14
$\begingroup$

I recently came across an answer that was copied & pasted in toto from another thread. I think the answer is high-quality--I had upvoted the previous iteration--but I am uncomfortable with this; I suspect this behavior (unabated) will clutter up the site. I am also in a quandary about whether to upvote an answer twice or leave a worthy answer un-upvoted, which may imply to future readers that the community didn't think much of it.

Certainly there are any number of questions that are fundamentally similar that arise regularly. This issue has been discussed before (How to stem the tide of duplicate Qs) and I continue to hold the opinions I stated in the previous discussion (tl;dr: this is inevitable, & harnessed properly can lead to positive knock-on effects). I don't think the best solution is to simply close and/or delete any question where the same answer could be given.

I left a comment to the answer at issue to the effect that it would be better to leave a comment to the question with a link to the previous answer or a brief answer with such a link. The OP disagreed on the basis that comments and original answers elsewhere could be deleted without notice.

I don't want to debate this with a single individual in the comments to an answer. I think it should be discussed here on meta. What policy should we adopt relative to this type of situation?

Note: I found this meta.SO thread, but did not find it to be very helpful. Here is another related meta.SO thread.

$\endgroup$
9
  • $\begingroup$ It is interesting that both meta.SO discussions seem to be OK w/ this. To show my hand a bit, I have several concerns: 1 there is an issue about how the duplicate should be treated vote-wise; 2 I think this will clutter up the site; 3 I think it will slow or prevent the positive knock-on effects I allude to (eg, many pages w/ the same answer doesn't attract the attention, Google or otherwise, of a single page w/ many links pointing to it & many votes, etc). $\endgroup$ Dec 9, 2013 at 19:21
  • $\begingroup$ In principle this worries me a bit; in practice hardly at all. It is not as if we are short of space. Some of the issues here seem to be about whether someone might get a bit too much reputation undeservedly; sorry, but I think that is taking the whole business too seriously. We should think statistically; all reputations are measured with error. $\endgroup$
    – Nick Cox
    Dec 9, 2013 at 22:33
  • 2
    $\begingroup$ For the record, I would much rather posting a link as a comment if SO respected user-content, e.g. did not delete questions, answers and comments without notice (+ without even given a possibility to retrieve the deleted content): I am totally against this SO destruction spirit but I can't do anything except complaining on meta, which I have already done (1, 2 and 3) to no avail. In the end, I find ctrl+c faster than dealing with all of that. $\endgroup$ Dec 10, 2013 at 0:24
  • 2
    $\begingroup$ @FranckDernoncourt, I understand your frustration & I agree. I think people should be notified, & I think there should be a way to recover your content. I think this is much less likely to be a problem on CV, however. Unless an OP deletes their own question, I think we hardly ever delete anything here. $\endgroup$ Dec 10, 2013 at 1:35
  • 2
    $\begingroup$ Most material on CV (with the exception of certain less-frequented corners of the site like tag wikis) never entirely disappears: even deleted comments are retrievable by moderators. High-rep users can see deleted Q's and A's and (I believe) vote to undelete them. Thus, I have little hesitation to delete or move material that clearly does not belong. Most of this is moving posts from answers into comments. $\endgroup$
    – whuber Mod
    Dec 10, 2013 at 2:31
  • $\begingroup$ You can see previous versions of tag wikis by clicking history. I'm not sure if that's what you are referring to. Even if you are not a moderator, once you reach 10k rep, you can see deleted answers, & edit them, or vote to undelete. You can see questions as well, but only if you can find them; the links to them tend to disappear. It may be possible to find them via the moderator tools page, but it isn't easy to do. $\endgroup$ Dec 10, 2013 at 2:37
  • $\begingroup$ Thanks for the clarifications, gung. I was referring (albeit not very clearly!) to the frequent experience of losing suggested edits to wikis and of losing wikis altogether when tags are merged or made synonyms. $\endgroup$
    – whuber Mod
    Dec 10, 2013 at 14:53
  • 3
    $\begingroup$ @Franck: I've made a mild edit to your first comment with the intent of preserving the message while removing the overly inflammatory language. Apologies for the irony associated with this. Cheers. :-) $\endgroup$
    – cardinal
    Dec 10, 2013 at 17:25
  • $\begingroup$ For future readers: I believe present policy is to close the thread as a duplicate of the original question. Even if the questions are not identical, threads can be duplicates if the same answer addresses both questions. $\endgroup$
    – mkt
    Sep 18, 2019 at 9:58

2 Answers 2

12
$\begingroup$

As a moderator I look upon duplicate answers with prejudice. Although there's nothing morally wrong with reproducing all or part of a previous answer that you authored (wherever it may have appeared) without specific attribution, there's nothing right about it either. If something you posted elsewhere might be useful as an answer, link to it instead and provide a synopsis.

By default readers will understand that what is posted on our site is original unless advised otherwise. Violating that expectation can be perceived as deceitful. Of course non-original materials may be reproduced under fair use guidelines and need to be properly attributed (that is, name the author and identify the publication in which the materials appeared). There is a spectrum of quotation: at one extreme, we ask that links be explained so that answers can stand alone; and at the other extreme, we ask that an answer that otherwise would be a wholesale extract of other work--even your own--be condensed to a brief summary with a link to the original.

Note that deleting a duplicate post destroys no information whatsoever.

Note, too, that there is extremely low risk in offering a link as a comment: nobody (at least nobody acting in good faith) is going to delete such a comment.

$\endgroup$
2
  • 2
    $\begingroup$ +1 I agree completely that duplicate answers should be removed. Where - as a small part of a more extensive answer - a section of another answer is quoted, I think that's fine as long as proper attribution and a link is given. Where it's the same user saying the same thing, they may say nearly the same thing as part of another answer more than once without consciously copying. I have sympathy with that, but if its more than a few lines in common with another answer, it should probably be limited to a brief, explicitly-quoted, extract and a link to the original; an edit may be sufficient then $\endgroup$
    – Glen_b
    Dec 10, 2013 at 2:57
  • 2
    $\begingroup$ This is a very grey area. For example, one active and one less active user contribute the same style of answer to many time series questions, namely that you should think of doing this instead and using our software instead. My view is that their posts are interesting and well-informed, if not always helpful in the way OPs asked, but some answers are close to the same stuff repeated. (I am not veiling this darkly; naturally the people concerned have every right to comment if they wish.) $\endgroup$
    – Nick Cox
    Dec 11, 2013 at 15:35
1
$\begingroup$

I do not believe that duplicate answers should be removed. If the answer is concise and answers the question the first time, it should not be summarized, and instead should be posted in total the second time as well (with credit given to original post). I think of the familiar situation in which you look up a term in the index of the back of a book, only to find "see ...", which leads you to another search term. This would create a needless waste of time considering we have no page limit on the internet. The problem I have with only giving a link is that you must take the time to read the other Q&A to make sure the situation adequately replicates your own and that it isn't just "something similar" to get you started. As a person interested in finding a quick, concise answer to a particular question, having a self-contained answer on the page in which the question was asked is most convenient.

$\endgroup$

You must log in to answer this question.

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