58

This question holds for the entire network of main and meta sites.


Looking through the activity tab for the Community user, I don't see any edit history for posts that have been randomly bumped (probably because no actual change is made to the post).

Is there any way to see the history of the Community user's bumping activity? (For example, through the API or SEDE.) I'd like to look at this data to see how effective bumping has been in the past, and see if we can come up with suggestions for improving it.

  • 9
    Seems the community user hasn't bumped this one in a while – bobobobo Jun 24 '13 at 20:12
19

As of about a week ago, there's a new PostHistory record added when Community bumps a post:

These are also in the Stack Exchange Data Explorer (PostHistoryType #50), so you can query and analyze the effects of bumping as much as you like (keeping in mind that there's no record prior to May 4th, 2016). Here's an example.

Credit here goes to: Konamiman (implementation), Jarrod Dixon (implementation guidance and moral support), jmac (identification of the problem, proposal for changes to address it, and at least an hour wearyingly trying to convince me that it was a real problem).

  • So... curious... does that mean that the questions I see bumped every week or so by community will collect dozens of these (if it's bumped regularly)? Or will they be collapsed into one with an event counter? – Catija May 10 '16 at 16:45
  • You should never see the same question get bumped more than once every 30 days, @Catija. That said, it is possible for a question to collect a disturbing number of these, at least in theory. – Shog9 May 10 '16 at 16:47
  • Ah. Must just be in my head, then. I feel like I see this cooking question bumped more often than once per month... though someone's upvoted one of the answers, now... so I won't see it any more. – Catija May 10 '16 at 16:55
  • If you go spelunking in the editing history of an unedited question, you should be prepared to be disturbed. – rumtscho May 10 '16 at 16:59
  • This is great, thanks. Is it at all possible to backfill these events into PostHistory for pre-May 4 bumps, or does that data just not exist at all? – senshin May 10 '16 at 21:28
  • It isn't, @senshin. We could guess at a few instances where a post has been bumped and remained completely untouched since, but since that would ignore both cases where the bumping prompted answers/edits and cases where questions were bumped repeatedly, it would probably be worse than useless (any analysis would be skewed if it didn't take this into account). – Shog9 May 10 '16 at 21:30
  • @Shog9 Gotcha. That's a bit of a shame, but it's definitely going to be helpful to have this data going forward. Thanks to whoever implemented this! – senshin May 10 '16 at 21:31
  • Where are my manners? I've added credits, @senshin. – Shog9 May 10 '16 at 21:50
  • @Shog9 collecting bumps is not a theory. Actually it's disturbing (ridiculous), that this particular post, which is obviously "troublesome" if it needs multiple bumps, get tweeted because of the bumps. – yo' Jul 7 '16 at 9:28
  • what's up with that question, @yo'? No one can verify the answer? – Shog9 Jul 12 '16 at 19:50

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .