2 ms·

I like this idea. I am skeptical that it will work. Necessary in this idea is an inflation curve to continue to provide the basic income and discourage hoardin

I like this idea. I am skeptical that it will work. Necessary in this idea is an inflation curve to continue to provide the basic income and discourage hoarding. It is also necessary to scale the user base to make the idea meaningful. This could lead to hyperinflation as the network bootstraps, creating a headwind that scales at least as fast as the idea. The requirement of actual meatspace peer-to-peer trust is an extremely hard problem as well. It seems like one option is people take this seriously and trust networks end up resembling current social networks with the inherent stratification. And you don't get internet money. Alternatively this requirement is eroded or ignored because it's too hard, until it becomes obvious that there are entire networks of fake accounts, everyone is holding FakeCoin, the risk can't be arbitraged out because of the 1 to 1 requirement, and now it's 2008 and FakeCoinCo is Too Big To Fail. Would love to hear why I'm wrong.

Jtsummers 9y ago
Regarding meat-space web of trust, I think solutions like Keybase can help with this (though I'd want to see the principle of it decentralized, rather than only one service as it is now). With validators, you also largely solve this problem. The vast majority of people are not hermits with no physical connections. I can create an account and go through my apartment building's validator. The apartment office has a vested interest in making sure my account is legit (I'm paying them!), and now I'm connected to a real world entity. Similarly, my employers can create a validator and connect my account during the normal on-boarding of a new hire. They also have a vested interest: a fake account impacts their reputation; they pay into that account and may even be paid from it (suppose I overspent on a business trip and need to reimburse them, they need me to pay back in a currency they trust). It does still run into problems, don't get me wrong, but this validator concept could mitigate many of the scaling issues of web of trust.
- lucozade 9y ago
  The validators idea to accelerate bootstrapping is a good one. And I think you're right; it be institutions like companies and banks that will provide the validation services. The issue is tying this to a non-fungible asset. The most obvious problem will be ghettoisation. You can envisage "invitation-only" validators at one end of the economic spectrum and a state sponsored validator at the other end. You could, possibly, force them to interact by mandating a common intermediary, presumably the state, so that there are connections across the population. But, at that point, the personal asset/currency idea seems superfluous. Might as well be an account linked to, say, your social security number that mints a fungible asset periodically. You can look at enhancements to the SSN idea to make it less forgeable but that's a separate issue. Having said all that, I've been quite skeptical of the social benefit of UBI. Largely as I couldn't see how you could sensibly start from here, as it were, and control its development. I still can't see how to do that but I'm less skeptical that it's possible.
- wallacoloo 9y ago
  I'm skeptical that there's enough to prevent validators from acting maliciously here. Firstly, the only leverage against a validator is that if they act poorly, others will no longer accept their currency, so their personal UBI is possibly lost. Secondly, the validator doesn't necessarily _gain_ anything from continuing to act in a trustworthy manner. Maybe they require membership fees or something, but that's not intrinsic to the system. Taken together, all it requires is for a validator to insert _one_ fake account into the trusted pool to break even, and _two_ fake accounts to profit significantly. For example, they could create 50 fake accounts at the start of the experiment and let them accumulate fake wealth on the side, then operate as a validator for two years, then introduce these 50 fake accounts into the trusted pool and exchange 100-man-years of UBI for actual value. This appears to be more than they would receive over a lifetime of acting in a trustworthy manner, it's trivial to pull off, and if anything it's a _conservative_ attack - if a validator can get a userbase of 10000 people, 50 fake accounts doesn't sound easy for another person to detect.

Jtsummers 9y ago

Regarding meat-space web of trust, I think solutions like Keybase can help with this (though I'd want to see the principle of it decentralized, rather than only one service as it is now). With validators, you also largely solve this problem. The vast majority of people are not hermits with no physical connections. I can create an account and go through my apartment building's validator. The apartment office has a vested interest in making sure my account is legit (I'm paying them!), and now I'm connected to a real world entity. Similarly, my employers can create a validator and connect my account during the normal on-boarding of a new hire. They also have a vested interest: a fake account impacts their reputation; they pay into that account and may even be paid from it (suppose I overspent on a business trip and need to reimburse them, they need me to pay back in a currency they trust). It does still run into problems, don't get me wrong, but this validator concept could mitigate many of the scaling issues of web of trust.

lucozade 9y ago
The validators idea to accelerate bootstrapping is a good one. And I think you're right; it be institutions like companies and banks that will provide the validation services. The issue is tying this to a non-fungible asset. The most obvious problem will be ghettoisation. You can envisage "invitation-only" validators at one end of the economic spectrum and a state sponsored validator at the other end. You could, possibly, force them to interact by mandating a common intermediary, presumably the state, so that there are connections across the population. But, at that point, the personal asset/currency idea seems superfluous. Might as well be an account linked to, say, your social security number that mints a fungible asset periodically. You can look at enhancements to the SSN idea to make it less forgeable but that's a separate issue. Having said all that, I've been quite skeptical of the social benefit of UBI. Largely as I couldn't see how you could sensibly start from here, as it were, and control its development. I still can't see how to do that but I'm less skeptical that it's possible.
wallacoloo 9y ago
I'm skeptical that there's enough to prevent validators from acting maliciously here. Firstly, the only leverage against a validator is that if they act poorly, others will no longer accept their currency, so their personal UBI is possibly lost. Secondly, the validator doesn't necessarily _gain_ anything from continuing to act in a trustworthy manner. Maybe they require membership fees or something, but that's not intrinsic to the system. Taken together, all it requires is for a validator to insert _one_ fake account into the trusted pool to break even, and _two_ fake accounts to profit significantly. For example, they could create 50 fake accounts at the start of the experiment and let them accumulate fake wealth on the side, then operate as a validator for two years, then introduce these 50 fake accounts into the trusted pool and exchange 100-man-years of UBI for actual value. This appears to be more than they would receive over a lifetime of acting in a trustworthy manner, it's trivial to pull off, and if anything it's a _conservative_ attack - if a validator can get a userbase of 10000 people, 50 fake accounts doesn't sound easy for another person to detect.