The Conviction Voting cadCad model is officially released. It can be used as a useful education and decision support tool for those who want to learn how to use cadCAD to model and simulate systems.
Original link:
Word Count: 2163 Author:
Translator: La Translation Agency: DAOSquare
This is the digital twin currently deployed at 1Hive.org. The following content is mainly extracted from the readme file of the model notebook. The initial model is a collaboration between Aragon, 1Hive, BlockScience and Common Stack. If you are willing to support the further functional development of the CV model, you can authorize contributions through this Gitcoin.
Open source FTW!
Question: Why rethink governance?
The traditional time-box voting process has limitations in efficiency in distributed communities, and early experiments in the blockchain space show that there is enough room for improvement. Direct democracy achieved through token voting requires a high cost of attention, which means that everyone needs to know everything that is happening at the same time, so that they can vote-for or against. This will lead to a lack of participation, which means that only 0.2% of token holders participate when voting for important decisions of some projects, which will lead to a solid foundation for whether these tools can lay a solid foundation for the future decision-making process. Voice of doubt.
However, combining the opportunities of blockchain technology with the new voting model can lower the threshold for voting and allow organizations to access rich data streams with continuous community preference signals. This makes decision-making more sensitive and flexible, opening up opportunities for a new era of on-chain signaling and governance.
Faith Voting 101
Conviction Voting is a new type of decision making process used to predict real-time collective preferences in a distributed job proposal system. Voters continue to express their preferences and invest their tokens on proposals they wish to approve. With the passage of voting time, their belief in their vote (ie, weight) also grows. The collective belief accumulates until it reaches the gate valve of the algorithmic setting of the fund ratio required for a proposal. When the accumulation of belief exceeds the threshold, the proposal will be passed, the fund will be released, and the project can begin.
Belief voting improves the traditional discrete voting process by allowing participants to vote at any time, and does not require a collective consensus on each proposal.
In order to reduce the attention consumption of token-based voting, Conviciton Voting is a tool that transforms the decision of the majority into full support for the proposal.
Voting is the signal processing that converts private continuous preferences into public decentralized events. Belief voting, as a token-based signal processing tool, provides some improvements compared to traditional binary voting during the conversion process.
When processing continuous personal preferences (for example: addressing global climate change) into effective shared results (for example: coordinating carbon emission reduction policies), you have reason to expect a continuous voting process, which is more suitable than our traditional decentralized voting system This signal conversion. Faith voting can more smoothly transform our evolving social beliefs (for example, on civil rights issues) into political actions.
1 Block scientific analysis of the token rights of the Hive.org belief voting proposal system
As our governance toolkit continues to expand with new tools such as belief voting, we can consider designing governance systems in the context of their communities. In the 1Hive community, holding Honey tokens can give you certain rights in the 1Hive organization. In order to design a distributed decision-making system, we need to consider the rights granted by governance tokens, how to control these rights, the attack vectors they present, and how to mitigate these vectors, now and in the future.
Belief voting takes time as the core part of collective decision-making and provides us with real-time signals of collective purpose. We hope this will promote more democratic dialogue and align our political system with the existing social system.
Voting of belief provides us with a new and transparent perspective, allowing us to understand the collective purpose of our community. It provides us with a richer signal about the sudden and dynamic preferences of a group so that we can better understand and discuss important issues in the community. It eliminates some attack vectors of temporary voting, such as last-minute voting fluctuations, and it allows people to vote at any time to promote flexibility in participation.
Belief voting mimics the natural distributed decision-making process, such as neurons burning in the brain. Will this be one step closer to real-time collective intelligence?
Different flavors of belief voting
The design space of this new governance tool is open to further exploration. From the academic origin of Dr. Michael Zargham’s research on robots and multi-agent coordination systems, belief voting is called social sensor fusion, which is a “fusion” of continuous personal opinions into collective emotional signals. This shows that belief voting can have many “flavors”:
1. Faith Fund (for decentralized proposals)
In this version of CV, the proposal has a pass threshold. When the accumulated collective beliefs pass the threshold required for support, the proposal will be passed and funds will be allocated to start work. An example of this kind of CV can be found in the 1Hive or Commons Stack model of the distributed work proposal system in which DAO allocates funds.
Faith Fund, the tokens that advocate supporting the proposal accumulate beliefs until the threshold can be passed, at which time the proposal is passed and funds are obtained.
2. Consensus of belief (for consecutive proposals)
In this version of CV, there is no single threshold to recommend termination of the proposal-on the contrary, group sentiment can fluctuate forever within a certain range, as a kind of “dynamic average consensus.” This may suit a community that wants certain aspects of its socio-economic system to be constantly determined by collective emotions. For example, the tax rate of community token entry/exit tax (also known as Tobin Tax), or the monthly community UBI amount raised by continuously minting local currency.
Consensus of belief, community parameters are dynamically set based on the continuous input of members.
This real-time governance tool may be more useful after community decision-making and decision-making. We look forward to continuing this research and creating an open source foundation for models that can be iterated into a wide variety of scenarios to promote collective wisdom. You can find some of the next steps we thought of in the conclusion section of the model notes.
Why vote for belief in cadCAD?
In cyber-physical systems such as the international power grid, the global flight network, or the social economic community ecosystem of the token project, engineers have modeled a simulated copy of their system, called a data twin. These data twins help us understand and manage complex systems, which may have trillions of data points and are constantly changing. These simulated copies guide information into the path, allowing humans to understand what is happening in the ecosystem at a high level, so that they can intervene in the right place and at the right time. (Just like pressing the breaker switch after clearing the fault in the power system).
A digital twin is a simulated replica of a real-world complex system, like an airplane engine or a token economy.
A data twin is a simulated replica of a complex system in the real world, such as an airplane engine or a token economy.
A data twin can be thought of as a flight simulator. It can be used to run your system through billions of different “tests”, changing one parameter at a time to see what effects will make your system out of balance. As engineers, the top priority is to ensure public safety. We must understand the boundaries and critical points of the system and ensure that appropriate mechanisms are established to push the system back into equilibrium when it enters the safe boundary conditions.
This cadCAD model is a data twin of belief voting, suitable for 1Hive DAO ecosystem. It can be used to provide operational support during the design phase and the decision-making process in the continuous governance of the 1Hive system, providing 1Hive members with an early form of Computer Aided Governance, computer aided governance.
What is cadCAD?
cadCAD (short for complex and adaptive dynamic computer-aided design) is a Python-based open-source modeling framework for research, verification, and computer-aided design of complex systems. cadCAD supports different multi-scale system modeling methods and can be easily integrated with common empirical data science workflows. Monte Carlo method, A/B test and parameter scanning functions are natively supported and optimized for the tool.
Given a model of a complex system, cadCAD can simulate the possible effects of a set of operations on it. This helps individuals or organizations make informed, rigorously tested decisions on how to best modify or interact with complex systems to achieve their desired goals.
Simulation & in-depth exploration notebook
As part of the modeling work, these notebooks were made to explain the deeper details of the belief voting parameters, derivation and modeling. These notebooks are a mixture of code snippets, descriptions, simulations, and a lot of background, which can make you more familiar with the concept of CV, and may even delve into the modeling of similar systems, or use cadCAD to further extend this model:
Notebook commentary video: Please browse the belief voting notebook with Andrew Clark and Jeff Emmett to find a high-level interpreter.
Simulation notebook (V3): The latest notebook iteration of belief voting, modeling the deployment of 1Hive.
Deduction of Belief Voting Algorithm: For a deeper understanding of the CV algorithm, including its mathematical derivation, please read this notebook.
Deriving the Alpha parameter: To learn more about the considerations of the alpha parameter, which sets the half-life of the belief and determines its “charge” speed, please read this notebook.
Explaining the Trigger Function: To learn more about the trigger function equation and how to pass suggestions from the candidate state to the active state, please read this notebook.
Model structure: more in-depth study of the structure of the cadCAD model. If you are learning how to use system modeling in cadCAD, please read this document.
How to use and improve the model?
It should be noted that although models can help system designers avoid certain hazards or optimize for certain results, they still need to be understood and used in an appropriate context. It is impossible to draw a “general conclusion” from these models, because deployments in different environments may vary greatly.
The cadCAD model should be regarded as a flight simulator. They cannot tell you exactly how to perform your tasks, because each task (ie deployment instance) is different. But they can be parameterized and customized to suit various implementations to provide operational decision support for your community. .
In this case, what can this model tell us? For example, it can provide us with information on the following issues:
What kind of systemic impact will aggravating beliefs too fast or too slow?
How can we optimize proposal fees to prevent spam proposals from destroying belief accumulation?
Can my ecosystem operate sustainably in the long-term (under existing model assumptions)?
It is also worth noting that these are the initial basic models, and can be improved and iterated by adding complexity. Real-time data in the pilot deployment can also help adjust the model to more accurately represent the reality.
Current belief voting experiment
Commons Stack Community Fund-Panvala League
The Commons Stack launched a pilot project of belief voting to determine how to match fund allocations through the Panvala League and Gitcoin Grants projects. This is all thanks to the hard work of the Panvala, 1Hive, BlockScience, Aragon and xDAI communities, as well as the efforts of @sembrestels for several sleepless nights.
Help us distribute the $170,000 $PAN match in the next round of gitcoin grants by sending your favorite grant using the Conviction Voting signal with CSTK tokens!
1Hive belief voting DAO
Since the beginning of 2019, the 1Hive community has been working with BlockScience and Commons Stack to actively develop belief voting contracts. They currently have a DAO on the xDAI network of 1hive.org, which uses local governance tokens (Honey) to distribute funds to various proposals through belief voting.
To see the belief vote to deploy smart contracts in the initial user interface, watch this video, interview 1Hive CV Dapp and Zartler, or check out 1Hive’s Github.
Public simulator
The Commons Stack has been working on the Commons Simulator, which is an educational tool (also a work of art!) that can help users understand these novel governance tools. The progress of the conviction vote in Commons Simulator can be viewed in the Commons Stack Github repository.
Support this important governance research and development
Through cooperation with 1Hive, BlockScience, Aragon and Commons Stack, the open source model will continue to be improved with additional features. It can be used as a useful education and decision support tool for those who want to learn how to use cadCAD to model and simulate systems. The modeling and simulation system integrates algorithm policy tools to pave the way for future computer-aided governance.
Please consider contributing to this project to support more work on the belief voting model. This grant will receive more funding through secondary matching:
👉👉
Or donate directly to this ERC-20 address:
👉
Participate in exploring the simulated governance environment and join the cadCAD community:
- cadCAD Discord :
- cadCAD Telegram :
- cadCAD , featuring tutorials & community calls
Thanks to the editors of this article Andrew Clark, Luke Duncan, Griff Green, Michael Zargham, and Jessica Zartler.
