Idea: Research Coin, second attempt
June 10, 2017
- Each university has 100 tokens and they can sell them at any time
- Token holders of a specific university can use these tokens to bet on university paper’s being accepted.
- Betting on papers from good university will be a low risk investment that will make the university token go up in price.
The assumptions are the same: there is a trusted oracle which is the conference accepting great papers. The motivations are the same: (1) create new funding for academia, (2) make good research, (3) incentivize peer review.
Disclaimer: I actually DON’T like this idea since it optimises for short-term good papers. I think universities should play a longer term than just “have papers accepted”
This idea comes from a midnight conversation with Matt from MIT. It’s an improvement (or different attempt) for a Research Coin from last time.
New token model
Each institution, research lab or group can create an “org”. Each org is assigned 100 org-tokens that are specific to the org. For example, the org “MIT” will have MITCoin, “Cornell” will have CornellCoin and so on. An org can decide whether to sell or not their 100 org-tokens.
Note: Even a specific lab, can decide to create an org, for example MIT PDOS can create a PDOS token.
A peer-to-peer network of peer reviewers can review the papers, and bet whether a paper will be accepted or not for a particular conference. On the conference deadlines, all the winning reviewers get rewarded. This system could re-create interesting behaviors, where peer-reviewers will try to look for great papers that need some more work and give comments to the authors.
Similarly to Research Coin, anyone can bet on a paper being accepted at a specific conference. However, one can bet on a paper by using the org-token representing the org used by the researchers. In other words, if I want to bet that the MIT paper will be accepted, I must bet tokens from MIT.
Bets run in rounds, for each conference, owners of tokens can bet org-tokens on papers from a specific org. When papers are selected by the conference, all the lost bets are split proportionally (however, I am not sure how, this is actually non-trivial) to the amount they bet.
The intuition here is that: as universities (and research labs) do great research, the value of their tokens would, since many would prefer to bet on their research. I expect that the token of a very successful university will have a higher price, but bets will have a low risk, hence a lower return.
I am overall convinced that this is NOT a good idea yet. I am not sure if betting on research paper passing through a conference is the best use of a currency.
- Would this lead to making great research or would this be optimize for short-term good publishable research?
- Can universities do “ICO”s for selling betting credit for their university?
- Would this really increase peer-review or there will just be speculation?
- Nicola Greco,
Keep on rocking the decentralized web