3 ms·

What does P=NP have to do with anything? Bitcoin can be attacked in polynomial time i.e. the "51% attack." I am also curious to know what security promises yo

What does P=NP have to do with anything? Bitcoin can be attacked in polynomial time i.e. the "51% attack." I am also curious to know what security promises you would be referring to.

oleganza 13y ago
51% attack is a myth. Do you seriously think that all BTC holders would just blindly jump to a suddenly revealed chain to allow some dud to revert 1000 BTC transaction? There would be some turbulence, no doubt, but people will quickly ban the source of disturbance or hire an army to go and shut it down. Most probably, 51% will never happen because it's never economically justified. Even craziest anti-heroes must have earned a lot of money to risk pulling this off for very dubious reward. Some scientists don't like possibility of it, but it's their own bet on how reality works. When I mentioned P=NP, I meant that Bitcoin rests on inability to quickly break hashes and ECDSA signatures. If we find such a method, it would probably be a proof of P=NP.
- betterunix 13y ago
  The 51% attack allows more than just reversing some transactions. It allows an attacker to selectively block transactions. It allows an attacker to deny the mining reward to miners. Really though, the point has more to do with formalizing Bitcoin's security. You claim that it relies on the security properties of cryptographic hashes and of digital signatures, but what property of Bitcoin actually reduces to the security of hashes or signatures? What are Bitcoin's security guarantees? When you can at least state some security property Bitcoin achieves, then maybe you can prove that Bitcoin's security depends on hashes and signatures. Without such proof you cannot go about making statements about P vs NP being relevant to Bitcoin's security; indeed, it is hard to speak about "security" for Bitcoin at all.
- peterwiese 13y ago
  It has been shown in a recent study that from a game theoretic point of view, it is in the interest of a rational player to form a cartel with other rational players in order to enable the 51% attack. Now, major stakeholders in the real world such as banks are getting in on bitcoins. Additionally, they may be interested in destroying the currency once and for all. Personally I think it is just a matter of time until the attack will be realized. Maybe it already happened. We wouldn't event know....
- dragonwriter 13y ago
  > There would be some turbulence, no doubt, but people will quickly ban the source of disturbance or hire an army to go and shut it down. Good luck with that if the "source of disturbance" is a major nation state hostile to Bitcoin.

What does P=NP have to do with anything? Bitcoin can be attacked in polynomial time i.e. the "51% attack." I am also curious to know what security promises you would be referring to.

oleganza 13y ago

51% attack is a myth. Do you seriously think that all BTC holders would just blindly jump to a suddenly revealed chain to allow some dud to revert 1000 BTC transaction? There would be some turbulence, no doubt, but people will quickly ban the source of disturbance or hire an army to go and shut it down. Most probably, 51% will never happen because it's never economically justified. Even craziest anti-heroes must have earned a lot of money to risk pulling this off for very dubious reward. Some scientists don't like possibility of it, but it's their own bet on how reality works. When I mentioned P=NP, I meant that Bitcoin rests on inability to quickly break hashes and ECDSA signatures. If we find such a method, it would probably be a proof of P=NP.

betterunix 13y ago
The 51% attack allows more than just reversing some transactions. It allows an attacker to selectively block transactions. It allows an attacker to deny the mining reward to miners. Really though, the point has more to do with formalizing Bitcoin's security. You claim that it relies on the security properties of cryptographic hashes and of digital signatures, but what property of Bitcoin actually reduces to the security of hashes or signatures? What are Bitcoin's security guarantees? When you can at least state some security property Bitcoin achieves, then maybe you can prove that Bitcoin's security depends on hashes and signatures. Without such proof you cannot go about making statements about P vs NP being relevant to Bitcoin's security; indeed, it is hard to speak about "security" for Bitcoin at all.
peterwiese 13y ago
It has been shown in a recent study that from a game theoretic point of view, it is in the interest of a rational player to form a cartel with other rational players in order to enable the 51% attack. Now, major stakeholders in the real world such as banks are getting in on bitcoins. Additionally, they may be interested in destroying the currency once and for all. Personally I think it is just a matter of time until the attack will be realized. Maybe it already happened. We wouldn't event know....
dragonwriter 13y ago
> There would be some turbulence, no doubt, but people will quickly ban the source of disturbance or hire an army to go and shut it down. Good luck with that if the "source of disturbance" is a major nation state hostile to Bitcoin.