« first day (1470 days earlier)      last day (51 days later) » 

6:20 AM
@SonictheAnonymousHedgehog here you go: data.stackexchange.com/stackoverflow/query/1624423
 
7:12 AM
@rene Not if the target doesn't have an answer, but if the duplicate doesn't have an answer. That's when the user is redirected if there's only one target.
 
7:32 AM
@SonictheAnonymousHedgehog ok, edited.
left the other option in as well. That might also have value, me thinks.
you can fork and then remove the T-SQL comment --
 
8:09 AM
@rene Thanks, that makes a great starting point. How would I now filter this to only include those posts that were edited by a human user (positive user ID, or null user ID that isn't named "URL Rewriter Bot") after the ClosedDate?
I think this can be simplified by looking at the last active date. Automatic edits don't update the last active date (which is why those edits don't bump posts).
 
on the closed post, right?
 
also done
 
Never mind, scratch my last statement. I'm finding a few examples where the last active date was after the closure date but no human edit was made later (I don't know why). It looks like one must check the user IDs/names as per my message before that.
 
maybe an answer was active?
Can you hand check that example?
 
8:20 AM
@rene Example post. Don't have 10k to confirm if a deleted answer was changed; question history shows nothing for the active date in May 2012.
 
that one is merged. it doesn't have answers
okay, let me check for a moment
 
Found out why. Seems like this answer was updated in May 2012, then was moved to the target in 2015, which is why the last active date for the merge stub is the former.
 
right. Let me do the edit check in posthistory as you suggested at first
 
But yeah, it seems we can't use last active date since it also includes answer updates. I totally forgot about deleted answers and answers that were merged into other questions. My first solution would only look at the question, which is what I was looking for.
 
8:36 AM
@SonictheAnonymousHedgehog a new version, I think the link is stable, right: data.stackexchange.com/stackoverflow/query/1624423
 
@rene Seems buggy, as it's including negative user IDs (and so has the same result as the initial version of your prior query that did no such exclusion).
It's taking into account the automatic insert duplicate link edits immediately after closure, it seems.
 
 
1 hour later…
10:28 AM
@SonictheAnonymousHedgehog slightly different approach now, much smaller set.
 
 
7 hours later…
5:52 PM
@rene Thanks! Seems to work quite nicely. Very few older unanswered duplicates that were later edited, it seems...
Just a question: how does it filter out Community user edits?
Also, this question was in the query results. If you don't mind, why was it just deleted this week?
 
 
4 hours later…
9:34 PM
@SonictheAnonymousHedgehog roomba
 
9:47 PM
@rene Criterion? RemoveDeadQuestions?
 

« first day (1470 days earlier)      last day (51 days later) »