Critical Update on DAO Vulnerability

The bug that was exploited here has been public for a week before someone decided to try it in practice. There was time to dispense back everyone's ether, had they taken it seriously. But taking security seriously requires an almost superhuman distance to your work.

The Ethereum developers is actively debating whether to put in logic to replay the blockchain in order to give back everyone's ether. While that's probably a good idea, it also means the company behind Ethereum can reverse any contract. That puts them in a difficult situation, as any smart contract platform will have dissatisfied parties at all times. (In comparison, none of the Bitcoin thefts have been reversed, and it's not clear they could have been as development is much less tightly knit.)

It's the most exciting thing since the fall of MtGox. The money at stake is comparable (the DAO is about a fourth of what MtGox was in perceived value).

I think we agree on this. The reality is that an active choice will still have to be made by a consensus group, though. It's not a matter of something being forced down the throats of a majority by an oligarchy, it's utilizing the existing consensus mechanism.

I think OP should be well warned here as to what is binding with smart contracts, for sure. This isn't any different than risk calculations in Bitcoin, though -- early days advice was to wait six blocks for transactions over $20, because the cost of subverting the network was very low. This is part of the give and take of mining and distributed consensus with current technologies.

So while there is no clear single party that makes decisions, active developers have the most say.

a) It is being bailed out by referendum, not mandate. Everyone gets a vote, if you don't want to install the patch, don't do it.

b) The funds to bail it out are those stolen by the attacker. Nobody is asking people without DAO tokens to suddenly contribute their ether to restoring the DAO or take debt.

That being said, saying it can't be bulletproof or watertight is too vague.

There is both a first and a second Gödel's incompleteness theorem.

The first shows that a system T which can do arithmetic has some statements that it can express, and which are 'true' , but which cannot be proven to be true by T.

The second shows that a system T which can do arithmetic cannot show that it is self consistent, unless it is not self consistent.

Neither of these seem to be a problem for smart contracts.

It is possible for a system to be self consistent. The smart contract or the system that the smart contract uses does not need to prove itself to be self consistent, so the second theorem is not a problem.

If there is some mathematical statement that can be expressed by the system that the smart contracts use, which the system cannot prove whether it is true or false, this is also not a problem. Which, Ethereum doesn't even have a proof checking thing built into it yet, so I don't see how this would be applicable.

I think that you are probably over-applying Gödel's incompleteness theorems.

Also, I don't think its so much complexity in the "wow these laws are complicated" sense, so much as "strength" in the "how many things can be talked about / shown to be true" sense.

You /might/ be able to do some weird program/proof analogy there, but I really don't think that applying it to law (by a law/program analogy) would really show all that much.

I would think that the law can be understood as being sort of like a function (these inputs result in these outputs), and a function can be both be complicated and total. The law doesn't really do much with formal proofs, as it is now anyway.

Keep in mind that there is also a Gödel's COMPLETENESS theorem.