I think there’s an asymmetry problem here. An early downvote hides the post from the frontpage and impedes it from getting more evaluations, so it’s a pit that’s hard to get out from (without, say, creating a different post asking why your first post is in the pit :P). An early upvote, on the other hand, exposes the post to more evaluations by keeping it longer on the frontpage, and it’s easy for later downvotes to push it back down, so it’s more like a slippery hill than a pit.
So I think a mechanism that would avoid premature burials of posts would overcome some of the noise and lead to more information being incorporated into the “final” evaluation of the post (the karma it stabilizes on).
I agree this is true for content by new users, but honestly, we kind of need to hide content from most users from the frontpage until someone decided to upvote it.
For more active users, their strong-vote strength gets applied by default to the post, which helps a good amount with early downvotes not hurting visibility that much.
I don’t think that’s enough? I have 3.5k Karma, which gives me a strong vote power of 7, but when I made this post the other post was on 5 Karma and long gone from the front page. It only started gaining karma and came back to the frontpage after I made this post.
And I kinda dislike “why am I getting downvoted” posts, so I would like mechanisms that make them unnecessary.
I mean, the difference between 7 and 5 karma on frontpage ranking is miniscule, so I don’t think that made any difference. The real question is “why did nobody upvote it”? Like, I think there physically isn’t enough space on the frontpage to give 5 karma posts visibility for very long, without filling most of the frontpage with new unvetted content.
I don’t think comparing 5 to 7 is correct, because we don’t want to compare to downvote to no-vote, we want to compare one ordering of votes to another ordering of votes. So, what would be the difference if it went up before it went back down again, rather than first go down like it has.
I think we do agree that if I didn’t ask why it was downvoted it would have remained at 5 rather than go up to 15, and that this is suboptimal, right?
To me it feels like mid-popularity posts are affected too much by noise and when they get posted.
I think there’s an asymmetry problem here. An early downvote hides the post from the frontpage and impedes it from getting more evaluations, so it’s a pit that’s hard to get out from (without, say, creating a different post asking why your first post is in the pit :P). An early upvote, on the other hand, exposes the post to more evaluations by keeping it longer on the frontpage, and it’s easy for later downvotes to push it back down, so it’s more like a slippery hill than a pit.
So I think a mechanism that would avoid premature burials of posts would overcome some of the noise and lead to more information being incorporated into the “final” evaluation of the post (the karma it stabilizes on).
I agree this is true for content by new users, but honestly, we kind of need to hide content from most users from the frontpage until someone decided to upvote it.
For more active users, their strong-vote strength gets applied by default to the post, which helps a good amount with early downvotes not hurting visibility that much.
I don’t think that’s enough? I have 3.5k Karma, which gives me a strong vote power of 7, but when I made this post the other post was on 5 Karma and long gone from the front page. It only started gaining karma and came back to the frontpage after I made this post.
And I kinda dislike “why am I getting downvoted” posts, so I would like mechanisms that make them unnecessary.
I mean, the difference between 7 and 5 karma on frontpage ranking is miniscule, so I don’t think that made any difference. The real question is “why did nobody upvote it”? Like, I think there physically isn’t enough space on the frontpage to give 5 karma posts visibility for very long, without filling most of the frontpage with new unvetted content.
I don’t think comparing 5 to 7 is correct, because we don’t want to compare to downvote to no-vote, we want to compare one ordering of votes to another ordering of votes. So, what would be the difference if it went up before it went back down again, rather than first go down like it has.
I think we do agree that if I didn’t ask why it was downvoted it would have remained at 5 rather than go up to 15, and that this is suboptimal, right?
To me it feels like mid-popularity posts are affected too much by noise and when they get posted.