Boosting: a prediction market on ideas

Boosting Curves

In the current boosting design, if people boost something early they will get a percentage of future Cred that flows through that thing. This does not, however, reward early boosters more than late stage boosters. This means that Cred whales can sit back and wait for ideas to get traction, then boost them to get a percentage of future Cred. When they do, they will get Cred in proportion to how much they boosted. This misaligns incentives between “Cred hunters” who seek out promising undervalued/appreciated ideas and “Cred whales” who can then swoop in to take the lion’s share of the benefits once ideas are discovered.

For example: if someone boosts something early on when it’s still underappreciated then they’re getting a small piece of a small Cred flow. Then as the thing becomes popular a Cred whale will be able to boost it and get a larger share of a larger flow of rewards. The Cred hunter found the valuable contribution, put Grain at risk to signal to the community that the contribution was valuable, but then once the contribution gets traction and more Cred starts flowing through it a Cred whale would be able to boost the contribution 10X what the Cred hunter did and thus get 10X the Cred rewards. This favors Cred whales.

A boosting curve could reward early boosters with more rewards than later stage boosters. This would acknowledge the risk that early boosters take on relative to later stage boosters. This would encourage boosting early and often. The curve might look as follows where Y is a percentage of Cred flows and X is total amount boosted.

boosting-curve

To make this work in practice perhaps a certain percentage of everything could be reserved for boosting (20/80?) and then the boosting percentage (20%?) could be split among boosters based on the curve.

This is just an idea, but I think it’s a good one to align incentives between various levels of engagement and Grain holdings. Curious if anyone sees any problems with this (first conceptually because we’re in the design stage, then with implementation) and/or ways this can be improved.

1 Like