I’m not sure if this is the right place to post this, but I couldn’t find a better place in 5 minutes of searching.
I think that the mix of “recent” and “upvoted” in the Latest Posts section skews way too heavily towards total karma, even for new posts with a tiny number of votes.
As a specific example, my link to a research post on polyamory, Rationality and correlated beliefs was strongly downvoted by the first person who saw it. This set its total karma to 1, which caused it to drop out of the front page within 2 hours of publication. It currently has 3 karma/3 votes. Now, when this happened before I just shrugged and assumed my posts were not good enough. But this is a super high-effort post containing a ton of novel data and analysis on a topic of interest to the community. I dare say it meets the frontpage criteria outlined above. It was linked by Tyler Cowen on Marginal Revolution and currently has 6,200 pageviews, but only 39 of them are from LessWrong.
I think that the “upvoted” part of the algorithm needs to either be downweighted, or its weight should be a function of the total number of votes. 0.5 karma/vote over 100 votes is a signal of low quality, even if the total karma is a healthy 50. 0.5 karma/vote over 2 votes is a signal of nothing at all, but its enough to kick the post out of the frontpage.
Hmm. Indeed I am surprised it has such a low karma score, it was a high-effort piece of data analysis about the community, and I learned useful things from it. (I have just strong-upvoted it, so it’s at 12 now.) I think this does update me that posts should be given more time in the recent posts list when posted, even if low-karma. (Thx for pointing this out.)
I’m not sure if this is the right place to post this, but I couldn’t find a better place in 5 minutes of searching.
I think that the mix of “recent” and “upvoted” in the Latest Posts section skews way too heavily towards total karma, even for new posts with a tiny number of votes.
As a specific example, my link to a research post on polyamory, Rationality and correlated beliefs was strongly downvoted by the first person who saw it. This set its total karma to 1, which caused it to drop out of the front page within 2 hours of publication. It currently has 3 karma/3 votes. Now, when this happened before I just shrugged and assumed my posts were not good enough. But this is a super high-effort post containing a ton of novel data and analysis on a topic of interest to the community. I dare say it meets the frontpage criteria outlined above. It was linked by Tyler Cowen on Marginal Revolution and currently has 6,200 pageviews, but only 39 of them are from LessWrong.
I think that the “upvoted” part of the algorithm needs to either be downweighted, or its weight should be a function of the total number of votes. 0.5 karma/vote over 100 votes is a signal of low quality, even if the total karma is a healthy 50. 0.5 karma/vote over 2 votes is a signal of nothing at all, but its enough to kick the post out of the frontpage.
Hmm. Indeed I am surprised it has such a low karma score, it was a high-effort piece of data analysis about the community, and I learned useful things from it. (I have just strong-upvoted it, so it’s at 12 now.) I think this does update me that posts should be given more time in the recent posts list when posted, even if low-karma. (Thx for pointing this out.)