@sdrebel -- on HT you wrote "i believe a fair comparison would be that faking a dkim signature is as easy as solving a btc block. it can happen, but it takes a shitload of computing power to do it"
Nope, that's not right, if by solving a block you mean a miner generating a new block.
It's more like faking a DKIM signature is like finding the private key for an already-generated address, i.e., being able to spend someone else's bitcoins without knowing their private key. Bitcoin is secure because that's so *extremely* computationally difficult. Much harder than generating a candidate block for the blockchain.
this is probably over my head...but isn't that basically the same as solving a block? or not because solving a block is an approximation and for the KDIM we would need the exact key?
Don't quote me on this, but the problem the miners solve is a completely different problem. You can generate a hash easily enough, but to be a candidate block for the block chain you have to find a hash that has a certain property -- it has to begin with a certain number of zeros. To increase the difficulty you just increase the number of zeros required. There are an enormous number of hashes that would be valid, but you just have to find one. So you generate as many as you can, as fast as you can, and hope you're the first to find a valid hash.
To forge a DKIM signature, or to spend someone's bitcoin without knowing their private key, you have to solve a problem that's much harder. There's only one right answer and you have to find that exact answer. As far as anyone (outside the NSA anyway) knows the only way to solve it is by brute force, and if that's true then even if you could harness all the hash power of all the bitcoin miners in the world your bitcoins would still be safe.