The worry is this: Two systems of voting/liking may be “equally good” in the sense that they e.g. incentivize 90% of good comments and disincentivize 10% of good comments, but the overlap of good things they disincentivize may be just 1%.
(This seems plausible given the differences in the mechanism, way how it is displayed, and how it directs attention)
It makes a difference if you are using two different randomly broken system, or two coppies of one.
It’s a different question.
The worry is this: Two systems of voting/liking may be “equally good” in the sense that they e.g. incentivize 90% of good comments and disincentivize 10% of good comments, but the overlap of good things they disincentivize may be just 1%. (This seems plausible given the differences in the mechanism, way how it is displayed, and how it directs attention)
It makes a difference if you are using two different randomly broken system, or two coppies of one.