By contrast, positive feedback does not appear to influence authors much at all.“Not only do authors of negatively-evaluated content contribute more, but also their future posts are of lower quality, and are perceived by the community as such,” they say. And it gets worse: “These authors are more likely to subsequently evaluate their fellow users negatively, percolating these effects through the community.”
Damn, that really spells out the challenge of creating a positive community experience.
I don't think this is probably a good interpretation of the data. What do they mean by "doesn't influence"? I suppose it probably actually means, "doesn't alter their observed behavior," which is totally different from "doesn't affect." Positive reinforcement affects all of us, but if everyone seems to appreciate what I'm doing, then why change? If someone tells me I'm an idiot, then maybe I'm going to do something different the next time (e.g. try harder, strike back, etc).By contrast, positive feedback does not appear to influence authors much at all.
No kidding. The linked article goes quite a ways to explaining much of what I saw yesterday on http://www.reddit.com/r/haskell/ . It was painfully apparent the user who posted the following question ( http://www.reddit.com/r/haskell/comments/25rzci/wanting_to_learn_haskell_then_i_hit_this_error/ ) was a newbie to both Reddit and Haskell. The result of the query was: two people tried to actively help the newbie (and, judging by the responses, the newbie seems to have understood this), one person was "helpful" (and the newbie responded in kind), there were several anonymous downvotes, and the single truly neutral question posed by the newbie “What are the ways the Haskell community have settled on to get better diagnostic messages out of the compiler?” went unanswered. Based on the article linked from medium.com, it would be fair to say the Haskell community has lost a new recruit. Sad, just sad.