RE: An Opponent of the Exponent: Making the Case for Vshare Linearity
This is gong to be a nonsense analysis that assumes voting behavior doesn't change under a completely different ruleset.
I don't think everything will stay exactly the same, voting habits will probably change. but i also don't think a change like this is enough to make it freaky friday.
And there is actually information that can be predicted with certainty (for example, the value of a users vote , provided we can count on vote utilization to remain relatively constant.)
The n log n curve that has been proposed to be used for a separate comment would seems like it would be fine to just use for everything. Most of the extreme effects you describe in your post due to n^2 would no longer exist.
Yeah, i think n log n would solve most of the problems and a compromise solution like that seems like it has way more support.
Then the change isn't worth making (to be clear I think it is worth making, but I think it would cause, over time, tectonic shifts in the voting behavior and user perceptions, both at the high and and the low end).