3 ms·

Sigh. The only sense in which it "solves" the Two Generals Problem is the sense in which it will, over the course of decades, create a rainbow table for it. Non

Sigh. The only sense in which it "solves" the Two Generals Problem is the sense in which it will, over the course of decades, create a rainbow table for it. None of the work being done is in any way socially useful (a crypto coin based on, say, folding proteins actually produces a social good and so could be equity based). Decentralized proof of a transaction having occurred is in principle valuable, but the token itself is not the store of value. The reason it is accurately described as a pyramid scheme is because holding bitcoin for the purpose of reselling it is exactly like holding a baseball card, or a share in Bernie Madoff's hedge fund: it is based on the premise that someone will come along and pay you more for it than you paid, despite no actual appreciation in any fundamental value.

olalonde 9y ago
> Sigh. The only sense in which it "solves" the Two Generals Problem is the sense in which it will, over the course of decades, create a rainbow table for it. None of the work being done is in any way socially useful (a crypto coin based on, say, folding proteins actually produces a social good and so could be equity based). That is by design. Doing useful work would hurt the network's security. > Decentralized proof of a transaction having occurred is in principle valuable, but the token itself is not the store of value. The reason it is accurately described as a pyramid scheme is because holding bitcoin for the purpose of reselling it is exactly like holding a baseball card, or a share in Bernie Madoff's hedge fund: it is based on the premise that someone will come along and pay you more for it than you paid, despite no actual appreciation in any fundamental value. What you described more closely fits the definition of a "bubble", not a "pyramid scheme". I agree that Bitcoin could be qualified as a "bubble" in the way you describe it. That is, similarly to paper money or gold, its price is completely disconnected from its "intrinsic" value.
- astrodust 9y ago
  > Doing useful work would hurt the network's security. Yeah, no. Any "proof of work" system that meets simple criteria is acceptable: Hard to solve, easy to verify. Hashes are just the easiest to implement, but they're not the only one that could work here. Factoring RSA keys of different lengths would also fit the bill. If the "difficulty" could be arbitrarily adjusted it'd be no different from SHA2 in terms of effort required.
makomk 9y ago
Do you have any idea how rainbow tables actually work and what it would take to "create" one for the SHA-256 hash used by Bitcoin mining? In order to create such a rainbow table, you would have to do about 2^256 SHA-256 computations. All of the Bitcoin mining ever done so far has carried out somewhere in the order of 2^87. To put this in perspective, the most power-efficient miners available right now require 0.1 J/GH of energy. There are roughly 10^24 stars in the visible universe putting out about 10^27 watts each. If you could use the entire energy output of all those stars to mine Bitcoins with the most efficient mining hardware currently available for the entire lifespan of the universe to the present day, you'd still get less than 2^210 hashes, about 1/10^74th of what you'd need to create a rainbow table for SHA-256. I'm pretty sure that even upgrading this hypothetical universe-wide computational machinery to hypothetical mining hardware that's as efficient as the laws of thermodynamics allow wouldn't get you close.
- FabHK 9y ago
  Isn't 2^210/2^256 = 1/2^46 ~= 1/10^14 or so? (1/10^74 is approx 1/2^245)

olalonde 9y ago

> Sigh. The only sense in which it "solves" the Two Generals Problem is the sense in which it will, over the course of decades, create a rainbow table for it. None of the work being done is in any way socially useful (a crypto coin based on, say, folding proteins actually produces a social good and so could be equity based). That is by design. Doing useful work would hurt the network's security. > Decentralized proof of a transaction having occurred is in principle valuable, but the token itself is not the store of value. The reason it is accurately described as a pyramid scheme is because holding bitcoin for the purpose of reselling it is exactly like holding a baseball card, or a share in Bernie Madoff's hedge fund: it is based on the premise that someone will come along and pay you more for it than you paid, despite no actual appreciation in any fundamental value. What you described more closely fits the definition of a "bubble", not a "pyramid scheme". I agree that Bitcoin could be qualified as a "bubble" in the way you describe it. That is, similarly to paper money or gold, its price is completely disconnected from its "intrinsic" value.

astrodust 9y ago
> Doing useful work would hurt the network's security. Yeah, no. Any "proof of work" system that meets simple criteria is acceptable: Hard to solve, easy to verify. Hashes are just the easiest to implement, but they're not the only one that could work here. Factoring RSA keys of different lengths would also fit the bill. If the "difficulty" could be arbitrarily adjusted it'd be no different from SHA2 in terms of effort required.

makomk 9y ago

Do you have any idea how rainbow tables actually work and what it would take to "create" one for the SHA-256 hash used by Bitcoin mining? In order to create such a rainbow table, you would have to do about 2^256 SHA-256 computations. All of the Bitcoin mining ever done so far has carried out somewhere in the order of 2^87. To put this in perspective, the most power-efficient miners available right now require 0.1 J/GH of energy. There are roughly 10^24 stars in the visible universe putting out about 10^27 watts each. If you could use the entire energy output of all those stars to mine Bitcoins with the most efficient mining hardware currently available for the entire lifespan of the universe to the present day, you'd still get less than 2^210 hashes, about 1/10^74th of what you'd need to create a rainbow table for SHA-256. I'm pretty sure that even upgrading this hypothetical universe-wide computational machinery to hypothetical mining hardware that's as efficient as the laws of thermodynamics allow wouldn't get you close.

FabHK 9y ago
Isn't 2^210/2^256 = 1/2^46 ~= 1/10^14 or so? (1/10^74 is approx 1/2^245)