4 ms·

> And even if bitcoin itself fails, I think the blockchain will be one of key technical innovations of this time period. A blockchain is only useful if it is h

by VMG 13y ago

> And even if bitcoin itself fails, I think the blockchain will be one of key technical innovations of this time period. A blockchain is only useful if it is heavily secured by hashing power. This is only the case if there is significant mining equipment devoted to validating blocks. That in turn only happens if there is a reward for validating a block. Long story short, it's currently very difficult to use a secure blockchain that is not based on Bitcoin, which puts the quote into perspective. Those who want to use blockchain technology have an incentive to use the Bitcoin blockchain because it has the greatest security.

dragontamer 13y ago
Or not? Proof-of-stake is arguably better than Proof-of-work. Peercoin, NXTCoin and so forth have fundamentally moved forward from the "waste more energy than the next guy" approach that is innate to Proof-of-work.
- nullc 13y ago
  Unfortunately, this doesn't appear to be true. So far all proposed proof of stake systems suffer from problem that miners don't consume a resource to attempt to contribute to the consensis. In other words, the problem of proof-of-statke is that there is nothing at stake. There is no reason for a miner to not attempt to extend every chain that they can— or at least every that they don't hate— and, in fact, doing so is the solution maximizing behavior. In proof of work, by comparison, miners must choose to expend a resource for every attempt and so an effort to try to mine on one fork requires a choice to not spend that resource on other ones. The best strategy then becomes putting all your effort on one chain which you think most likely to survive. Optimal POS strategy is to mine all forks you can, just in case any survive. This issue seems fundamental, but perhaps someone someday will find a clever way around it be changing assumptions. Peercoin was attacked this way— a miner grinding out alternative histories and finding ones where they were awarded almost every block— _immediately_ when POS mining became possible on it. The attack was thwarted because PPC has a central control mechanism where its developer cryptographically signs blocks to force the network to accept them over longer, otherwise better, chains... and it was later closed off by requiring POW blocks to select the POS blocks that can mine. (But now, the security of that coin reduces to the security of POW and the security of the signing key controlled by its pseudonymous developer). See also section 5 of https://download.wpsoftware.net/bitcoin/asic-faq.pdf https://download.wpsoftware.net/bitcoin/asic-faq.pdf
bdcs 13y ago
With merged-mining[1], any alt-coin can hijack Bitcoin's security via PoW without any additional mining, see, e.g., Namecoin (which is unfortunately defunct for unrelated reasons). [1] http://bitcoin.stackexchange.com/questions/273/how-does-merged-mining-work http://bitcoin.stackexchange.com/questions/273/how-does-merg...
- jnbiche 13y ago
  Namecoin isn't defunct: https://github.com/namecoin/namecoin https://github.com/namecoin/namecoin
- vertex-four 13y ago
  Namecoin isn't defunct. In fact, it has several developers. The blockchain has been secured since the incident I assume you are talking about, and there's work on numerous things ongoing - a lite client, an implementation of Namecoin as a library, a proxy allowing any program to connect to Namecoin domains. There is an IRC channel at [0], a mailing list, and a website with a wiki and forums at [1]. [0] irc.freenode.net#namecoin [1] http://namecoin.info/ http://namecoin.info/
- gnaritas 13y ago
  False, just because a miner can merge mine doesn't mean they will, you don't automatically get bitcoin's hash power just because your coin can be merge mined.
sktrdie 13y ago
Actually Ripple's consensus algorithm can be used for a variety of different applications, the ones you would think to build using Bitcoin's blockchain. Because Ripple solves the double-spending problem without the need of mining [1]. The main thing that Bitcoin adds is of course the democratic distribution of coins based on hashing power. But if you don't intend to use "blockchain technology" as a currency, then Ripple may very well serve your needs. 1. http://bitcoin.stackexchange.com/questions/7550/how-does-ripple-solve-the-double-spend-problem http://bitcoin.stackexchange.com/questions/7550/how-does-rip...
- makomk 13y ago
  Ripple's consensus algorithm makes it effectively a centralized system though - there's a central clique of nodes run by the Ripple creators and a handful of people they trust, the consensus they come to cannot be influenced by anyone outside the clique unless they choose to allow it, and if any nodes decide on a different consensus this effectively hard-forks the Ripple transaction history into two branches (their version and the official version).
- gwern 13y ago
  Ripple doesn't solve the double-spending problem, as I understand it: it punts the problem to the user and tells them 'well, if someone abuses the trust you gave them, you obviously screwed up. Sucks to be you.' and lets them figure out who they can trust (if anyone). While a bitcoin is a bitcoin is a bitcoin.
philrapo 13y ago
>A blockchain is only useful if it is heavily secured by hashing power. Ripple consensus keeps a distributed database in sync without requiring mining. And the network's security has no correlation to the amount of hashing power involved. (There's no 51% attack on Ripple) I actually think PoW based systems, because of the 51% attack, present an extremely undesirable amount of systemic risk for broad adoption in the financial system.
- Phlarp 13y ago
  >the network's security has no correlation to the amount of hashing power involved. (There's no 51% attack on Ripple) I think you are misunderstanding what the majority of users consider "security". Typically it involves a situation where someone else can't arbitrarily change the rules on a whim or influence the outcome of transactions on a one-off basis. I'd also like to point out this account is used almost exclusively to promote ripple here on HN while rarely disclosing that the owner, Phil Rapoport, is currently involved with Ripple (the corporation) and has been since 2011 [1] [1] https://ripple.com/blog/interview-with-phil-rapoport-director-of-markets-and-trading-2/ https://ripple.com/blog/interview-with-phil-rapoport-directo...
  - philrapo 13y ago
    You are mistaken. The consensus process does not allow for someone to "arbitrarily change the rules on a whim or influence the outcome of a transaction on a one-off basis". Here is a youtube explanation of consensus: https://www.youtube.com/watch?v=pj1QVb1vlC0 https://www.youtube.com/watch?v=pj1QVb1vlC0 and here is an explanation on the Ripple wiki: https://ripple.com/wiki/Consensus https://ripple.com/wiki/Consensus and a brief explanation on stackexchange: https://bitcoin.stackexchange.com/questions/7550/how-does-ripple-solve-the-double-spend-problem https://bitcoin.stackexchange.com/questions/7550/how-does-ri... You are correct that I am involved in the project. I think Ripple consensus is a more practical, faster, more secure, and less wasteful solution to the double spend problem than proof of work. Happy to entertain your arguments to the contrary.
    - 13y ago
canvia 13y ago
If miners had to be registered and were restricted to a certain amount of hashing power per unique registered user, or had to purchase a standard mining rig (maybe one that doubled as a POS, and restrict mining to merchants) in order to participate, I think that would mitigate the risk of a 51% attack. Right now the imbalance in ability to mine not only creates a risk of an attack, but it skews the distribution of newly generated bitcoin. In my opinion those are the two biggest problems with bitcoin and good enough reasons for it to fail in the long run. It would take a new crypto to address those issues since the initial distribution of bitcoin is so imbalanced.
pbreit 13y ago
I guess that is the question: is there a way to get the (superior) blockchain without the (inferior) currency?

by VMG 13y ago

dragontamer 13y ago

Or not? Proof-of-stake is arguably better than Proof-of-work. Peercoin, NXTCoin and so forth have fundamentally moved forward from the "waste more energy than the next guy" approach that is innate to Proof-of-work.

nullc 13y ago
Unfortunately, this doesn't appear to be true. So far all proposed proof of stake systems suffer from problem that miners don't consume a resource to attempt to contribute to the consensis. In other words, the problem of proof-of-statke is that there is nothing at stake. There is no reason for a miner to not attempt to extend every chain that they can— or at least every that they don't hate— and, in fact, doing so is the solution maximizing behavior. In proof of work, by comparison, miners must choose to expend a resource for every attempt and so an effort to try to mine on one fork requires a choice to not spend that resource on other ones. The best strategy then becomes putting all your effort on one chain which you think most likely to survive. Optimal POS strategy is to mine all forks you can, just in case any survive. This issue seems fundamental, but perhaps someone someday will find a clever way around it be changing assumptions. Peercoin was attacked this way— a miner grinding out alternative histories and finding ones where they were awarded almost every block— _immediately_ when POS mining became possible on it. The attack was thwarted because PPC has a central control mechanism where its developer cryptographically signs blocks to force the network to accept them over longer, otherwise better, chains... and it was later closed off by requiring POW blocks to select the POS blocks that can mine. (But now, the security of that coin reduces to the security of POW and the security of the signing key controlled by its pseudonymous developer). See also section 5 of https://download.wpsoftware.net/bitcoin/asic-faq.pdf https://download.wpsoftware.net/bitcoin/asic-faq.pdf

bdcs 13y ago

With merged-mining[1], any alt-coin can hijack Bitcoin's security via PoW without any additional mining, see, e.g., Namecoin (which is unfortunately defunct for unrelated reasons). [1] http://bitcoin.stackexchange.com/questions/273/how-does-merged-mining-work http://bitcoin.stackexchange.com/questions/273/how-does-merg...

jnbiche 13y ago
Namecoin isn't defunct: https://github.com/namecoin/namecoin https://github.com/namecoin/namecoin
vertex-four 13y ago
Namecoin isn't defunct. In fact, it has several developers. The blockchain has been secured since the incident I assume you are talking about, and there's work on numerous things ongoing - a lite client, an implementation of Namecoin as a library, a proxy allowing any program to connect to Namecoin domains. There is an IRC channel at [0], a mailing list, and a website with a wiki and forums at [1]. [0] irc.freenode.net#namecoin [1] http://namecoin.info/ http://namecoin.info/
gnaritas 13y ago
False, just because a miner can merge mine doesn't mean they will, you don't automatically get bitcoin's hash power just because your coin can be merge mined.

sktrdie 13y ago

Actually Ripple's consensus algorithm can be used for a variety of different applications, the ones you would think to build using Bitcoin's blockchain. Because Ripple solves the double-spending problem without the need of mining [1]. The main thing that Bitcoin adds is of course the democratic distribution of coins based on hashing power. But if you don't intend to use "blockchain technology" as a currency, then Ripple may very well serve your needs. 1. http://bitcoin.stackexchange.com/questions/7550/how-does-ripple-solve-the-double-spend-problem http://bitcoin.stackexchange.com/questions/7550/how-does-rip...

makomk 13y ago
Ripple's consensus algorithm makes it effectively a centralized system though - there's a central clique of nodes run by the Ripple creators and a handful of people they trust, the consensus they come to cannot be influenced by anyone outside the clique unless they choose to allow it, and if any nodes decide on a different consensus this effectively hard-forks the Ripple transaction history into two branches (their version and the official version).
gwern 13y ago
Ripple doesn't solve the double-spending problem, as I understand it: it punts the problem to the user and tells them 'well, if someone abuses the trust you gave them, you obviously screwed up. Sucks to be you.' and lets them figure out who they can trust (if anyone). While a bitcoin is a bitcoin is a bitcoin.

philrapo 13y ago

>A blockchain is only useful if it is heavily secured by hashing power. Ripple consensus keeps a distributed database in sync without requiring mining. And the network's security has no correlation to the amount of hashing power involved. (There's no 51% attack on Ripple) I actually think PoW based systems, because of the 51% attack, present an extremely undesirable amount of systemic risk for broad adoption in the financial system.

Phlarp 13y ago
>the network's security has no correlation to the amount of hashing power involved. (There's no 51% attack on Ripple) I think you are misunderstanding what the majority of users consider "security". Typically it involves a situation where someone else can't arbitrarily change the rules on a whim or influence the outcome of transactions on a one-off basis. I'd also like to point out this account is used almost exclusively to promote ripple here on HN while rarely disclosing that the owner, Phil Rapoport, is currently involved with Ripple (the corporation) and has been since 2011 [1] [1] https://ripple.com/blog/interview-with-phil-rapoport-director-of-markets-and-trading-2/ https://ripple.com/blog/interview-with-phil-rapoport-directo...
- philrapo 13y ago
  You are mistaken. The consensus process does not allow for someone to "arbitrarily change the rules on a whim or influence the outcome of a transaction on a one-off basis". Here is a youtube explanation of consensus: https://www.youtube.com/watch?v=pj1QVb1vlC0 https://www.youtube.com/watch?v=pj1QVb1vlC0 and here is an explanation on the Ripple wiki: https://ripple.com/wiki/Consensus https://ripple.com/wiki/Consensus and a brief explanation on stackexchange: https://bitcoin.stackexchange.com/questions/7550/how-does-ripple-solve-the-double-spend-problem https://bitcoin.stackexchange.com/questions/7550/how-does-ri... You are correct that I am involved in the project. I think Ripple consensus is a more practical, faster, more secure, and less wasteful solution to the double spend problem than proof of work. Happy to entertain your arguments to the contrary.
  - 13y ago

canvia 13y ago

If miners had to be registered and were restricted to a certain amount of hashing power per unique registered user, or had to purchase a standard mining rig (maybe one that doubled as a POS, and restrict mining to merchants) in order to participate, I think that would mitigate the risk of a 51% attack. Right now the imbalance in ability to mine not only creates a risk of an attack, but it skews the distribution of newly generated bitcoin. In my opinion those are the two biggest problems with bitcoin and good enough reasons for it to fail in the long run. It would take a new crypto to address those issues since the initial distribution of bitcoin is so imbalanced.

pbreit 13y ago

I guess that is the question: is there a way to get the (superior) blockchain without the (inferior) currency?