BitShares Forum
Main => General Discussion => Topic started by: pgbit on November 26, 2013, 12:50:10 am
-
Need arbitration on a problem or dispute? Let neural-networks and global voting systems be part of the process. Both parties might agree to a binding decision, and allocation of dues as a result. Why would people in dispute do this? If they believed a system to be inscrutably fair and just, they might consider it over current costly alternatives. Those shareholders weighing-in on a decision might lose or gain depending on the final majority consensus, and they might profit by the outcome decision, but the key here, though (surely?), is to dissociate the algorithmic decision by the DAC and a public vote. Public opinion on multiple decisions, however, might (somehow, independent of market forces) be fairly programmed to affect subsequent decisions, ultimately to evolve into an autonomous independent self-learning DAC.
Team, anyone?
1. Logo designer
2. International lawyer
3. Coin developer
Ideas welcome...
-
I like the idea of an automated arbitration system that essentially presents the thing that needs a decision to a random % of the holders of SolomonShares. The more specialized your need for arbitration, the more expensive it is. The cheapest service would be a summary judgement with 1% of the network polled and it's a random 1%.
A reputation system could be enacted to determine which arbiters are "better". What does better even mean? Maybe it means % voting with the majority decision
-
This kind of DAC is very possible and profitable.
1) People must PAY to submit a dispute to the DAC, some of these fees are paid to the voters and some paid to the share holders.
2) Voting works for binary decisions but how could we use this to settle a divorce?
3) This works well for public disputes, how could it be used in the btc-u hacking dispute?
In a prior life I was working on a centralized debating forum with the aim of finding the truth of anything by having people debate in a new kind of web forum. Recognizing that Republicans and Democrats will fail to apply reason to their votes and instead vote party lines or based upon a biased agenda I knew that I couldn't just have people vote 'true' or 'false' on any given assertion.
So I came up with the idea of having people vote on the relationship between two pieces of information: Does argument A support or contradict argument B. I didn't reveal the result of the votes to the users. The weight I gave a user's vote depended upon how close they were to the mean vote of all other users. In effect, those that were able to correctly guess what others would guess were considered more trustworthy and less biased. Someone with an agenda who voted out of line in an attempt to defy logic would get penalized.
Ultimately the whole system was too complex for people to grasp and I let it die.
That said, JuryShares would be a good way of establishing 'guilt' if you were willing to accept a judgment by random individuals.
-
This kind of DAC is very possible and profitable.
1) People must PAY to submit a dispute to the DAC, some of these fees are paid to the voters and some paid to the share holders.
2) Voting works for binary decisions but how could we use this to settle a divorce?
3) This works well for public disputes, how could it be used in the btc-u hacking dispute?
In a prior life I was working on a centralized debating forum with the aim of finding the truth of anything by having people debate in a new kind of web forum. Recognizing that Republicans and Democrats will fail to apply reason to their votes and instead vote party lines or based upon a biased agenda I knew that I couldn't just have people vote 'true' or 'false' on any given assertion.
So I came up with the idea of having people vote on the relationship between two pieces of information: Does argument A support or contradict argument B. I didn't reveal the result of the votes to the users. The weight I gave a user's vote depended upon how close they were to the mean vote of all other users. In effect, those that were able to correctly guess what others would guess were considered more trustworthy and less biased. Someone with an agenda who voted out of line in an attempt to defy logic would get penalized.
Ultimately the whole system was too complex for people to grasp and I let it die.
That said, JuryShares would be a good way of establishing 'guilt' if you were willing to accept a judgment by random individuals.
Creating a new tier of justice, with not just a local jury, but a global voice of opinion, might be the way to develop this system in the learning phases, so that in the first implementations of such a system users of the process would be encouraged to adjudicate (online jury service, would you do it?) on decisions and processes.
-
In effect, those that were able to correctly guess what others would guess were considered more trustworthy and less biased.
I've been thinking about DACs that involve individual opinion for a certain topic, but since in a blockchain P2P system, every piece of data is publicly available, many people would wait to see what other people 'said' before declare their opinions (which can be automated by a robot for the interest of an individual).
I haven't figured out a solution after talking to certain people.
-
In effect, those that were able to correctly guess what others would guess were considered more trustworthy and less biased.
I've been thinking about DACs that involve individual opinion for a certain topic, but since in a blockchain P2P system, every piece of data is publicly available, many people would wait to see what other people 'said' before declare their opinions (which can be automated by a robot for the interest of an individual).
I haven't figured out a solution after talking to certain people.
Maybe, the blockchain could just broadcast the act of voting? and embed an encrypted vote decision; later the overall decision stats could be decrypted publicly when time-sensitive information is safe to be released.
-
could the legal "rules" ie regs and common law of a jurisdiction be held in a machine readable database/wiki where the algorithm could take into consideration all the factors a conventional judge would take in deciding say, the equitable distribution in a divorce case and then the jury could decide the fact questions like usual?
-
could the legal "rules" ie regs and common law of a jurisdiction be held in a machine readable database/wiki where the algorithm could take into consideration all the factors a conventional judge would take in deciding say, the equitable distribution in a divorce case and then the jury could decide the fact questions like usual?
Any such system would be subject to manipulation of the corner cases and has no way to validate inputs / special cases.
In the case of a divorce, I highly recommend couples avoid 'shared property' and instead always know who owns what. They can of course lend it freely to their partner, but it should always have one and only one owner.
Also... there is a very simple way to divide assets in a divorce that is fair assuming you want a 50/50 allocation. You simply auction everything between the two parties. High bidder gets the asset, low bidder gets the cash. Have the bids done in secret to force a Nash equilibrium on the valuation of assets. Unfortunately most women / divorces are situations where the parties don't want 50/50 but want to hurt / gain advantage and thus neither party is likely to agree to such a division strategy.