3 ms·
I think Levine did a good job debunking the idea that a bit of disclaimer in the Terms of Service would preclude a court of law from considering facts beyond ju
by zaroth 10y ago
I think Levine did a good job debunking the idea that a bit of disclaimer in the Terms of Service would preclude a court of law from considering facts beyond just the [exploitable] code of the smart contract. In the real world courts are fully equipped to handle situations like a faulty contract being exploited by insiders and outsiders alike. Fortunately the law is not a Turing machine. I think if the hacker who exploited this bug ever was uncovered it would not go well for them.
But the idea that any miner choosing a fork which blacklists this recursive split exploit is liable for damages to the hacker, that's just laughable. Can you imagine a universe where miners are legally bound to continue investing their hashrate in a particular fork?
The entire system is predicated on miners independently choosing whichever fork they want to support. The attacker is welcome to mine as much as they want on their preferred fork, just as any miner is free to do the same.
In any case, the market will decide how much the Eth on either fork should be worth. You can argue which fork you would value more, or which universe you would rather live in, but what's neat is ultimately the miners and the market will decide.
Whatever direction the Eth ultimately goes, one thing we can be sure of is that blacklisting is always possible to the extent that miners are willing or able to be compelled into implementing them. A campaign against blacklisting this particular address seems to me to be misplaced, because other than enriching the hacker it accomplishes nothing. Miners upgrade their code to express their preferences all the time. Miners can run "official releases" or variations on the same at any time. I think arguing that loading or not loading a particular commit proves anything about the future nature of Etherium is pretty speculative. You can blacklist an address today and not blacklist tomorrow. You can not blacklist today and blacklist tomorrow.
My own opinion is that Eth is very much still in Beta (or Alpha) and the community is still in Kindergarden learning how to write secure contracts in Solidity. We should expect these contracts to fail like this more often than not. If the miners vote for a "redo" on this particular bug in this particular contract I think it doesn't matter much either way. The reality is that Eth has a very long road before it can be safely used for its intended purpose, and that realization (whether it should have been obvious or not) will have a larger economic impact than this particular hack being rolled back or not. It's not like this is the only bug in "The DAO" code.