ethberlinzwei / Find-A-Team

Team formation repo - just file issues with ideas!
MIT License
13 stars 11 forks source link

[NEW] Long-term Climate Prediction Markets #18

Open tobias-kaiser opened 4 years ago

tobias-kaiser commented 4 years ago

Introduction

Hey all,

I am a blockchain researcher on Tokenomics and decentralized business models. I have a very interdisciplinary background that combines applied computer science, cognitive science, economics, game theory, and many more. Currently, I am working on a project where we seek to realize a decentralized incubator and marketplace for ideas. As my research entails writing a paper on predictive markets, I will be attending Blockchain Week to hopefully get some inspiration and of course to get into contact with others in the industry.

Since Dappcon is hosted by Gnosis and since climate change has been a hot topic this year (pun intended), my idea is to combine the two into a prediction market for long-term climate change.

Idea description

The problem with traditional, centralized climate research is that, nowadays, it cannot be conducted objectively anymore. While the climate models compiled by the IPCC are still our best guess regarding long-term climate predictions, the financial incentives in climate research favor extreme positions.

On the one hand, it shouldn’t be a wonder to anyone that the scientists most involved in climate research are also the most alarmist when it comes to anthropogenic climate change and its implications on humanity and the ecosystem. At the very least, they would jeopardize themselves financially, if they stated otherwise. The same is of course true for private climate-sceptical research institutions that are oftentimes funded by the oil industry. On the other hand, governments abuse climate change as a vehicle to introduce new taxes, or raise existing ones. Besides the usual publication bias existing in modern science, the politically-laden notion of scientific consensus regarding anthropogenic climate change creates a toxic environment, in which climate scientists may become reluctant to come forward with evidence that challenges the mainstream climate models.

Prediction markets are a useful tool for forecasting the future and have been successfully implemented for events relating to sports, popular culture, election results, weather, epidemic outbreaks, and others. When such markets are liquid enough and attract a large diversity of opinions, they typically outperform all other prediction methods. Transferring the idea of prediction markets to climate change would solve the problem of credibility, as anyone who wants to make a claim must back it up by putting money on the line. Moreover, if the liquidity is high enough, this can create financial incentives to conduct independent, decentralized climate research in order to find a more accurate prediction of long-term climate change.

Skillset

Frankly, I must admit that I am not a developer and this is my first Hackathon ever. I am therefore planning this as a research project in order to come up with a viable business/tokenomics model and a position paper or, when there is enough time, a full-fledged whitepaper. I am therefore looking to brainstorm with people who have a similar skill set as me (tokenomics, decentralized business models, game theory, mechanism design, policy making, governance, etc). Bonus points if you have prior knowledge of prediction markets, market makers, or climate research. Since we will probably need to overcome some regulatory hurdles, legal experts might be helpful as well.

Some of the question I'd like to discuss:

I do however think that it is within the realm of possibility to implement an MVP, as the basic smart contracts that are needed for a prediction market are available from Gnosis. We would only need to customize them and build a nice UI. If we want to take this step, we need both backend (Solidity/JavaScript) and frontend developers.

Communication

Alternatively, you can comment below, or talk to me during one of the events at Blockchain Week.

elliotolds commented 4 years ago

Hey, I'm also interested in working on prediction markets at the hackathon.

IMO a charity/nonprofit makes a lot of sense for a business model for this idea. We set up a market on Gnosis and put a UI on top of it, then have a separate contract that people donate to which funds the market. Their incentive is they want to make the market more accurate by providing liquidity. We figure out how to make the market as accurate as possible, given a certain amount of funding.

I have a project that I'm starting outside of the hackathon about getting from our current situation to Futarchy as soon as possible. 'Donating to prediction markets' is one of the ideas I've been thinking about: https://elliot.miraheze.org/wiki/Improving_societal_decision_making_(abridged)

tobias-kaiser commented 4 years ago

I'm rather thinking about a tokenomics solution to funding and liquidity, but let's have a chat at either Dappcon or the Hackathon. I'm sure we can work together on our ideas.