Re:The Quora post that killed Bitcoins. Please discuss if hi...
by jerf on Thursday June 02, 2011 @10:19AM
(#2612051)
"There is absolutely nothing preventing people from taking the open-source code and creating an alternative "Bittoken" currency which changes problematic design elements such as the minting/issuance model."
Broadly speaking, the more desirable properties you try to obtain at once, the more you constrain the design, and it is absolutely possible to constrain the design to the point where there are zero solutions. Arrow's theorem shows an example of that in a voting context. You can not have all of the five desirable properties laid out in the theorem, and it isn't a matter of finding a clever algorithmic solution; they are actually contradictory. You have to loosen the requirements somehow.
There is no guarantee that any given set of desirable properties has a solution. There is no guarantee that BitCoin is even close to having a solution, there's no guarantee that it is merely one requirement over the line, it may well contain multiple contradictions. There's no guarantee that there is anything more optimal than government fiat money.
I'm in favor of experimentation to find out, in general. But it should be pointed out that there is not necessarily going to be a solution at all, and I certainly would say it's not going to be easy to just "take BitCoin and fix the problems" or any variant thereof. There is something stopping people from just "creating an alternative", it's going to be freaking hard, and again, I don't just mean hard in the way that creating an open source operating system from the efforts of thousands of not-centrally managed programmers is, I mean, potentially hard in the way that proving P != NP is, really, actually, hard. Starting from BitCoin isn't necessarily even starting from a useful base.
Broadly speaking, the more desirable properties you try to obtain at once, the more you constrain the design, and it is absolutely possible to constrain the design to the point where there are zero solutions. Arrow's theorem shows an example of that in a voting context. You can not have all of the five desirable properties laid out in the theorem, and it isn't a matter of finding a clever algorithmic solution; they are actually contradictory. You have to loosen the requirements somehow.
There is no guarantee that any given set of desirable properties has a solution. There is no guarantee that BitCoin is even close to having a solution, there's no guarantee that it is merely one requirement over the line, it may well contain multiple contradictions. There's no guarantee that there is anything more optimal than government fiat money.
I'm in favor of experimentation to find out, in general. But it should be pointed out that there is not necessarily going to be a solution at all, and I certainly would say it's not going to be easy to just "take BitCoin and fix the problems" or any variant thereof. There is something stopping people from just "creating an alternative", it's going to be freaking hard, and again, I don't just mean hard in the way that creating an open source operating system from the efforts of thousands of not-centrally managed programmers is, I mean, potentially hard in the way that proving P != NP is, really, actually, hard. Starting from BitCoin isn't necessarily even starting from a useful base.