Vote 📣 Pray 🙏 Love 🤍

Polkadot Blockchain Academy 2022 Cohort Final Exam Project.

Summary

This is a pallet quadratic voting pallet based off substrate collective pallet. The additional feature is the slashing mechanism which provides an incentive for voters to collaborate in the decision making process.

Navigate to pallet/slashing-voting to see tests and implementation

NOTE: This is an experimental pallet, no research has been conducted to actually prove economic costs of this governance system.

Idea

The idea behind is to introduce the quadratic voting, make it anonymous and slash-reward the voters. Let's break it down.

Any account with identity can join a voting council to be a part of governance
When the user joins the council, fixed amount of voting tokens is allocated the account
The user must reserve some currency to have skin in a game
When the proposal is created, the length in blocks is specified
The voters submits votes anonymously. The votes are measured on a quadratic scale
When the voting is over, the reveal phase begins
Voters have limited time to reveal their actual votes
Votes are calculated and the result is deduced
If the voter is in minority (i.e. on the losing side). 10% of their stake is slashed and deposited to the "pot"
If the voter is in majority, they receive even proportion of the reward from the "pot"
If the vote is a tie, both parties get slashed and the money go to proposer
Once the voter has finished all proposal, they can leave the organisation and cash out

Name

The name is quite random. I just like how it sounds :)

One way to interpret it: Vote for choices, Pray for luck and Love the new governance system.

Motivation

The classical voting system has very little incentive for voters. The average vote turn-up does not exceed 10%. These ratio can not provide the true representation of opinion of the population. Therefore, the additional incentive has to be provided. I was interested in running this sort of system as a game theory experiment to see how voters behave under such incentive.

Anonymous voting ensures that voters make choice motivated by personal choice and rationally and not profit-seeking. The slashing ans reward system incentives voters to actually participate in the voting process while quadratic voting system has benefits of allowing voters to express strong preference of their choice. Quadratic voting has a linear dependency and not tied to the economic influence of particular entity.

Different solution for anonymous voting have been considered. Here is the summary of the main two:

Shared public key

Author creates a proposal and generates a keypair for this proposal
The public key of the proposer is distributed to voters
Voters encrypt their vote using the public key and sign transaction with their private key
Once timeout and enough votes are collected, the proposer is ready to reveal the results
The proposer uses their private key to decrypt votes and calculate the outcome of vote

If the proposer reveals results before the timeout -> slashing If the proposer tries to inside-trade the intermediate vote results -> no solution, might be worth using nominating random voters to generate keypairs and use multi-sigs to collectively reveal the results

Commit and Reveal

Author publishes a proposal
The voters commit their decision
Once timeout is out, the voters have some time to reveal their choices
If the voters does not reveal their results -> slashing

While the first approach may seem more convenient for the voter since they only need to submit a single transaction to represent a vote, it harm the global integrity of a governance protocol. The voter can simply inside-trade the actual votes and give away votes of other voters before the end of voting phase. The commit and reveal approach ensures trustlessness of a solution, hence, a suitable solution.

While it may seem a burden for user and proposer, the economic incentive actually motivates any voter to end the commit phase and reveal phase ASAP to potentially collect reward from voting.

Objectives

A summary of what has been implemented and what's planned:

Stretch goals

More tests
Callable intrinsics for proposals
Advanced treasury management

Compromises

The current system only supports simple text based proposals. Callable intrinsics will be added later
Weights of dispatchable functions have not been benchmarked and do not represent true economic value
If the voter reveals the choice after the reveal phase deadline, the slashed funds go into pot and stay there (burned). A better approach might be considered how to distribute those funds among truthful voters later.

Running

Simply run script build-run.sh script to build a chain in release mode and run a dev node. Otherwise, /scripts contains additional scripts to run your chain in docker.

Refer to substrate setup instructions to start hacking

Account commit signatures for voting

Alice - Yes - salt: 10 - f26b35c9565f76a01b06ac1dd80832027a59306de8ac31b1019dff75890ec76c3376fb11985323f75788750a9c9ec0e17fc26440a49914d00ff2b7ef2d1a588f
Bob - No - salt: 10 - b67cec01fabdc17233dc1080f09c4ffb86d0f19077f6c4f601f951a7f5a851175d127bb64142a24074596e2496388a9f01b00cd3a9db21a458e76757deda8585

Gist for generating signatures

Resources

License

Apache License, Version 2.0 (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)

SkymanOne / vote-pray-love