Timeline for answer to How often are closed questions re-opened? by CRABOLO
Current License: CC BY-SA 4.0
Post Revisions
14 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 18, 2021 at 12:13 | history | edited | CommunityBot |
replaced http://data.stackexchange.com/ with https://data.stackexchange.com/
|
|
| Jul 10, 2015 at 0:27 | comment | added | CRABOLO | @loli in that specific case the OP has enough rep to cast reopen vote which would send to reopen queue. If the OP didn't have enough or someone else didn't have enough rep to reopen they could start a meta post about the specific question or go to a chat room and ask about it and see what others think. | |
| Jul 9, 2015 at 16:56 | comment | added | loli |
Anytime that the OP edits their question of a closed post, it gets sent to the reopen queue But what if the question is good and doesn't need any edit? like stackoverflow.com/questions/4706696/…
|
|
| Jul 29, 2014 at 1:23 | comment | added | Rachel | @VotetoClose That will work for Users because all the user ids are sequential, so its easy to pick out the missing ids to find "deleted" users. But that won't work for Deleted Questions because the Posts table contains both Questions and Answers, and we can't just assume that every missing id number is a Question (although it is safe to assume that every missing Id is a Post, even if we can't tell if it's a Question or an Answer). | |
| Jul 28, 2014 at 16:53 | comment | added | Rachel | I'm really confused about how your missing questions query works. It looks like it takes all questions, sorts them by id, then gets the row number of each record and assigns it to seqid. Then it joins that to the table containing all posts (questions and answers) using rownumber/seqId = posts.id, and counts any record where rownumber/seqId is null as being "deleted". I don't see how that gets you an accurate count of deleted questions. Am I misunderstanding how this works? | |
| Jul 25, 2014 at 13:26 | vote | accept | James Kingsbery | ||
| Jul 25, 2014 at 11:28 | history | edited | CRABOLO | CC BY-SA 3.0 |
added 878 characters in body
|
| Jul 25, 2014 at 2:34 | comment | added | Robert Crovella | +1, Thanks for the interesting data mining you did. | |
| Jul 25, 2014 at 2:09 | comment | added | Ken White | Your phrasing is off. There's nothing in your calculation that gives "the chance of a question being reopened", which implies odds of it happening. There is a 100% chance of a closed question being opened if it isn't a duplicate, it's edited to improve it, and the right people being the first to see it in the reopen queue (or see it as on hold and vote to reopen it on the spot). You're citing odds when you mean "average based on the data at this time". (Yes, I maybe am being a little pedantic, but this is a technical site. :-) ) | |
| Jul 25, 2014 at 0:33 | comment | added | user541686 | Also, it'd be awesome if you could interpolate between the two cases. Rather than just "closed" or "reopened", is there some way to show how many reopen votes they currently have? (full votes would mean reopened) That way we can figure out how likely it is for a question to gather enough votes for a reopen when it has at least some votes. | |
| Jul 25, 2014 at 0:32 | comment | added | user541686 | Kinda funny you did this given your name and all. | |
| Jul 24, 2014 at 23:18 | comment | added | Braiam | Actually the proportion is off, since most close questions are deleted, hence not in the data dump. | |
| Jul 24, 2014 at 22:08 | history | edited | CRABOLO | CC BY-SA 3.0 |
added 140 characters in body
|
| Jul 24, 2014 at 22:03 | history | answered | CRABOLO | CC BY-SA 3.0 |