rep_level level median_next_post median_next_edit
--------- ---------------------------- ---------------- ----------------
5 participate in meta 13 hours 70 hours
10 remove new user restrictions 11 63
15 vote up 8 53
20 talk in chat 8 48
50 comment everywhere 6 33
75 set bounties 6 26
100 edit community wiki 5 25
125 vote down 5 23
250 view close votes 4 19
500 access review queues* 2 3
1000 established user 3 13
1500 create tags 5 13
2000 edit questions and answers 4 8
2500 create tag synonyms 3 8
3000 cast close and reopen votes 4 6
5000 approve tag wiki edits 4 6
10000 access to moderator tools 6 5
15000 protect questions 9 12
20000 trusted user 10 15
| rep_level | level | median_next_post | median_next_edit |
|---|---|---|---|
| 5 | participate in meta | 13 hours | 70 hours |
| 10 | remove new user restrictions | 11 | 63 |
| 15 | vote up | 8 | 53 |
| 20 | talk in chat | 8 | 48 |
| 50 | comment everywhere | 6 | 33 |
| 75 | set bounties | 6 | 26 |
| 100 | edit community wiki | 5 | 25 |
| 125 | vote down | 5 | 23 |
| 250 | view close votes | 4 | 19 |
| 500 | access review queues* | 2 | 3 |
| 1000 | established user | 3 | 13 |
| 1500 | create tags | 5 | 13 |
| 2000 | edit questions and answers | 4 | 8 |
| 2500 | create tag synonyms | 3 | 8 |
| 3000 | cast close and reopen votes | 4 | 6 |
| 5000 | approve tag wiki edits | 4 | 6 |
| 10000 | access to moderator tools | 6 | 5 |
| 15000 | protect questions | 9 | 12 |
| 20000 | trusted user | 10 | 15 |
rep_level level median_next_post median_next_edit
--------- ---------------------------- ---------------- ----------------
5 participate in meta 13 hours 70 hours
10 remove new user restrictions 11 63
15 vote up 8 53
20 talk in chat 8 48
50 comment everywhere 6 33
75 set bounties 6 26
100 edit community wiki 5 25
125 vote down 5 23
250 view close votes 4 19
500 access review queues* 2 3
1000 established user 3 13
1500 create tags 5 13
2000 edit questions and answers 4 8
2500 create tag synonyms 3 8
3000 cast close and reopen votes 4 6
5000 approve tag wiki edits 4 6
10000 access to moderator tools 6 5
15000 protect questions 9 12
20000 trusted user 10 15
| rep_level | level | median_next_post | median_next_edit |
|---|---|---|---|
| 5 | participate in meta | 13 hours | 70 hours |
| 10 | remove new user restrictions | 11 | 63 |
| 15 | vote up | 8 | 53 |
| 20 | talk in chat | 8 | 48 |
| 50 | comment everywhere | 6 | 33 |
| 75 | set bounties | 6 | 26 |
| 100 | edit community wiki | 5 | 25 |
| 125 | vote down | 5 | 23 |
| 250 | view close votes | 4 | 19 |
| 500 | access review queues* | 2 | 3 |
| 1000 | established user | 3 | 13 |
| 1500 | create tags | 5 | 13 |
| 2000 | edit questions and answers | 4 | 8 |
| 2500 | create tag synonyms | 3 | 8 |
| 3000 | cast close and reopen votes | 4 | 6 |
| 5000 | approve tag wiki edits | 4 | 6 |
| 10000 | access to moderator tools | 6 | 5 |
| 15000 | protect questions | 9 | 12 |
| 20000 | trusted user | 10 | 15 |
I estimated enthusiasm following a privilege level by doing a longitudinal study of median time-to-next-action. Looking at the 868 Stack Overflow users with more than 20k whose accounts were created after 2010年10月05日 (the date privilege notifications were first issued privilege notifications were first issued), we can see how quickly they posted and edited after gaining each new privilege:
* The access to First Post and Late Answer review queues is an outlier because it was introduced around August, 2013 introduced around August, 2013. It’s a good guess that most of these users got notified of their new privilege when they had far more than 500 reputation.
I estimated enthusiasm following a privilege level by doing a longitudinal study of median time-to-next-action. Looking at the 868 Stack Overflow users with more than 20k whose accounts were created after 2010年10月05日 (the date privilege notifications were first issued), we can see how quickly they posted and edited after gaining each new privilege:
* The access to First Post and Late Answer review queues is an outlier because it was introduced around August, 2013. It’s a good guess that most of these users got notified of their new privilege when they had far more than 500 reputation.
I estimated enthusiasm following a privilege level by doing a longitudinal study of median time-to-next-action. Looking at the 868 Stack Overflow users with more than 20k whose accounts were created after 2010年10月05日 (the date privilege notifications were first issued), we can see how quickly they posted and edited after gaining each new privilege:
* The access to First Post and Late Answer review queues is an outlier because it was introduced around August, 2013. It’s a good guess that most of these users got notified of their new privilege when they had far more than 500 reputation.
I'm suggesting these privileges because comments are one of the few things that elected and appointed diamond moderators can do something about that the rest of the community cannot. Other than flagging, there's no way for regular users to fix problems with the comments on a post. Having a powerful privilege to aspire to provides a better incentive than providing somewhat less-powerful privileges. Anecdotally, people most look forward to the 10k privileges 10k privileges which introduces the final piece of community moderation: post deletion.
These are exceptional users, so they participate (as a group) far more than others. If we look at users with more than 5 reputation, the median first post after getting participate on meta participate on meta is 215 hours. Compared to their later behavior, trusted users start slow. Notice that they also slow down again after hitting 10k or so. Critically, the slow down after 10k also occurs for 30k users (N=429), so it’s not just an artifact of the cutoff.
There’s circumstantial evidence that increasing the top privilege level increased participation at higher levels. More to the point, the relatively underpowered 15k 15k and 20k 20k levels appear to be less effective at motivating users than 10k powers. So introducing something really helpful, such as comment moderation, seems in order.
However, I think comment moderation is mostly a waste of time. I'd much rather let the system handle comments the system handle comments. Unfortunately, there are already so many comments that should have been edits (or even answers) instead that we can't blindly hide/delete old comments. Creating this privilege will mean committing to manual comment moderation for the foreseeable future. Comment moderation is tedious and unrewarding so I'd rather have our top users do something more productive with their time.
I'm suggesting these privileges because comments are one of the few things that elected and appointed diamond moderators can do something about that the rest of the community cannot. Other than flagging, there's no way for regular users to fix problems with the comments on a post. Having a powerful privilege to aspire to provides a better incentive than providing somewhat less-powerful privileges. Anecdotally, people most look forward to the 10k privileges which introduces the final piece of community moderation: post deletion.
These are exceptional users, so they participate (as a group) far more than others. If we look at users with more than 5 reputation, the median first post after getting participate on meta is 215 hours. Compared to their later behavior, trusted users start slow. Notice that they also slow down again after hitting 10k or so. Critically, the slow down after 10k also occurs for 30k users (N=429), so it’s not just an artifact of the cutoff.
There’s circumstantial evidence that increasing the top privilege level increased participation at higher levels. More to the point, the relatively underpowered 15k and 20k levels appear to be less effective at motivating users than 10k powers. So introducing something really helpful, such as comment moderation, seems in order.
However, I think comment moderation is mostly a waste of time. I'd much rather let the system handle comments. Unfortunately, there are already so many comments that should have been edits (or even answers) instead that we can't blindly hide/delete old comments. Creating this privilege will mean committing to manual comment moderation for the foreseeable future. Comment moderation is tedious and unrewarding so I'd rather have our top users do something more productive with their time.
I'm suggesting these privileges because comments are one of the few things that elected and appointed diamond moderators can do something about that the rest of the community cannot. Other than flagging, there's no way for regular users to fix problems with the comments on a post. Having a powerful privilege to aspire to provides a better incentive than providing somewhat less-powerful privileges. Anecdotally, people most look forward to the 10k privileges which introduces the final piece of community moderation: post deletion.
These are exceptional users, so they participate (as a group) far more than others. If we look at users with more than 5 reputation, the median first post after getting participate on meta is 215 hours. Compared to their later behavior, trusted users start slow. Notice that they also slow down again after hitting 10k or so. Critically, the slow down after 10k also occurs for 30k users (N=429), so it’s not just an artifact of the cutoff.
There’s circumstantial evidence that increasing the top privilege level increased participation at higher levels. More to the point, the relatively underpowered 15k and 20k levels appear to be less effective at motivating users than 10k powers. So introducing something really helpful, such as comment moderation, seems in order.
However, I think comment moderation is mostly a waste of time. I'd much rather let the system handle comments. Unfortunately, there are already so many comments that should have been edits (or even answers) instead that we can't blindly hide/delete old comments. Creating this privilege will mean committing to manual comment moderation for the foreseeable future. Comment moderation is tedious and unrewarding so I'd rather have our top users do something more productive with their time.
- 80.7k
- 35
- 249
- 350