Predicted outcome: within a couple of weeks, blue/green will have understood but undocumented positive/negative associations. Votes will be noisier, though, thanks mostly to confused newcomers and the occasional contrarian pursuing an idiosyncratic interpretation. Complaints about downvotes, and color politics jokes, will both become more common.
p = 0.7 contingent on implementation for core claim, .5-6 range for corollaries.
Proposed chaotic refinement: Blue/green, but switch them every 18 to 30 hours (randomly sampled, uniform distribution).
(ETA: Upon reflection days or weeks would be better, to increase chaos/noise ratio. Would also work better with prominent “top contributors for last 30 days” lists for both blue and green, and more adulation/condemnation based on those lists.)
Other refinements: each person is randomly permanently assigned either: blue/green OR they see blue/green but it’s actually green/blue behind the scenes. This makes any explicit discussion of blue/green more difficult.
Or: Each person actually has grue and bleen buttons. At some time t, they are suddenly voting for the other colours. An extended form of this looks similar to your ETA.
Predicted outcome: within a couple of weeks, blue/green will have understood but undocumented positive/negative associations. Votes will be noisier, though, thanks mostly to confused newcomers and the occasional contrarian pursuing an idiosyncratic interpretation. Complaints about downvotes, and color politics jokes, will both become more common.
p = 0.7 contingent on implementation for core claim, .5-6 range for corollaries.
0.7 strikes me as low.
Proposed chaotic refinement: Blue/green, but switch them every 18 to 30 hours (randomly sampled, uniform distribution).
(ETA: Upon reflection days or weeks would be better, to increase chaos/noise ratio. Would also work better with prominent “top contributors for last 30 days” lists for both blue and green, and more adulation/condemnation based on those lists.)
Other refinements: each person is randomly permanently assigned either: blue/green OR they see blue/green but it’s actually green/blue behind the scenes. This makes any explicit discussion of blue/green more difficult.
Or: Each person actually has grue and bleen buttons. At some time t, they are suddenly voting for the other colours. An extended form of this looks similar to your ETA.