The topic of cross-site duplicates has come up in many Physics.Meta questions so far, for example:

Based on reading those question, I understand that cross-posting is generally discouraged on Physics.SE, though one answer pointed out that "Cross-posting is not a violation of any written rule", and other sites might allow a question from Physics.SE to be cross-posted there.

My question is, what to do when a question from Physics.SE has been asked on a different site in the SE network, and received a good answer there, but still takes up space in the unanswered queue here on Physics.SE.

For example, this question on Physics.SE: Can we invert Density Functional Theory through sufficiently accurate experiment? received very little attention in the last 26 months (0 upvotes, 0 answers), but the same question when cross-posted received an answer within 7 days. Now, do we keep the question on Physics.SE opened, or should I flag the question to bring attention to the diamond moderators if the consensus is to close or delete such questions?

Another example is here: Does the Schrodinger Equation yield a unique wave function and density?, and whereas the cross-posted version of the question got an answer more than 1 year ago, the question on Physics.SE remained unanswered until about 6 months ago. Considering that Physics.SE got an answer to this question about 6 months ago, perhaps both versions of the question will get more answers over time, but those answers will be split between two SE sites rather than all in one place. When someone notices something like this (say the other site got a well-received answer first), would it make sense to flag for a diamond moderator to close/delete the Physics.SE version, or should we keep both questions and their respective answers on each site?


I don't think there's any reason to assume anything has substantially changed from the "we strongly discourage this, but sometimes it's okay".

In general, cross-posting the exact same question without mentioning the cross-post in the question itself is bad because it leads to duplication of effort: People on one of the sites won't see whether the question might already have received answers elsewhere, and might just write the same answer someone else already has written elsewhere. So, at the very least: If you cross-post, mention the other versions of the question directly in each version so users are aware of this.

Furthermore, posting the exact same question verbatim to different communities is rarely the optimal way of phrasing that question for any particular community: Each SE site has idiosyncrasies in what it likes to see in a question, and if you're going to post your question to multiple communities you should tailor it to these communities. I am much more likely to vote to close a careless copy-paste as a cross-post than something where effort on the part of the asker is evident.

As for what to do with cross-posts: Since our stance is "discouraged, but not always forbidden", moderator flags are not an appropriate way to deal with this, since moderators then need to make a judgement whether or not this is a case that should be discouraged or not. It is much preferable to just vote to close the question (with custom reason, e.g. "because it is cross-posted", with a link to one of the meta discussions we've had about that) and let the ordinary community review decide whether this is a case of an unwanted cross-post or not.


I think ACuriousMind's answer largely covers everything. As a small addendum: if a question is on-topic here, has been open for a while (i.e. too old to migrate), and then gets re-posted and answered elsewhere, I think it's entirely appropriate to post a brief answer saying roughly "This question has been answered at [link]" and providing a brief synopsis of the answer. This allows the answer stub to be accepted, thus removing the thread from the Unanswered listings.

  • $\begingroup$ That answer stub should probably be community-wiki in order to not duplicate reputation for the same answer twice? $\endgroup$ Jan 4 at 18:08
  • 3
    $\begingroup$ @user1271772 It's a good idea, but I wouldn't insist on it. Such stub answers are unlikely to get much rep anyway, so I don't think this is much of a substantive issue. To the extent that rep is accrued (an accept plus a few votes) it's a reasonable reward for what's essentially curatorial work (which is work). $\endgroup$ Jan 4 at 18:15

You must log in to answer this question.

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