4 ms·

I might have missed something, but how can we be sure that malicious nodes do not just steal your money?

by zapt02 13y ago

oleganza 13y ago
Insurance deposits. Both nodes lock up much more money than being "promised". If one node does not deliver, another node will automatically destroy both deposits. Nodes thus are motivated to follow the protocol and not cheat. Since the process is automated, there's no one to play extortion games. The insurance contract is a special type of transaction that I invented: http://blog.oleganza.com/post/58240549599/contracts-without-trust-or-third-parties http://blog.oleganza.com/post/58240549599/contracts-without-...
- Nevermark 13y ago
  I like your transaction idea and have an improvement. The trick is to reduce the costs of one-time punishment. This makes transactions less risky for contract keepers and more risky for contract breakers: Terminology: I would call a contract where both parties lock up funds to be released on successful contract completion a "co-investment contract". Then I would add third parties we can call "co-investment insurers". A co-investment insurer could be any party with a solid reputation of their own to keep, and who has a long term relationship with you (your bank for instance). For a small fee, (1%?) the co-investment insurer agrees to cover your investment loss if a contract fails. Then the economics work like this: 1) People will not be afraid to lock up co-investments when their investment is insured because the worst that can happen if a contract is broken is they are out the small insurance fee. 2) Any party who gets insurance will still be incentivized to complete the contract successfully, so as not to harm their reputation with their long term co-investment insurer. (Who could increase their insurance fee or cancel their account if many transactions failed.) 3) Not everyone needs to buy co-investment insurance for this to have a big impact. The mere possibility of insurance makes punishment a stronger threat and therefore contracts more reliable. 4) As the systems's reputation for reliably punishing contract breakers increases, the insurance costs go down, making it even more affordable. This further increases the system's reliability for punishing contract breakers. Given that all this can be automated, and reinforces incentives to keep contracts, I would think that insurance fees of only 1% or even less would become possible.

I might have missed something, but how can we be sure that malicious nodes do not just steal your money?

by zapt02 13y ago

I might have missed something, but how can we be sure that malicious nodes do not just steal your money?

Insurance deposits. Both nodes lock up much more money than being "promised". If one node does not deliver, another node will automatically destroy both deposits. Nodes thus are motivated to follow the protocol and not cheat. Since the process is automated, there's no one to play extortion games. The insurance contract is a special type of transaction that I invented: http://blog.oleganza.com/post/58240549599/contracts-without-trust-or-third-parties http://blog.oleganza.com/post/58240549599/contracts-without-...

Nevermark 13y ago
I like your transaction idea and have an improvement. The trick is to reduce the costs of one-time punishment. This makes transactions less risky for contract keepers and more risky for contract breakers: Terminology: I would call a contract where both parties lock up funds to be released on successful contract completion a "co-investment contract". Then I would add third parties we can call "co-investment insurers". A co-investment insurer could be any party with a solid reputation of their own to keep, and who has a long term relationship with you (your bank for instance). For a small fee, (1%?) the co-investment insurer agrees to cover your investment loss if a contract fails. Then the economics work like this: 1) People will not be afraid to lock up co-investments when their investment is insured because the worst that can happen if a contract is broken is they are out the small insurance fee. 2) Any party who gets insurance will still be incentivized to complete the contract successfully, so as not to harm their reputation with their long term co-investment insurer. (Who could increase their insurance fee or cancel their account if many transactions failed.) 3) Not everyone needs to buy co-investment insurance for this to have a big impact. The mere possibility of insurance makes punishment a stronger threat and therefore contracts more reliable. 4) As the systems's reputation for reliably punishing contract breakers increases, the insurance costs go down, making it even more affordable. This further increases the system's reliability for punishing contract breakers. Given that all this can be automated, and reinforces incentives to keep contracts, I would think that insurance fees of only 1% or even less would become possible.