6
$\begingroup$

I occasionally see reasonable questions get deleted by their authors after being closed as duplicates. Here's one recent example:

https://stats.stackexchange.com/questions/588900/does-fixed-effect-and-mixed-effect-model-have-same-prediction

It seems to me that this undermines the value of the close-as-duplicate system; having linked questions makes finding good answers here easier - and probably via google too.

Is this a situation in which voting to undelete the question is acceptable and a good idea? I'm aware of the conditions in which we can, but I've not seen much advice about when we should.

I'm also aware that authors cannot delete their question if it

  • has an answer with upvotes (even if that answer has a net zero or negative score)
  • has an accepted answer
  • has multiple answers (even if there are no upvotes)
  • has an answer with an awarded bounty
  • has at least one other question that is marked as a duplicate of your question

The quoted text above is from How does deleting work? What can cause a post to be deleted, and what does that actually mean? What are the criteria for deletion?

$\endgroup$
5
  • 2
    $\begingroup$ Has the close-as-duplicate system been shown to have value? In my experience, it's bewilderingly and unevenly applied. $\endgroup$
    – dipetkov
    Sep 29, 2022 at 10:44
  • $\begingroup$ @dipetkov That could be worth posting as its own thread! I'm not sure exactly how one would show it has value, to be honest. One goal for the site is to maintain a corpus of high-quality Q&As, and I personally find it tidier to point more and more traffic towards canonical threads rather than have the information diffused across a bunch. That said, one could reasonably argue for the opposite. $\endgroup$
    – mkt
    Sep 29, 2022 at 13:18
  • $\begingroup$ @dipetkov Unevenness is I think unavoidable. A tiny set of people have the rep & time to vote on these, many questions are never seen by them, and there's not universal consensus on what the standards for closing as duplicates are. But overall I think it's better than not having the duplicate system at all. That said, each time I get new 'privileges', I find there's a bunch of obscure and unwritten rules about their use, part of which is because the SE platform has to have some commonality across sites but different communities have different needs. It can be rather opaque as a newcomer. $\endgroup$
    – mkt
    Sep 29, 2022 at 13:24
  • $\begingroup$ If the finding duplicates can work as advertised, it would be a great thing to have. It doesn't seem to be so in practice (personal opinion here). I would prefer, however, that duplicates are not closed but only link to the related posts. And I think it would be helpful for some of older posts to link back to a few of the newer related posts. This could work more as an exchange of ideas, rather than as promotion of the one best (which usually means most highly voted) answer. $\endgroup$
    – dipetkov
    Sep 29, 2022 at 13:33
  • $\begingroup$ @dipetkov It's definitely suboptimal, and part of that is site design stuff that's beyond our control. But if you have ideas, you could consider raising them at meta.stackexchange.com. But if you think we can use them better here given the constraints, open a thread here on CV meta. $\endgroup$
    – mkt
    Sep 29, 2022 at 13:36

2 Answers 2

Reset to default
3
$\begingroup$

There is not enough benefit of keeping such questions to undelete them. Yes, they may make finding relevant questions easier, but on another hand, they add to the overall "noise" of the multiple questions and answers we have. I would say that there is no harm in keeping them, but also no harm if they are deleted. We want to keep the good duplicates, not all the duplicates.

See also

Should duplicates be deleted?

In general, no: most duplicates stay around. Having multiple copies of the same question with different wording is useful as search fodder, because people looking for an answer may use different wording too. On the other hand, duplicates that are word-for-word copies or that are so poorly written that they are not useful may be deleted by users with sufficient privilege.

So again, we want to keep most of the duplicates, but there is no need to keep them all.

$\endgroup$
2
$\begingroup$

I did this on Math.SE yesterday.

My question was (correctly) closed as a duplicate. Because I had made some incorrect assumptions in the question, I decided that it would be best to delete it and keep newcomers from reading the incorrect information and skipping the comment pointing out the mistake (without giving a correction).

I would expect Math.SE to respect my decision to delete, and I expect the same from Cross Validated.

$\endgroup$

You must log in to answer this question.

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