How should we as a site deal with answers which are nothing but merely a copy paste from another article or site?

It's is no on, so I was wondering, what's the status of such post here?


5 Answers 5

The first rule is that you should quote and attribute any direct quotations. There may also be limits on what size of quotation represents fair use. These two issues are important from an ethical and legal perspective. Attribution is also useful to the reader.

Then there is the perspective of what makes a good answer. In general, the voting mechanism sorts out whether an answer is appreciated, although there are limits (e.g., spam, comment answers, not an answer at all, etc.).

That said my general advice on incorporating quotes would be:

  • Put the quote in a quotation block
  • Attribute the quote with a reference, preferably also with a working url
  • Relate the quote to the question
  • If needed, bold aspects of the quote to highlight elements that directly answer the question and edit the quote to make the key points clearer (e.g., put three periods "..." to indicate omissions or use brackets "[]" to make the quote make sense).

Ultimately, I believe it ought to be a balance between quoting (with attribution), paraphrasing and summarising... preferably with a diagram or two and a TL:DR if it is a lengthy answer.

I am not impressed by "wall of text" replies and generally don't upvote them (nor do I necessarily downvote them).


As @JeromyAnglim pointed out, a case that's literally of the sort you've described poses legal problems. I assume Meta Stack Overflow has its own policy about plagiarism, and I hope the moderators here know it and know (better than I would) what to do if they see plagiarism. As for users like me, I'd say it's sort of a matter of personal discretion whether to comment to notify the answer's "author" of one's concern, or whether to just flag it right away for moderator attention (and explain, of course). If one comments but doesn't flag, I'd say there's a little added responsibility to check back and see if anything gets done about it later, but it's the same sort of voluntary responsibility we all have for participating in the site: someone else can do it, but some amount of social loafing is likely, so anyone who notices the problem should try to take responsibility. If the plagiarist doesn't take responsibility, a moderator (or maybe the Stack Exchange lawyers) will ultimately have to deal with it one way or another (unless no one ever cares to sue, which is likely enough in most cases after all, I suppose), but anyone who notices plagiarism should either help the moderators take over right away by flagging, or should check back after commenting and flag it after the plagiarist has had a reasonable amount of time to (not) attribute the source. If someone else had already commented a while ago about plagiarism before I got there, and nothing had been done to fix it since, I'd probably just flag it right away, and maybe downvote too.

Since attribution is necessary, I would also discourage upvoting any answers that are clearly plagiarized, and hesitating to upvote when there's suspicion. Once attribution is in place, it becomes another question entirely. I have no objections to answering entirely with properly attributed quotes if this is the best option, but of course there are times when paraphrasing, summarizing, etc. would be preferable. Problems like that might be solved by commenting or editing, and shouldn't necessitate moderators IMO.

In my experience as a psychology teacher, I've noticed that plagiarism is a big problem in my field. As I said before, I'm sure Meta Stack Overflow has its policy in general, but I'm glad this question came up here separately, as we may need to have our own nuanced policy as well to handle a higher volume of cases in the long run. In my experience, a lot of plagiarism occurs due to mere ignorance of the governing rules (some of which are laws and have teeth), and can be sorted out easily by explaining the rule (or directing the plagiarist to a resource that explains it, which might be useful if someone could recommend a good one in this thread!), and bugging the person to follow it. Some willful plagiarism occurs knowingly for reasons with which I might sympathize, so sometimes I might prefer to see an answer deleted than referred for prosecution...but given the legal risk (for Stack Exchange too), I'd say every instance should be taken seriously and remedied sooner rather than later, by a moderator if not the author.

BTW, I can't quite decide how I feel about inserting the source myself if I happen to know it or can find it, instead of telling the plagiarist to do so. On one hand, it solves the immediately obvious problem, reduces the motivation for a lawsuit, and probably makes a lawsuit very unlikely to succeed; on the other, the evidence of the crime is still there in the edit history, and the person's behavioral pattern (to whatever extent it can be inferred from the case(s) at hand) hasn't been addressed. I'd appreciate any comments on this sub-issue, if it's worth considering.

The likelihood of a lawsuit is very very low –  caseyr547 Jan 17 '14 at 15:07
Citation needed... –  Nick Stauner Jan 17 '14 at 19:39
like youtube i think this is the shield law… –  caseyr547 Jan 17 '14 at 20:15

No, we don't want copy-and-paste answers, especially if they are links to wikipedia. If your answer is just a copy-and-paste of wikipedia or the abstract of an article then just give that link as a comment, not as an answer.

Copy-and-pastes are almost never good answers (with the exception of asking what some historical figure thought on something, or a formal definition). They are usually a lazy way for people to provide answers without tailoring them to the exact question that was asked, or even bothering to understand the source they are quoting. In almost all cases, it is preferable to summarize the source (as long as your do it faithfully, an attribute) and tailor the answer to the specific question instead of just quoting a single person.

In the cognitive sciences, it is also easy to get source on either side of a question, so a good answers involves multiple sources and a synthesis of several references.

Here are two related discussions:

"Synthesis of several references" is exactly what I wanted to say, but put better than I would have done :-) –  Jonathan Deamer Jan 19 '14 at 12:11
FWIW, Wikipedia sometimes gives perfectly good summary answers that can't be improved by the answerer, and SE policy is to provide answers that don't "just give that link." I don't see any problem with doing this in a comment as you've suggested, so long as someone else eventually answers the question independently...but if no one does this, an answer would be better than a comment for the sake of this site's stats (see, if the question deserves it. A question that can be best answered by Wikipedia doesn't show a lot of research effort though... –  Nick Stauner Jan 21 '14 at 0:54

We do want copy and paste answers because expertise and time are used to isolate a relevant reference even in the age of Google.

I disagree. Finding a wikipedia link should a comment, an "answer" that just copy-and-pastes from one or two wikipedia links is not a good answer. –  Artem Kaznatcheev Jan 18 '14 at 22:32
@ArtemKaznatcheev your welcome to disagree but some questions don't require a dissertation and many of "answers" i've seen are tldnr –  caseyr547 Jan 18 '14 at 22:38
pasting long quotes from wikipedia that are marginally related is the most "too long" version that I've seen. It is almost always better to summarize the relevant parts and tailor them to the question than to copy-and-paste. If a question can be completely answered by a wikipedia article, then the question should be closed as lacking initial research (since we are not an alternative to wikipedia) and the link left as a comment. We don't want to promote poor questions or weak answers. –  Artem Kaznatcheev Jan 18 '14 at 22:43
@ArtemKaznatcheev idk i thought tldnr about your answer here… –  caseyr547 Jan 18 '14 at 22:54
Your judgment of "tldnr" doesn't make an answer useless to others. If an answer is too long for you to read, it either means it's off-topic, or you don't care enough. If it's off-topic, criticize it for that reason, if you wish. Otherwise, consider whether anyone else might care enough to read a longer answer than you. I suspect the person asking the question will usually care a lot more than you. In cases like this, especially cases where all answer content is relevant, saying that the answer is too long for you to read is counterproductive, as well as marginally offensive. –  Nick Stauner Jan 21 '14 at 1:06
@NickStauner meh i never said it was useless. its just invalid. the problem is many of the tldnr don't actually even answer the question. people get lost listening to themselves talk or type. artem's example i point out above clearly answers the question with no. then goes into great detail explaining something the user didn't even ask about. if you can't comfortably define intelligence you shouldn't answer questions about intelligence. moreover links to your own blog posts are not valid sources of scientific information. –  caseyr547 Jan 21 '14 at 1:20
Regardless of the reasons you thought "tldnr," it is useless as criticism. TLNR is ambiguous, dismissive, and often irrelevant. It does not convey your reasons. What you just wrote to justify it is what you should've said the first time instead. Everyone is already trying to balance thoroughness with terseness. Readers can judge for themselves when to skim if a writer errs on the side of thoroughness according to their own values. If you don't care enough to read something, why comment on it at all? Judging by your explanation, you still read it. Did thinking TLNR really even matter to you? –  Nick Stauner Jan 21 '14 at 4:52
@NickStauner it would be nice if readers skimmed in an accurate fashion but comprehension doesn't drop it plummets especially with higher grade equivalent material so your still conveying the wrong message. –  caseyr547 Jan 21 '14 at 5:07
such is the reason for abstracts –  caseyr547 Jan 21 '14 at 5:23
So why don't you tell readers not to skim instead? If you're worried about an answer being too ambiguous, you are not helping by making a comment that only discourages long answers. Making an answer shorter means taking details or related information out; how do you think that affects comprehension of the intended message? "tldnr" is a great example of a message that's too short to make any useful point clearly. –  Nick Stauner Jan 21 '14 at 5:52
@NickStauner people are not capable of following such orders. details can be omitted and replaced with hyperlinks for further study and citations which can be delimited for the reader to skip. tldnr begins a discussion about what can be removed or where it got off topic. –  caseyr547 Jan 21 '14 at 6:01
Yeah, that first question was rhetorical. Sorry that was unclear...I blame the character limits on comments! ;-) I agree that those options you list would be useful sometimes, and wouldn't discourage you from offering those suggestions. If you want to start a discussion about such suggestions, I think you'll get a much better conversation by giving your specific suggestions about specific sections that are excessively detailed or otherwise off-topic right away. "tldnr" is just so blunt and indiscriminate by itself; I can't imagine it starting any conversations besides the one we're having now. –  Nick Stauner Jan 21 '14 at 6:48
@NickStauner yes i do need to work on being less blunt :) –  caseyr547 Jan 21 '14 at 6:51
You're not alone there! The internet makes us all sound more blunt than we mean to be, I think. I try to resist it too, but that feeds into the same conflict between "too blunt; didn't know how to react" and "tldnr"...I can't seem to avoid erring on both sides at least once in a while! –  Nick Stauner Jan 21 '14 at 7:18

You must log in to answer this question.

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