3 ms·

I find the idea that papers are judged by a simple up-vote/down-vote terrifying. Popular academics will post their new paper on Twitter, which will immediately

I find the idea that papers are judged by a simple up-vote/down-vote terrifying. Popular academics will post their new paper on Twitter, which will immediately lead to a round of upvotes from their friends and followers. People who are currently out of favour will be downvote-bombed. I see no suggested way of stopping this -- and I hope noone thinks the answer is academics are "above that". I'd suggest changing from "people may leave a comment if they wish when they downvote" to "any vote requires at least 200 words of justification". In general I think it's important reviewing requires thought and justification.

dbingham 4y ago
This is helpful feedback, and requiring responses with votes was already something we were pondering. And I'm increasingly leaning in that direction. One thing that wasn't covered in the blog post was the intention for downvotes to cost reputation, same as on SE right now. That should help disincentivize brigading to some degree. The field system may also help counter that tendency to some degree. By choosing which fields you publish in, you can choose the subset of people able to review and vote. But there is also a degree to which we will try to counter this with culture. The cultures of internet communities can be extremely stick and resistant to change, both good and bad. Just think about Reddit vs StackExchange. Roughly similar systems (though with very different goals and guidelines), very different cultures. If we can establish a strong culture on the platform from the get go - a culture of low ego collaboration and seeking good science above all else, that can potentially help counter some of the worse tendencies of wider academia as more and more of it moves into the platform. At least in the early days, before we can afford active moderation. In the long run, active moderation, and tools for community moderation, can help as well.

dbingham 4y ago

This is helpful feedback, and requiring responses with votes was already something we were pondering. And I'm increasingly leaning in that direction. One thing that wasn't covered in the blog post was the intention for downvotes to cost reputation, same as on SE right now. That should help disincentivize brigading to some degree. The field system may also help counter that tendency to some degree. By choosing which fields you publish in, you can choose the subset of people able to review and vote. But there is also a degree to which we will try to counter this with culture. The cultures of internet communities can be extremely stick and resistant to change, both good and bad. Just think about Reddit vs StackExchange. Roughly similar systems (though with very different goals and guidelines), very different cultures. If we can establish a strong culture on the platform from the get go - a culture of low ego collaboration and seeking good science above all else, that can potentially help counter some of the worse tendencies of wider academia as more and more of it moves into the platform. At least in the early days, before we can afford active moderation. In the long run, active moderation, and tools for community moderation, can help as well.