Nicely written. This seems like the best evolution of the multi-tiered white list idea.
Since we do want to encourage excellence in science without discouraging new/unpopular projects, I like the solution outlined in the first proposal where project #1 receives 2X GRC compared to project #26.
How close do you think we could get to a Nash equilibrium with these solutions?
The other side of the coin is, how do we tally the votes for the ranking? It would have to be based on the current voting system with a lot of weight going to who has the most coins. Is this the best way?
In the first two, at equilibrium everyone would be crunching the top project, that's the problem. In the current setup, there's massive discrepancies, but eventually at NE each project would get exactly the same amount of computational power. Here, theoretically, there's no limit, though in practice it would probably cause similarly-sized discrepancies between projects that we have now (depending on the multiplier factor), but it would be intentional and subject to community input.
In the third, what's likely is that NE wouldn't be reached and the same scenario that we have now would occur - some users could/would still get more GRC than others. The difference here is, 1) it's more random - i.e. not easy to predict which will get more - as opposed to the current setup where by default less popular projects give more GRC, and 2) the massive discrepancy in received GRC between these projects would be greatly diminished, so while crunchers might still get more GRC with their hardware on some projects rather than others, these effects would be greatly reduced. If this system actually reached NE, every project would get computational power equal to its share of the GRC.
Good question. I consider the voting-weight issue a completely separate one from this, since these proposals could theoretically be implemented with any given voting system, including the one we have.