I don't like the lack of down-voting. If you're afraid of good things being down-voted too soon, why don't you do a nonlinear transformation from the number of

I don't like the lack of down-voting. If you're afraid of good things being down-voted too soon, why don't you do a nonlinear transformation from the number of down-voters to the negative score? * example: until the 10th downvote, none of the downvotes count * example: based upon a similarity score of people's upvotes, look for broad-based support for downvotes Prospect Theory implies we need downvotes.