. Here’s the reality: A smart contract is a software program. It’s written in one of a variety of mostly brand-new languages, chief among them Solidity. A smart contract is the exact equivalent in the blockchain world of a “stored procedure” in the database world; this means that it’s embedded in the blockchain and has access to its internal functions.
At first glance, smart contracts can seem like a clever idea that enables endless extensions to the underlying “immutable distributed ledger” technology in which they’re embedded, greatly extending their flexibility and fields of application. Let’s take a look at how this first glance holds up under scrutiny.
Here’s a typical explanation of smart contract:
“Here’s a very reductive way of establishing a smart contract: let’s say you and I have agreed that if I write you a history of bitcoin, you’ll send me $10 on my birthday this year. We can do that via a legally enforceable contract, which involves lawyers, notaries, and so on — or we can do it via Ethereum. In the latter case, you put $10 worth of smart coins in escrow, and when the terms of the contract are met, those coins are released to me. If I don’t meet the terms of our agreement, the coins are released back to you.”
The key, innocent-sounding phrase in the above description is “when the terms of the contract are met.” When I write you a history of bitcoin, will you accept it if it’s a piece of crap? Probably not. Here’s what has to happen with the deal:
- We have to agree on the terms of the deal
- This happens verbally, no matter what mechanism is used.
- We have to express the deal terms in a mutually acceptable way.
- In the land of normalcy, either our verbal agreement would be OK, or we’d have an email or paper exchange.
- In smart-contract-land, someone would have to write a program in an acceptable language such as Solidity. We would both have to have wallets and accounts in the same crypto-currency among the hundreds that are out there. We would have to agree that the Solidity program expressed our mutual agreement. How good are your Solidity reading and writing skills? You would also have to deposit $10 in your account.
- I would write up my history and send it to you
- In real life, I’d get it to you using paper or email.
- In smart-contract-land, there is no good way to send an email – and putting an email address into a smart contract would make it available to the public! There are some clever hacks involving external services that monitor accounts in the blockchain, but there’s no direct solution. So I’d have to get my history to you using plain old real-life methods.
- A decision would be made about whether the history I send satisfied the contract.
- This would be done in real life by you reading the history and making a judgment.
- Smart-contracts would remain blissfully unaware of this crucial step, unless and until an amazing advance in AI/NLP is somehow embodied in them.
- If the terms of the contract are met, the money would be sent to me.
- In real life, you’d hand me the $10, mail it to me, or electronically transfer it to me via one of the widely available methods, for example Venmo, which handled over $12 billion in transactions in 2018.
- In smart-contract land, you’ve had $10 tied up in your crypto account since the contract was agreed to and the program – oops, I meant the “smart contract” – was created by one or both of us. You would then send a transaction to the contract to transfer the money to my account, after which I could convert it to normal money, with fees taken out along the way.
Someone please explain to me how using a smart contract to embody and execute our agreement is an improvement on normalcy?
In addition to the problems mentioned along the way, we’ve got these:
- If you love my history but want to cheat and not pay me, how does the smart contract help?
- If I’m worried about you paying me and want an enforceable contract, how does the smart contract help?
- If we make a written agreement, even a simple email one, at least we can probably both understand it. What are the chances that we are both fluent in Solidity?
- Assuming we somehow manage to write the code in some language, what if we’re less than perfect and have a bug in the code? The teensy weensy little problem with smart contracts is … they’re immutable (i.e., can’t be changed), along with the blockchain in which they’re stored!
- The key part of any agreement of this kind is the characterization of the history I’m supposed to write and the acceptance criteria. How can that be expressed or evaluated in lines of code?
- In terms of just moving the money quick and easy with minimal overhead, how is anything in crypto-land easier or better in any way than Venmo?
Smart contracts are a really cool idea. The best way to use them is in a sandbox where really smart, unemployed people can play games and make experiments, keeping them out of trouble and far away from real life and real problems that need to be solved.
This post originally appeared in Forbes.
Comments