CrowdDynamicsLab / QV-paper

Raw tex files and editing process for the paper: "I can show what I really like.'': Eliciting Preferences via Quadratic Voting
MIT License
0 stars 0 forks source link

Add a little more on QV related work detail #17

Closed a2975667 closed 3 years ago

a2975667 commented 3 years ago

“In section 2.2. the authors state: “QV inherits the concept where an individual has to pay the price that other people lose from that outcome [48].” Since I was left wondering about the complexity this adds, I would have appreciated a sentence from the authors explaining why this is desired. Currently, the authors have not justified this and left me wondering about its usability trade-off.” — 2AC

a2975667 commented 3 years ago

This question refers to pp107-108 of the radical markets book.

What's important is not so much the total cost of each number of votes, but that the marginal cost of casting the next vote grows proportionally to the number of votes cast. [...] For this reason, this quadratic rule is the only one that induces rational individuals to buy votes in proportion to how much they know and care about the issue. [...]

For example, X buys 16 votes while Y buys 8 votes. The exact number of votes X and Y buy depends on their estimates of how likely it is that they will be pivotal voters or other reasons that X and Y are motivated to vote. [...]

QV achieves a perfect balance between the free-rider and the tranny of the majority problem. If the cost of voting increases more steeply, [i.e. 4th power], those with strong preferences would vote too little and will revert back to  tranny of the majority. [...] If the cost of voting increased more slowly, [i.e. linear] [...] as a partial free-rider problem would prevail.
a2975667 commented 3 years ago

Forgot to edit the rebuttal doc on this.