On: Blockchain at Berkeley

[xg15]

If you talk about the general idea of an append-only distributed database with no central point of control, I agree with you.

However the proof-of-work/auto-adjusting-difficulty method of asserting truth still seems horrible to me from a technical perspective.

Politically, I don't want to live in a world where this principle catches on and an ever-growing % of energy production would be spend on solving meaningless math problems.

[simias → xg15]

Yeah the proof of work is probably the less elegant aspect of it all, I can agree with that. But theoretically you could replace by any other function that's difficult to compute but easy to verify. Maybe it's just a solution in search for a problem.

[ufo → simias]

I don't understand how that would fix proof of work. The whole point is that the voting power is proportional to how much computational power you are willing to burn away, in order to protect against Sybil attacks.

[simias → ufo]

I was thinking that eventually we may discover a proof of work that would actually compute something useful instead of burning electricity away. One can hope at least.

[arthurcolle → simias]

Ethereum

[xg15 → arthurcolle]

I'm not savvy in the technology, but are the contracts actually executed as part of mining?

[akhosla → xg15]

Recommend you check out Proof-of-Stake. It solves the problem of power wastage. ;)

[rocqua → akhosla]

That requires solving the 'nothing at stake' problem. I recall reading some proposed solutions, but they were complicated.

If anything is great about proof-of-work, it's how simple it is.

[Klathmon → akhosla]

But many people (including those who are actively developing it) aren't even sure if Proof-of-Stake is even fully possible.

As far as I know there are still some pretty big holes in PoS that still need to be filled in.