Should we close questions as "cross-site duplicates"?

Question

The question The largest Saturday number was closed by five users for the following reason (quoting a comment from the first close-voter which was apparently used as a close reason by the other four too):

I’m voting to close this question because it already has an answer at math.stackexchange.com/questions/411897/… (cross-site duplicate)

Should this be a valid close reason? I've never before seen a question (on any SE network site) being closed just because it's previously been posted on another site.

---

To clarify the scope of this meta question:

• This is not about questions that are more maths problems than maths puzzles, and therefore might fit better on Maths SE than here. We can close those using the existing custom close reason. Here I'm asking about puzzles that would normally be considered on-topic here if other sites are disregarded.
• This is not about questions that the same OP has copy-pasted across different sites ("cross-posting"). Those are normally handled either by tailoring the question to be different for different communities, or by closing/migrating one of the questions. Cross-posting is discouraged as a network policy, but here I'm asking about puzzles that just happen to exist on another site already, maybe from years ago.
• Further clarifications can be added if required.

Answer 1

No, this shouldn't be a close reason.

Each individual SE site's scope and policies should exist in a vacuum, separate from other sites. Of course there are some network-wide policies which we have to follow since we're using SE's platform (although even some of those aren't applied for Puzzling specifically), but otherwise we should be able to decide our own rules without caring about anything on Stack Overflow or Mathematics SE or Gardening & Landscaping SE. Just like we don't allow their scope policies to shape ours, we don't have to care at all what's on those sites - we should be able to decide what we do in a way that's independent of whatever they do, so our work can continue even if their sites shut down. Like the thief and cop on the open interval, our universe is restricted to the domain of a single interval site, and nothing outside that exists for us.

So the existence of a question on another SE site from years ago shouldn't mean that we close it here. We should completely ignore other SE sites when deciding what to close or leave open on our site. (This is assuming it's not the same OP cross-posting questions across sites, in which case we're back to network-wide policy.) Aside from this general philosophy of our universe being bounded to a single site, there are also other reasons why it can be better to have a question existing on two sites.

Any two sites have different communities and viewpoints, and maybe would answer the same question in different ways. This isn't the case for the Saturday number question, which is a plain mathematical puzzle that can only be solved in one way. But it might apply to e.g. What is the best method of scrambling a rubik's cube? - maybe Maths SE would answer this from the viewpoint of theoretical randomness and bounds while Puzzling SE would answer it from the viewpoint of what actual puzzlers and puzzling organisations do. Or it might even apply to some plain mathematical puzzles: since they're more about existing mathematical knowledge and we're more about solving stuff ourselves, maybe a Maths SE answer would simply say "this is a case of a known puzzle that's been studied before, here's the name, you can go and read about it" while a Puzzling SE answer would provide a solution from scratch. Both could be useful contributions. I've seen that Maths SE answers sometimes take the form of hints ("try to make this substitution and see what happens") while those would be considered poor partial answers here, so it might be that something previously posted at Maths SE isn't even answered well by Puzzling SE's standards.

Sometimes the older answers might not be very good, and we can do better with a new post. This might be an issue of different sites having different standards, as mentioned in the previous sentence, but it might also simply be that a question has been asked on Maths SE years ago and just not answered very well, or even answered well but not as well as someone is willing to do on Puzzling SE today. I saw an example of this recently with two other SE sites: a question was asked in 2012 on Science Fiction & Fantasy, where it got several answers contributing different bits of information, all now highly voted; the same question was asked in 2021 on Literature, where I provided a single answer collating all the information from the different SFF answers and more. Nothing wrong with the original question, or with information trickling in gradually, but the internet would've been a slightly poorer place if the new question had been closed as a cross-site duplicate rather than being allowed to attract a better answer (an answer which would've languished at the bottom of the pile, or even deleted for providing no new information, if posted on the original question).

Answer 2

Just found this meta post, so just to chip in with my original thinking when writing that close reason:

• if it were any other site on the internet where an earlier version of the puzzle were found without being attributed, it would be closed using the canned response:

  This looks like a puzzle you found elsewhere. For content you did not create yourself, proper attribution is required. If you have permission to repost this, please edit to include (at minimum) where it came from, then vote to reopen. Posts which use someone else's content without attribution are generally deleted.

• there seemed a more relevant option initially, "Duplicate", but on attempting to use that it only lets one select posts from the same Stack Exchange site.

• the most relevant option seemed to be: "This question belongs on another site in the Stack Exchange network", which clearly it does, because it was indeed already on that other site and indeed already had an answer, but the admin overhead of us shuffling the question over there only for them to need another set of votes to mark it as a duplicate seemed too much hassle.

• an earlier comment had already described it as a "cross-site duplicate", which seemed a better fit for the situation than the canned responses.

I note that the post for question itself still doesn't attribute a source, making it falsely appear that the OP created it as an original puzzle, which could in itself be cause for closing and deleting, but I've already used my close vote on that question now, albeit with "the wrong reason".

It could also be possible to edit the question to attribute MSE as the source, but that might not have been the source the OP used anyway.

However, I think the most appropriate response to a similar future situation is:

use the normal canned response for non-attributed content, rather than muddying the issue with a "cross-site duplicate" comment