The right bar that goes off the page is so far unexplained for me − 921 users joined in September 2011, more than three times the number in the months before and after it. If you happen to know what caused that, I would be very interested in finding out.
My prediction is something HPMOR related- either more links to lesswrong in the Author’s Notes, or HPMOR itself had a spike that month.
Another way to cut down on new-new interaction is to limit the number of comments someone can make in a time period- if people can only comment once an day until their karma hits 20, and then once an hour until their karma hits 100, and then they’re unrestricted, that will explicitly encourage lurking / paying close attention to karma among new members. (It would be gameable, unless you did something like prevent new members from upvoting the comments of other new members, or algorithmically keeping an eye out for people gaming the system and then cracking down on them.)
[edit] The delay being a near-continuous function of the karma- say, 24 hours*exp(-b karma)- might make the incentives better, and not require partitioning users explicitly. No idea if that would be more or less effort on the coding side.
Problem: Limiting the number of posts doesn’t limit the number of comments, so they’d still be able to overwhelm older users with newbie comments or create their own culture in the comments. I think this idea would be ineffective unless, by “posts” you meant “comments” (or added some similar plan for comments).
Thanks, Vaniver. The OP has been updated. I also used find to see whether there were other ambiguities around the word “post” in the OP. Caught a few. (:
Limiting the posts would cause new users to lose momentum. A lot of them might lose steam after joining and give up. That would be risky. Also, because a large proportion would give up, this would filter users. We’d end up with a larger proportion of the type of user persistent enough to tolerate this. I don’t know what that sort of person looks like.
I’ll add the idea to the pile, but I can’t really sell it if those things aren’t addressed.
My prediction is something HPMOR related- either more links to lesswrong in the Author’s Notes, or HPMOR itself had a spike that month.
Another way to cut down on new-new interaction is to limit the number of comments someone can make in a time period- if people can only comment once an day until their karma hits 20, and then once an hour until their karma hits 100, and then they’re unrestricted, that will explicitly encourage lurking / paying close attention to karma among new members. (It would be gameable, unless you did something like prevent new members from upvoting the comments of other new members, or algorithmically keeping an eye out for people gaming the system and then cracking down on them.)
[edit] The delay being a near-continuous function of the karma- say, 24 hours*exp(-b karma)- might make the incentives better, and not require partitioning users explicitly. No idea if that would be more or less effort on the coding side.
Problem: Limiting the number of posts doesn’t limit the number of comments, so they’d still be able to overwhelm older users with newbie comments or create their own culture in the comments. I think this idea would be ineffective unless, by “posts” you meant “comments” (or added some similar plan for comments).
I meant “comments” by “posts.” I’ll edit the grandparent to be clearer.
Thanks, Vaniver. The OP has been updated. I also used find to see whether there were other ambiguities around the word “post” in the OP. Caught a few. (:
Limiting the posts would cause new users to lose momentum. A lot of them might lose steam after joining and give up. That would be risky. Also, because a large proportion would give up, this would filter users. We’d end up with a larger proportion of the type of user persistent enough to tolerate this. I don’t know what that sort of person looks like.
I’ll add the idea to the pile, but I can’t really sell it if those things aren’t addressed.