4
$\begingroup$

There is some healthy overlap between Astronomy SE and some other sites, just for example:

When questions are asked in one SE site that have good answers in a different SE site, good samaritans often call attention to the existing answers elsewhere by adding a comment, like "related in X SE" or "answered in X SE" followed by a link to a question with applicable answers.

But recently two different comments called questions here "duplicates" of questions in other sites, and this word has a very specific and different meaning in SE. A duplicate question (at least currently) must be in the same SE site as the question in question, and duplicate is a specific close reason.

Calling questions on other sites as duplicates may confuse some folks, especially new users, that the question is somehow less than 100% legitimate and answerable here, and attract erroneous close votes or down votes.

Examples:

Space.SE duplicate: (link to Space SE) here

Duplicate of (link to Physics SE) here

Question: Should we in fact avoid calling questions that have answers in other SE sites as "duplicate", which already has a specific and different meaning in SE?

$\endgroup$
6
  • 4
    $\begingroup$ Who is the "we" in this question? Is it you, so you can enforce your views? Is this something you want moderators to do, so they can enforce standards, and maybe delete such comments? Or is it all members of the community? (And good luck with this final meaning.) $\endgroup$ Jul 15 at 11:08
  • $\begingroup$ Of your two examples, the comment that points to a Space.SE Q&A as a supposed duplicate is not a duplicate. The Space.SE question asks what causes the diffraction spikes; neither the question nor answers address removing said spikes. The Astronomy.SE question specifically asks about removing them. This new question would not be a duplicate of the prior question even if that prior question had been asked on Astronomy.SE. Writing something along those lines would have been the appropriate response. $\endgroup$ Jul 15 at 11:49
  • 1
    $\begingroup$ Regarding the latter example, that is a duplicate. One of the best and most prolific members of the Astronomy.SE community (ProfRob) answered the question over on Physics.SE. A copy-and-paste version of his Physics.SE answer would work quite nicely on the Astronomy.SE question, but why bother? $\endgroup$ Jul 15 at 11:51
  • $\begingroup$ @DavidHammen It's great when "... good samaritans often call attention to the existing answers elsewhere by adding a comment, like "related in X SE" or "answered in X SE" followed by a link to a question with applicable answers." We should simply avoid saying those duplicates because the term already has an accepted meaning in Stack Exchange which doesn't apply. $\endgroup$
    – uhoh
    Jul 15 at 19:11
  • $\begingroup$ Arguably, this question is a cross-site duplicate - meta.stackexchange.com/q/4708/236563 $\endgroup$
    – Chenmunka
    Jul 27 at 14:05
  • $\begingroup$ @Chenmunka Per the description there: "which belongs on another site" - no, this is on topic here; "same question is posted in two places at the same time" - also no. $\endgroup$
    – called2voyage Mod
    Jul 27 at 18:59

2 Answers 2

7
$\begingroup$

In short, yes. I would advise against calling these duplicates.

However, I feel given the discussion here, this deserves a little bit more of an answer.

I don't think it's desirable or realistic for moderators or other users to police what people call things, but having meta discussions like this can help guide people's actions.

So what is the appropriate course of action?

If a question is a cross-post...
Flag it for moderator attention. If you cast a close vote, use the custom close reason and put "Cross-posted from [link]". If you believe a question is cross-posted, but can't recall from where, leave a comment but please don't close. It's important to verify that it's really a cross-post before closing.


If a question is not a cross-post, that is, it is not an identically worded question posted by the same user from a different site, then stop and reflect before taking action.

If it's from the same user, but the question has been meaningfully tweaked in a way that elicits information best provided by the expertise on this site...
Leave it alone. Unless the changes don't actually result in materially different answers (e.g. a resident astrophysicist like Rob already answered in depth on Physics in a way that addresses the altered question body for Astronomy), we allow this kind of inquiry here. If it is the case that it's already fully addressed elsewhere, then leave a comment saying so and flag for moderator attention. If you must vote-to-close, please use the "Cross-posted from [link]" format as that helps mods and other users appropriately determine whether it's an unnecessary duplication of effort.

If it's from a different user, but the answer sought is identical to an existing question on another site...
Handle with care. If it's really on topic here, the best course of action is usually to leave it open here, and allow our community to develop our own base of knowledge to answer the question. We can even use references from other SE site answers if applicable. If a Q&A covering this subject on another SE site is already so thorough, that it really can't be improved upon here, then flag for moderator attention. If you must vote-to-close, then use the custom close reason and put something along the lines of "While this is on topic on this site, I believe it would be an unnecessary duplication of effort to address this here." I do not recommend this. Question closures should really only be used for these cases: "the community has determined this topic is inappropriate here", "this question is definitively unanswerable or impossible to answer factually", or "this question could be reopened after some work".

What should really be done? Well, really we should simply be commenting and informing the asker of the Q&A that might address their question. If it does address their question, the asker can delete their question. If it doesn't, they can edit their question to clarify what additional information they're seeking. In extreme cases, I could potentially see a moderator deleting a question for the author. For example, if it looks like they're just trying to ask a question from another site to get rep or to self-answer with a personal "theory". That's why flagging for moderator attention is helpful.

If it's from a different user and the answer sought is not identical to that on another site...
This should not be considered a duplicate or a cross-post at all. That question should be fully welcomed on our site, and I'm not sure why anyone acting in good faith would want to turn it away, unless they misunderstood the question. If you treat a question as if it is identical and someone else in the comments points out that they believe the question is not identical, then you should either wait for the asker to clarify, or in lieu of clarification from the asker, vote to close as unclear.


But what if the question is not on topic here?

Then you shouldn't be closing it as a duplicate at all. You should be closing it as off-topic, with the explicit reason why it is not in scope here. If it is answered elsewhere, you are not obligated to guide the asker to it, but if you want to be helpful you can give them a link in the comments. The only time you should be concerned about an off-topic question being answered elsewhere is when the question is being recommended for migration to that site. In that case, you can flag for moderator attention and inform us of the existing Q&A so that we can take that into consideration when determining whether it's a good candidate for migration. (Sometimes, with the permission of the target site mods, we still migrate these so they can mark them as duplicates once they get there, as having different ways of phrasing a question can be helpful for people searching.)

But what if it's not a good faith question?

Then flag for moderator attention. It's unlikely that one of our other close reasons doesn't apply in these cases, but if they really don't or you don't want to take the time to figure out the best way to close, just pass it off to the mods. Moderators need to be aware of users acting in bad faith, so we can make sure that the behavior is addressed appropriately.


Why is this important?

It's important because overloading terms that have defined meanings on this site obscures communication and makes it harder to ascertain what is going on when there is a disagreement over a question. On Stack Exchange, a duplicate is a question that is on-topic on the site it is asked and has already been asked on that site. Asking a duplicate question is not in and of itself a negative activity. In some cases (but not all), it can be a sign that the asker was unwilling to do even a trivial amount of research before asking. Sometimes, though, it is because the question that was previously asked was hard to find. Questions closed as duplicates remain a part of the site, and they help people locate answers more easily by creating multiple points of entry.

There is no mechanism to close a question as a duplicate of a question on another site, and this is by design. If a question is off-topic, it should be migrated. The migration creates a link that helps direct people to the answer, similar to the duplicate process. If a question is on-topic, the SE philosophy is to leave that question up and allow the site to answer it on their own, in the belief that this gives the opportunity for others to answer who have different expertise.

Simply closing a question answered elsewhere with a custom close reason of "duplicate" obscures the real reason why you believe that specific question should not remain open here.

$\endgroup$
14
  • 1
    $\begingroup$ One more thing I didn't end up addressing in the answer: if you personally disagree with something being in scope here, but there is not a close reason for it and there is not an existing meta discussion about it, you should be bringing that up on meta. Closing a question as a duplicate of a question on another site just because you personally disagree with the community decision to allow that topic here is bad-faith activity. In that case, you are intentionally obscuring your real motivations to bypass the model the community uses to moderate the site. $\endgroup$
    – called2voyage Mod
    Jul 19 at 18:33
  • 3
    $\begingroup$ Pro tip: If you're passionate about this issue, please keep in mind that we strive to create a friendly environment here. If you see someone use "duplicate" contrary to the standard SE sense of the word, take a step back and make sure you really need to engage. Is it just a comment? Has the question received downvotes or close votes that seem unjustified? If not, it's best to leave it alone. $\endgroup$
    – called2voyage Mod
    Jul 19 at 21:27
  • 1
    $\begingroup$ If someone has used "duplicate" as a custom close reason and you don't believe it is a cross-post, consider asking, "Is this off topic here?" or "I don't think this is the same user. Why do you think it's a duplicate?" to start a discussion to clarify intent, rather than "That's not what duplicate means here" which comes across as unfriendly and combative. $\endgroup$
    – called2voyage Mod
    Jul 19 at 21:33
  • $\begingroup$ Here's a very nicely worded almost-example. I prefer "remind" over "engage" and linked to two of these in a short period of time (8 minutes!) by two experienced users. The concern expressed here is that it "may confuse some folks, especially new users that the question is somehow less than 100% legitimate and answerable here, and attract erroneous close votes or down votes." Similar concerns of other users seeing and picking up on the practice expressed here. $\endgroup$
    – uhoh
    Jul 26 at 21:22
  • $\begingroup$ Basically new users pick up a lot of "how SE works" by following the example of more experienced users, so it's good if more experienced users stick to rules and guidelines and accepted norms, newer users can pick some of them up by some combination of emulation and osmosis $\endgroup$
    – uhoh
    Jul 26 at 21:28
  • $\begingroup$ As was pointed out in the comment thread of the aforementioned question, the purpose of the terse comment was to point out that the answers there do provide an important addition to the question. @uhoh might have pointed out their own answer on the subject, I would have preferred it over being told how to walk and talk; the concerns were replied to and the intentions clarified. I don't know that this question has brought us as far; but some of my sentiments seem to be echoed. --- I'll say this here instead of there: 90 views, 5 score. $\endgroup$
    – Rob
    Jul 27 at 11:11
  • $\begingroup$ @Rob I'm not sure which question you're referring to, so I'm having trouble following your comment. $\endgroup$
    – called2voyage Mod
    Jul 27 at 14:31
  • $\begingroup$ @called2voyage the first linked example, 2nd last paragraph of the question. $\endgroup$
    – Rob
    Jul 27 at 15:01
  • 1
    $\begingroup$ @Rob Ok, I understand your comment now, and yes, I agree. It was not a friendly way to approach the topic. Also, per my guidance here in the comments, it really didn't need to be addressed at all (only possible justification would be the single downvote on the question, but it was still positively scored). $\endgroup$
    – called2voyage Mod
    Jul 27 at 17:40
  • $\begingroup$ @called2voyage I can certainly envision other ways to write the comment; it seems I'd just copy/pasted the earlier comment I'd left on another misuse of "duplicate" made 8 minutes earlier. That part is well-taken and taken to heart. But to " it really didn't need to be addressed at all" at this point I still disagree and for the reasons stated in my question here and in my two comments above (and links therein) which go beyond immediate evidence of downvoting. $\endgroup$
    – uhoh
    Jul 27 at 20:38
  • $\begingroup$ @Rob My comment certainly could have been worded differently for sure to avoid the "how to walk and talk" aspect. Got it. To "the purpose of the terse comment was to point out that the answers there do provide an important addition to the question" that's great(!) and I've referred to such things in my question here as "good samaritans often call attention to the existing answers elsewhere". My comment there was to suggest alternative "tersities", like "answered in..." or "related in" so as not to make an OP feel like they've done something wrong (those with more personal experience... $\endgroup$
    – uhoh
    Jul 27 at 20:39
  • $\begingroup$ @Rob ...having questions closed as duplicate react differently to the word than those that haven't) nor new users getting the notion that SE duplicates can be cross-site. As to "I don't know that this question has brought us as far" I do. called2voyage has taken the time and gone to great lengths to address several different aspects of this topic in a thorough and thoughtful way, and answers like this shape site policy for the future. As always, the purpose of SE questions is to provide an opportunity for the writing of good answers, which are of immediate use but also benefit future readers $\endgroup$
    – uhoh
    Jul 27 at 20:40
  • $\begingroup$ @Rob final note; the way I think and express myself doesn't always align with neurotypical folks and while I'm sorely aware of that, the awareness is only the first step in closing the gap. I'm happy to delete the comment; haven't until now because I'd linked to it but now that there are answers here it's last remaining potential raison d'etre has expired. Is it okay with you if I delete it now? $\endgroup$
    – uhoh
    Jul 27 at 21:00
  • $\begingroup$ In response to this comment, the reply was offered in this context, and was not made in response to a specific question, apparently. - You can cleanup, or leave this so others know what has already been covered. $\endgroup$
    – Rob
    Jul 27 at 21:31
-1
$\begingroup$

Technically you are correct. We should avoid calling things duplicates unless they are clear duplicates, and if the duplicate is a carbon-copy of the same question on two different StackExhanges, we should not call them duplicates. We should call those out as cross-posting. Moreover, if the "duplicate" obviously is not a cross-post (e.g., nearly same question was asked years apart on more than one SE), it's not quite a "duplicate".

However, I think you are overreacting here, uhoh. If nearly the same question was asked on multiple SEs, and if the related question has answers that answer the question at hand, that is useful information. If the user who provided that information called it a "duplicate" rather than "related", so what? If someone feels strongly enough to post a custom reason to close regarding the duplication and enough people join in, the question is closed. You apparently do not like custom reasons to close. Other users disagree.

If on the other hand, someone calls out a question as a duplicate of a related question when in fact the new question is not a duplicate, does it matter that the so-called duplicate is not? And does it matter if the non-duplicate is on Astronomy.SE or elsewhere?

You are overreacting.

$\endgroup$
2
  • 1
    $\begingroup$ My reaction is appropriate and carefully calibrated. Stated concern: "Calling questions on other sites as duplicates may confuse some folks, especially new users, that the question is somehow less than 100% legitimate and answerable here, and attract erroneous close votes or down votes." Stated remedy: "...avoid calling questions that have answers in other SE sites as "duplicate", which already has a specific and different meaning in SE" My reaction: post a question in meta citing examples and suggesting a pretty simple remedy, use words based on their accepted meaning. $\endgroup$
    – uhoh
    Jul 14 at 21:37
  • 1
    $\begingroup$ I'm looking forward to answer(s) from folks who have some actual experience having questions downvoted and closed as well. Invariably when people disagree with my statements about question closing and negative commenting under questions from new users, they turn out to be users with very few question posts themselves. The correlation is so high it can't be a coincidence. $\endgroup$
    – uhoh
    Jul 14 at 21:44

You must log in to answer this question.

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