Copying, Linking, Attributions, and Plagiarism

I have two related questions that I would like to see some discussion on:

  1. Do we need to improve our Help Center’s text to remind people that they have to cite their sources by name?

    UPDATE: I’m marking this first question completed as of 2014-07-08 04:38:38 ᴜᴛᴄ.

  2. What should we do about postings that lack proper attribution?

    • Leave a comment.
    • Downvote.
    • Edit in the attribution if known.
    • Delete the posting.
    • Flag the posting. (and if so, as what?)
    • Add a blue moderator note.
    • Something else.

Edit: Apparently, ours is not the only site with these problems. Most of the issue mentioned there apply here, notably including this quote from that posting of Shoggoth’s:

Do not tolerate answers consisting primarily of text copied from other sources

[. . .]

We require that answers consist primarily of the words of their author, and that all quotes be clearly marked as such and attributed to their respective authors.

A bare link next to verbatim, copied-in text is not “attributed”, and answers that are just text copied from elsewhere provide no words from that author. If you read his posting referenced above, there are specific steps given as guidance.


Background

In our Help Center’s section on “How to reference material written by others”, it reads in part:

When you find a useful resource that can help answer a question (from another site or in an answer on English Language & Usage Stack Exchange) make sure you do all of the following:

  • Provide a link to the original page or answer
  • Quote only the relevant portion
  • Provide the name of the original author

Example:

According to Ernest Hemingway - Biographical on Nobelprize.org, Hemingway saw combat when he was a teenager. It says:

After the United States entered the First World War, he joined a volunteer ambulance unit in the Italian army. Serving at the front, he was wounded, was decorated by the Italian Government, and spent considerable time in hospitals ....

[other sources, quotes, explanations, etc. necessary to complete the answer]

However, this is not happening. I believe that it should be happening, and I would like for the community to decide how to approach missing attributions.

This omission is especially noticeable in answers to questions tagged , where the answerer will copy in a dictionary definition without saying where they got the definition from. However, it occurs throughout our site.

Sometimes a link is given, and sometimes it is not. Now in some cases, one cannot provide a link because the source is from a printed book. In those cases, posters are a bit better with supplying the name of the work they are quoting.

However, in the case of links, posters are very bad at this. In some cases, it’s so bad that their would-be answer is nothing but a word that’s hyperlinked to some online resource. Link-only answers aren’t real answers.

But even when there is more, actual text copied out, a link by itself is not an attribution. It does not include in plain text the name of the work linked to. This is burdensome; how can one judge the authority of the cited source if there is no source given for the citation? There is a world of difference between:

  1. Citing formally curated resources like the OED, the American Heritage Dictionary, or even CGEL.
  2. Citing crowd-sourced resources like Wikipedia, Wiktionary, the Free Dictionary, Etymonline, or Urban Dictionary.
  3. Citing John Q. Public’s random private web page.

Even within each of those three types, there is obviously a hierarchy of “trustworthiness”, but if the source is not named, readers of the posting will not be aware of which one it is.

I would like to see the source named so that we can tell how good of a source it is. I believe that all citations, link or no link, need to provide the actual name of where the text has been copied from. That is what our Help Center says, but people are not doing it.

You should not have to punch through a link to find out how trustworthy the citation is, or even where it has come from. Telling people to “hover” won’t work not just because it is an undue burden, but also because it does not work on the mobile interface to SE.


The Good, the Bad, and the Ugly: Examples Galore

Edit: Some of the examples listed below as being in one of three buckets seem to have gotten unwittingly tossed into the wrong bucket. In particular, some posts were erroneously tossed in the “Ugly” bucket and do not deserve to be there. My apologies.

Here are examples of good, bad, and completely absent attributions:

share
    
I like "Leave a comment." though that doesn't seem forceful enough. –  Mitch Jul 7 at 21:35
3  
When does a link count as a citation that must be attributed in text? I frequently supplement my answers with inline links that provide additional detail on words or concepts that I've introduced, but since I'm answering on my own authority as a Man Of Letters, I don't see any need to say "this is a link to Wikipedia, by the way" if it doesn't flow with what I'm writing. (Example.) –  phenry Jul 7 at 21:39
1  
@phenry You’ve asked a fine question (and I hope that wasn’t one of my unattributed link examples; if so, I’ll remove it). My take on your example is that that isn’t really copied-in text, so I don’t think it needs to be inline-attributed the way the stuff in the Help Center talks about. It’s merely links to things supporting your argument. I’m talking about actual text that is copied in verbatim, not original work with links for support. –  tchrist Jul 7 at 21:47
1  
You'd be better taking some of this to proper SE meta. When I add a reference to any URL the footnote is already there with an [n] next to the text that will be used as the hyper-link. When I post my answer the site automatically removes the footnote. –  Frank Jul 8 at 9:00
1  
After reading your dialog with @Erik Kowal, I feel the need to clarify my upvote of the question. As my answer states, I agree with much of what you said, but disagree with some. I also welcome a discussion of the points of view about this topic. My vote is not intended to help someone win the debate. –  bib Jul 8 at 12:02
    
@Frank You have to write a footnote explicitly if you want one. Markdown does not provide footnote functionality by itself. –  Andrew Leach Jul 8 at 17:42
1  
@AndrewLeach I realise that, but my point is the actual URL is there in the post, it could be shown as a footnote rather than hidden in a hyper-link. That would avoid the issue about not knowing where a link might take you on a small device that doesn't support press and hold for link identification. –  Frank Jul 8 at 17:53
    
@Frank Even if the URL were visible, that is not the same thing as naming the provider. A URL is merely an address, whereas the provider is the name of the organization or individual. Reasonable abbreviations can be made, such as OED or COCA, but as you see, these are still different things from URLs. –  tchrist Jul 8 at 20:27
9  
You seem to use the term plagiarism too broadly. You should take more care, given that it is extremely pejorative. It's obvious to any reasonable reader that quotation marks/markup indicates thoughts other than one's own, yet there are many such examples in your "plagiarism" category. –  j.i.h. Jul 9 at 16:26
1  
@j.i.h. Are you sure? “. . . you need to say where you got quoted material from; otherwise, this amounts to plagiarism.” –  tchrist Jul 12 at 2:48
5  
Frankly, I have no interest in doing anything more than linking to a dictionary or Wikipedia entry when I quote them. I think anything more than that is an absurd waste of space and time and a complete misunderstanding of how the internet works. –  MrHen Jul 14 at 18:47
1  
@MrHen But as Andrew observes, links are not always preserved when the content is reused via various SE-sanctioned APIs — at which point it becomes blindly copied text from an unacknowledged source. The courtesy of a plaintext attribution takes no time worth mentioning. –  tchrist Jul 14 at 18:52
3  
@tchrist: And, for dictionaries and Wikipedia, I am perfectly okay with that. The only reason I even quote an actual dictionary is to give people a place to go look up related information. I could "write" my own dictionary entries but why bother when there are so many existing online dictionaries? I would rather not quote any dictionary than bother with figuring out a proper citation for the various free, online dictionaries. I see little benefit to it. –  MrHen Jul 14 at 19:04
    
@MrHen Well just do as you think best then; I trust you. –  tchrist Jul 14 at 19:09
1  
@AndrewLeach and I am saying to accuse someone of plagiarism is a very grave accusation especially when it's evident it is not true. How difficult is it to click on a link to a dictionary anyway? We're on the Internet! The following is an example of how a perfectly-formatted answer could not possibly mislead a reader into thinking the poster is writing in his own words. How can that be defined as being "ugly"? –  Mari-Lou A Aug 5 at 7:18

9 Answers 9

up vote 8 down vote accepted

The help text at How to reference material written by others is common to all sites. Here’s Math.SE’s version for comparison — it still has the Hemingway quote.

That means that it’s a site-wide norm and expected of all posters on every Stack Exchange site.

It’s certainly no more difficult to reference a dictionary with its name than it is to create a link on the headword itself: you still have to create a link — just create the link on the text “[ODO]” or whatever instead of the headword.

It’s a service to those who come after you: it immediately identifies the source of the quotation and the reader may or may not feel the need to click through to check it. They certainly don't have to do that (or hover, if that’s available) to find out where the text came from.

It’s good manners towards the source of the quote. They hold the copyright, and while Fair Use may allow a quotation for the purpose of reference or argument, the person or organisation who did the work to provide the material should merit an explicit referencing citation.

It creates a more scholarly look and feel to answers to have even a minimal citation. For an online link a full MLA citation isn’t necessary, but a nod in that direction can’t be bad, especially since it’s not difficult.

A reason that Stack Exchange likes the actual text identifying the origin to appear in the reference (and thus why the How to reference pages mandate it) is that the content of SE is made available via APIs for display elsewhere. Links may not survive. Including the citation in plain text ensures that it is also included wherever SE content is shown elsewhere. This helps increase the value of the content, and of the site as a reliable resource; and SE itself can’t be accused of plagiarism.

I certainly don’t recommend going through and altering thousands of answers to make them match SE policy. But it’s not unreasonable to expect future answers to follow that policy especially since it isn’t particularly onerous. If you can edit in a citation on a post without one when you are tidying things up, so much the better.

This post used to mention “unattributed material may be deleted”, which — while not inaccurate — isn’t the whole story and is open to misinterpretation. Plagiarised material may be deleted summarily without warning. If you mark a quotation as a quotation, then it’s obviously not plagiarised. In this case, it’s unlikely to be deleted without warning. However, if a citation isn’t added after a reminder, or an edit to add one is rolled back, then the quote or the post containing it might be subject to deletion. Each case will be dealt with individually.

Note that a simple link out to another site isn’t normally quoting external material. It's normally original text containing a link. This sort of link isn’t covered by the quotation policy at all.

There are some worked examples in another answer.

share
3  
I don't disagree with the principles behind any aspect of the policy stated here. This whole discussion has been about relative burden, and a minor burden at that, not about best practices. But the format now cited in the Hemingway quote, as illustrative of the site policy, didn't exist until yesterday. Before that, the offered epitome was a link without a textual reference to the source. If the expanded reference approach is our new policy, I am happy to conform. –  bib Jul 9 at 0:15
2  
Is the policy changed based on this discussion ? Hemingway example is updated as I can see. Who decided to update that? –  ermanen Jul 9 at 0:51
1  
I have no idea whether this discussion has influenced SE policy or not. Any change is surely likely to have been in the works for the customary six to eight weeks, so if it isn't pure coincidence then I suspect the most that can have happened is that the discussion has accelerated publication of the SE-wide policy. But if you take the first two paragraphs of this answer out, the remainder still stands. –  Andrew Leach Jul 10 at 6:40
1  
Unlike @bib, I'm actually not happy to conform to the expanded citation policy. Adding a whole bunch of text is very distracting and in my view decreases the value of the answer, because it is harder to read. It also discourages people from answering, because it is harder to write. Many people seemed happy with the hover-or-follow-the-link style of attribution, including several high-rep, well-respected regulars here. I will comply with the policy, but definitely not happily. –  John Y Aug 26 at 2:40
    
After adding "proper" attribution to a useful answer, the author of the answer understandably took offence and deleted the answer in response to perceived false accusations of plagiarism. Congratulations on implementing a rule that is detrimental to the site. –  oerkelens Sep 4 at 7:09
    
@oerkelens Which answer on which question, please? Some unnecessary comments and edits have been noticed, so it would be helpful to have full details to pass to TPTB if the network policy is detrimental. –  Andrew Leach Sep 4 at 7:20
    
@AndrewLeach: This answer - link from ELL. The comment from the poster is here - link from ELL meta. The original answer was clearly contrary to new policy by the letter of the policy. –  oerkelens Sep 4 at 7:27
    
@oerkelens Thanks. We'll take a look at a practical outworking of how the policy is being implemented within the community. That question doesn't appear to have been migrated from ELU, so we'll need to work with ELL mods. –  Andrew Leach Sep 4 at 7:49
    
@AndrewLeach: I hope that working with ELL mods is not too scary :) I try not to see ELU and ELL as two entirely different entities, and I hope policies are more or less the same (it would be very confusing for those of us who are active on both otherwise :) ) –  oerkelens Sep 4 at 8:31

Just to add to the chorus: if you quote a dictionary, it is perfectly sufficient to link to the definition - the link itself becomes the author citation. Explicitly citing the dictionary's name would imply that that particular dictionary is somehow to be preferred for this definition, which 99% of the time is not the case. (I have a list of dictionaries bookmarked, and when I need "a definition, any definition", I will randomly choose one of those bookmarks.)

If your answer involves comparative definitions ("dictionary A says [x] but dictionary B says [y]"), then naturally, the names of the dictionaries must be part of the answer. But otherwise, adding the dictionary's name would be worse than unnecessary: it would be possibly misleading.

I do agree that answers that are 99% quoted, or are quoted with no attribution, need to be addressed somehow: fixed if possible, downvoted into oblivion otherwise. But please, don't let's start offending good contributors with counterproductive nitpicking.

Edit This is becoming ridiculous, folks. A LINK IS AN ATTRIBUTION!!!!!!!!! If this were a print medium, then forcing people to add awkward, possibly misleading verbiage about the source of a quote would be understandable, but last I checked, this is the effing INTERNET.

share
2  
So, do you feel that the Help Center text regarding this is wrong? –  tchrist Jul 8 at 16:15
1  
It is interesting to note that the updated Help Center text indeed squarely contradicts this answer. I very much agree with this answer, though. Somehow it seems we went to a black/white situation, all or nothing. Because some people would not include any reference to their source, we now have to overstate our sources. –  oerkelens Sep 3 at 7:11
2  
I cannot for the love of me see what is wrong with "one word you could use is obligatory". Instead I have to contort my phrase into "you could use the word obligatory (see this link to merriam-webster for a definition)" –  oerkelens Sep 3 at 7:12
1  
Wish I could +1 again for the addition of the last paragraph! –  oerkelens Sep 3 at 18:00
1  
@oerkelens Part of the issue with "one word you could use is obligatory" is that it doesn't explain what that word means or why it fits the OP's request. As an added point, I linked to related word instead of the word I wrote. Especially for single-word-requests, embedded attribution encourage sloppy answering. Cf."Merriam-Webster defines obligatory as binding in law or conscience. I feel obligatory is a good fit because it carries the connotation of a burden of conscience." –  KitFox Sep 3 at 19:38
    
@KitFox: I never said that a link alone is enough as an answer. Of course I expect an actual quote from the dictionary, and/or further explanation if that benefits the answer. My answer here would be illegal under the new rules because I did not write "This looks like (a variation on) what is described as negative raising in this answer on ELU". Does "what is described as, "this answer on ELU" really add so much to my answer that without it, I am plagiarizing and it deserves deletion? –  oerkelens Sep 3 at 19:56
    
This answer of mine is a better example. I include a link when I mention omilophobia, but because I do not say where the link points to before giving the (short, but verbatim) description (fear of sermons) that is mentioned in the link, I am plagiarizing. If I would not have taken the effort of providing the link, which makes it easier for the reader to follow up on what I write, my answer would pass muster under the new rules. As is stands, it deserves deletion? –  oerkelens Sep 3 at 20:01
    
@oerkelens If I were inclined at this moment, I would edit your answer to include the source (Defintion-Of.com, which I'd never heard of before and which looks extremely dubious). You don't actually copy the given definitions verbatim, but if you did and refused to attribute it, then that part of your answer might be deleted. Or we could just edit in the source and not worry about it. If you hadn't included the reference, then your claim would be unsubstantiated and that's a whole nother can of worms. –  KitFox Sep 3 at 20:08
    
The word I used is indeed not very commonly used, and I just wanted to illustrate that it actually is not a completely made-up word. I as good as copied the whole definition. The site itself asks me to link to the word definition in exactly the way I did it. But SE has decided that even if I comply with what the author wants, I am not allowed to follow their wish. Since internet seems to have stopped at the borders of SE I will indeed edit my answer to comply. –  oerkelens Sep 3 at 20:28
    
After spending more time editing that answer than I think I ever put into writing it, I can only agree more with the remark that "This is becoming ridiculous, folks. A LINK IS AN ATTRIBUTION!" (quote from answer by user Marthaª to question number 4973 on meta.english.stackexchange, posted originally Jul 8 at 1:42). Or, in the old style: this answer –  oerkelens Sep 3 at 20:46
    
@oerkelens From your edits, I must conclude that you really don't understand what is expected. If it won't provoke you further, I would be happy to edit your post to demonstrate how the policy is expected to be applied. –  KitFox Sep 4 at 0:17
    
@Kitfox Feel free to edit it. I do indeed not understand what is possibly wrong with the attributions as they are in that post at the moment. Then again, I do also not understand what was wrong with them in the first place. –  oerkelens Sep 4 at 6:52
    
On ELL, one user has deleted their answer after feeling unjustly accused of plagiarism. It is sad, but I fully understand their reaction. Even though sources were clearly mentioned in a way nobody would misunderstand, and absolutely no credit was taken for someone else's work, even though the author used the standard SE style to reference external sources, the user did not "properly attribute" their sources under the new regime. I had some hope that I might be the only one feeling less inclined to contribute, but it seems this new regime drives more people away. –  oerkelens Sep 4 at 7:13
    
@oerkelens slightly OT but can you tell me how to add those footnote links. They make a post with a lot of attribution info much tidier. –  Frank Sep 4 at 7:19
1  
@Frank: I literally manually typed <sup>**[1]**</sup> in-line and copied it at the bottom. They do make for a nice and tidy post, and a lot of work (keeping your numbering straight) if you have a lot of links in your post. I am too lazy to make this a standard habit, probably a default disclaimer with plain text references to common resources is much handier. –  oerkelens Sep 4 at 7:23

I want to disagree with you strenuously. But I can't. That doesn't mean I can't disagree with you mildly.

I wholly agree with your discussion about The Ugly. We should strongly discourage quotes, including definitional quotes, that do not give the reader the ability to check out the source. A quote that provides neither the name of the source nor a link to that source, at the very least least, borders on plagiarism.

I also agree that an answer that consists of little more than a linked word is also substantially less than helpful, and outside the realm of how we provide answers. It fails because the information that is most important to the reader, the actual definition, is not on this site.

I am troubled by the view that an answer, often a single word, accompanied by a written definition (and sometimes even an example of use), coupled with a link to an established online dictionary or other standard online reference, is an unacceptable response. I am flattered that the first example of a Good answer is one in which I name the source, provide a link, and quote the applicable definition. I am chagrined that this answer was edited by me to read this way as a sarcastic riposte to @tchrist 's chiding.

I get the problem. Someone can post a quote and a link to a site of limited or even spurious repute, and the unwary may simply take the data on our site without vetting the underlying source.

And yet, the most critical information is often the surfacing of the word. The provided definition may clarify and confirm that the answer suits. In many cases, whether the particular definition is from ODO or American Heritage (that sounds so much better than yahoo) is often (but not always) immaterial. That some authority lists this definition as fitting is often enough.

For the reader who is trying to determine the trustworthiness of the source, the link is a better guide than a mere reference to the name of the source. The link can provide context (alternative definitions, examples, etymology) that the textual cite cannot.

Ironically, the section @tchrist cites in this site's Help Section reads

According to this biography, Hemingway saw combat when he was a teenager. It says:

After the United States entered the First World War, he joined a volunteer ambulance unit in the Italian army. Serving at the front, he was wounded, was decorated by the Italian Government, and spent considerable time in hospitals .... [other sources, quotes, explanations, etc. necessary to complete the answer]

[Supplemental note: The above section and the reference to it in the above question have been modified since this answer was written. I think this actually supports my point as noted in comments below.]

The name of the source is not given, just a link to it.

The difference between a fairly complete answer that provides a word, or term, coupled with a definition and a link to the source, and that same answer with the name of the source added seems to parallel the difference between footnotes and end notes. Footnotes are clearly more informative during the process of reading. You know exactly who said what. But they are often distracting (have you ever read a law journal? Save me!). End notes provide the same information, but some of the material is not immediately present. However it is readily accessible if you need it.

I can't disagree that a simple textual cite to the name of the source (along with other critical elements) is a bad idea. I'm not sure that it should be a rule.

share
1  
Unacceptable would in some of those cases be putting it too strongly. Incomplete or suboptimal might be a better fit. Oh, and I can’t believe yours came up first: I honestly randomized the lists! –  tchrist Jul 7 at 23:09
    
@tchrist Can I please have a cite (linked and textual) on the randomization process you used [and the statistical support for the degree of randomization achieved]. –  bib Jul 7 at 23:33
    
Why certainly! Enjoy: # randline - tchrist@perl.com use strict; use open qw(:std :utf8); my @addr = (0); my $i = 0; my $file = shift; my $temp = 0; if (!$file) { $temp = 1; $file = "/tmp/randline.$$"; system "cat > $file"; die "bad cat" if $?; } unless ($file and -f $file and -T $file) { die "usage: $0 file\n"; } open(F, "< $file\0") || die "can't open $file: $!"; unlink $file if $temp; pop @addr; srand(time() ^ ($$ + ($$ << 15))); my @lines = <F>; while (@lines) { print splice @lines, rand(@lines), 1; } –  tchrist Jul 7 at 23:44
2  
@tchrist Well thank goodness. I was afraid I was going to have to withdraw my upvote, flag the question, downvote and seek closure. And that's exactly how I would have done it (if I could code and understood statistics). –  bib Jul 7 at 23:49
1  
Ugh. See, I should've just stuck with Lewis Carroll instead of gettin' all fancy. Fixed. –  Shog9 Jul 8 at 4:39
    
@Shog9 Thanks. Updated question to include new text. –  tchrist Jul 8 at 15:48
    
@tchrist But I think that your original comfort with the old example, the one lacking a textual author cite, supports the view that it is not always necessary. –  bib Jul 8 at 16:15
    
@tchrist When you cited it, there was no textual author reference, neither in the Help Section, nor in your quote. There was a link. I suspect you read it at least twice, once on help and once in your post. Apparently it did not jar you as being improper since you cited it as a guide for what is proper. Again, this is not to say that adding the textual source is not a good idea, just that it may not b e essential when linking to readily available institutional sources. I do agree that a textual reference is essential when quoting individuals or less standard sources. –  bib Jul 8 at 16:25
    
@tchrist What meant was established reference works, such as the range of dictionaries at onelook.com or long standing sites, such as nobelprize.org. –  bib Jul 8 at 17:15
1  
I don’t understand why you believe that “established reference works or sites of long standing” should somehow be exempted from the requirement of proper attribution. I see nothing in the SE Policy on References to support this notion — and everything to the contrary. Even the OED expects attribution. Furthermore, if what you are claiming were in fact true, then we would be stuck with everyone making their own personal decisions about what is and is not in that exempt category of recognition or longstandingness — which I’m sure you recognize is hopelessly subjective and therefore unworkable. –  tchrist Jul 8 at 17:22
2  
@tchrist That is because we disagree on what is proper. I think a link, without text, is proper for many references. I agree that sites that do not rise the level of reference works (yes, that's vague; use good judgment) need to be made clear. But in any event, a simple click through will reveal the nature of the source. –  bib Jul 8 at 17:52
    
Interestingly, both the quote from the Help Section in the answer and the Lewis Carroll link in the comment are now forbidden as per the updated help section. Is that really what was intended? –  oerkelens Sep 3 at 7:14

Are you sure that you are not misinterpreting this rule? (for the part about giving the exact name of the link as a text)

I started seeing your comments about this in my answers. (I think you should have waited for our opinions first.)

For example: Adjective to describe someone who is knowledgeable, resolute, and calm

Why would I write the exact name of the link as a text in my answer? I'm copying the relevant part and attributing to the link. This covers the first two rules in Help Center:

  • Provide a link to the original page or answer
  • Quote only the relevant portion

And the third rule does not apply: "Provide the name of the original author"

Should I provide the name of the author of the online dictionary here then? I do not think so.

I provide the name of the authors only when I give citations from books. Even the example in Help Center looks like my answer.

[Edit: Hemingway example in Help Center is updated after this answer]

I see a lot of answers that attributes to the link but that does not give the exact name of the link as a text. Even moderators have these kind of answers.

Why would it be a problem now after all this time?

Note: You are right about the answers that copy from a source but that does not provide any link or source.

Note2: If there is a really big problem about this, more details can be added to the instructions in Help Center.

share
    
Yes, I specifically do think you should indeed provide the name of the online dictionary in plain text. Otherwise people won’t know how to judge it. Some are from random web pages, some are from Urban Dictionary, and some are from say Oxford. I think knowing which is which is important; don’t you? It improves your answer, especially when you are citing a source more reputable than my own web page or whatnot. :) –  tchrist Jul 7 at 22:49
    
So, is this nice to have or mandatory? I can give the name of the link using <sub> tag at least if it is going to make it better. What does everyone think about this? –  ermanen Jul 7 at 22:55
4  
A link to the source of a dictionary definition, plus a colon, is quite sufficient as far as I'm concerned if the link is to a reputable online dictionary -- in other words, when the link is to a non-controversial source. (If a reader is too idle, or the topic is insufficiently important to them to click on the link, that is up to them.) In my opinion, explicit inline attribution only becomes necessary if it is apparent that a controversy exists regarding either the definition or the general subject of the posting, or where the issue being discussed is a complex one. –  Erik Kowal Jul 7 at 23:05
6  
@tchrist - Two points: 1) Despite your professed ambition that "community consensus can be reached, and for that to happen, multiple viewpoints must be heard", your speedy rebuttal of my points shows you are more interested in trying to impose your own views than first hearing what others have to say; and 2) I still don't see why a non-controversial definition has to be explicitly attributed to within an inch of its life when a single click can take an interested party straight to its source. Frankly, I think your attitude to this issue is nitpicking, controlling and unnecessarily bureaucratic –  Erik Kowal Jul 8 at 1:18
4  
@tchrist - You've just demonstrated exactly the kind of approach that I was referring to in my previous comment... Why should I have to post this as a separate answer, thus divorcing it from the context in which I was making it? –  Erik Kowal Jul 8 at 2:45
    
@tchrist : In regard to your most recent comment to ErikKowal 'Because then people can vote on your position', don't the upvotes on his comments count as people voting? If not, then what's that function's purpose? –  Souta Jul 8 at 23:53
    
@tchrist If one disagreed, one could simply argue their point in the comment section. One could also argue, that the rebuttal argument could be seen as a downvote to the argument they are going against. –  Souta Jul 9 at 0:01
    
@Souta One is not supposed to argue in comments. –  tchrist Jul 9 at 0:02
1  
@tchrist Erik was giving more information on an answer he apparently agreed with. Something that the comment section states is its purpose. –  Souta Jul 9 at 0:08

Downvotes are for poor answers so while some postings may also need to be downvoted I don't think downvoting is appropriate for missing citations.

Commenting gives the best hope of fixing the answer, but there's no guarantee. In the meantime we're hosting content that's essentially stolen; this is not the academic way.

Editing to add attribution sounds fine, but won't always apply. If it can be done then I say go for it.

Flagging may create a lot of work, and what exactly is accomplished? The question/answer/comment hasn't been improved, and the content is still there.

I think deletion is the most appropriate. We may loose content, but was it quality content? It's not that hard to cite your source.

In regards to adding the Author's name. I think that's overkill. The point of a citation is give credit. So long as the source can be found credit has been given. However I think leading by example would be useful here since including the author's name is good practice; I just don't think it needs to be required.

share
    
I agree that flagging would be just putting work on others, which I am not in favor of at all. I put it out there because it has been a suggestion made, that it be flagged as plagiarism for a completely unattributed verbatim copy. I don’t want to make more work for moderators; I would like to see people do a better job at attributing that citations correctly. –  tchrist Jul 7 at 23:30

Since I am especially interested in EL&U questions that involve the historical development and alteration of word meanings, or the first occurrence of a word or phrase, providing source information is crucial to the practical value of my answers. A bit of advice I received (some time ago) in a comment from Hugo about the usefulness to others of linking to (as well as quoting) online references has encouraged me to try to be thorough in documenting my sources.

The drawback of this approach is that such information adds an element of stiffness, formality, and physical length to answers, which some answerers may find incompatible with their writing style.

I also wonder whether, for answerers who embrace the "competition for reputation" aspect of this site, the desire to be quick with an answer in order to get it posted (and collecting Up votes) as soon as possible doesn't serve as a constant (though probably in most cases low-level) motive not to show their work. To judge from some of the scores earned by answers with unattributed quotations, many voting readers don't have a problem with the lack of annotation.

Ultimately, I think, it's important to believe that good and useful content drives out (or at least outperforms) bad and useless content. To the extent that it does, we can all contribute by endorsing good questions and answers, and by constructively criticizing—or at least refraining from approving—deeply flawed ones; to the extent that it doesn't, we may simply be dealing with an imperfect website in an imperfect world.

share

It is nice to see that even though the rule now is that "moderators are instructed to delete on sight without further warning any content that is not properly attributed", practice shows that editing extra text is still an option, as in this ELU question.

The moderator's comment suggests this is a one-of show of leniency...

I am waiting for this delete-on-sight policy to be put into practice.

I have so far flagged a number of answers, including very good and useful ones, for moderator attention (that is, instant deletion).

I once read something about SE being moderated by the community, but now it seems that the community is being told not to attract any moderator attention if they do not want to see potentially useful information being deleted.

That seems strange. If moderators are being instructed to delete on sight, then certainly users should be encouraged to assist moderators in their work, as they have always been.

If certain content (like an obvious link to an on-line dictionary without an explicit mention to the name of the specific on-line dictionary) is something that the powers that be deem so unwanted on any SE site that it deserves immediate deletion without reprieve (the same way spam is treated), it is hard to understand why users should not be encouraged to treat those terrible offensive questions and answers in exactly the same way as spam.

Telling users to look the other way while instruction moderators to execute on sight is a contradiction. Are users, in order to preserve useful information, now urged to comment on posts, urging the author to "quickly add correct references before a moderator sees it"? I remember those things from the playground... but since we are all over 13, it strikes me as strange.

If such a paradox is the only way to preserve useful content on this site, there is something very wrong in the rules that give rise to this strange construction.


How can the attribution in this answer be sufficient, by the way?

The images are, or seem to be, well-attributed:

All images provided by the respective Wikipedia article referenced immediately above each one, with one obvious fair-use exception.

However, there are no Wikipedia articles referenced! Yes, there are several links to Wikipedia articles, however, a link is not sufficient attribution. I guess there should at least be yet another addition to the post to state that the several unattributed links refer to Wikipedia.

Or, as another obvious option, I simply fail to see the difference between those links and links to dictionaries with just the referenced word in the link text.

share
    
I notice that you have not provided attribution to the quoted moderator directive as required by SE rules here final para.. Your answer has been flagged for deletion on sight. Have a nice day. –  Frank Sep 3 at 16:13
    
@Frank: I am assuming the author to be ELU (strictly SE) since copyright for original content is transferred to SE as per the rules of this site. Unless I am terribly mistaken, in which case I understand deletion of my answer. –  oerkelens Sep 3 at 17:16
1  
@Frank: your comment is exactly the kind of comment that seems to have to appear under a vast amount of useful posts, and it illustrates very nicely the problem I have with this rule. Thank you for the example! :) –  oerkelens Sep 3 at 17:17
1  
What would be most useful is to have a moderator written meta post (locked) that explains the rule and provides a number of example methods of attribution (for both text and images whether from the internet, scanned manually or transcribed) and some statement that if the attribution is not provided in 'xx' hours the question/answer will be deleted. That way any user could provide a standard comment 'Please attribute your sources after reading link to moderator meta post', flag it for the attention of a moderator and they could then follow up 'xx' hours later and delete if required. –  Frank Sep 3 at 17:27
1  
@Frank: Such a reference source would be handy, but we shouldn't forget to reference that post correctly in the comment. I still find it strange that we would treat people who provide an in-text link completely in good faith the same way as someone who copy/pastes without even admitting it is not their own handiwork... –  oerkelens Sep 3 at 17:32
    
Copy/pasting with zero attribution is obviously the easiest way to skirt the whole issue. To delete such a post based on it being plagiarism, the moderators would need to find the plagiarised source. Not an easy task sometimes. –  Frank Sep 3 at 17:36
    
Please flag my posting for immediate deletion since it offends you so badly. Or simply edit it yourself until it meets your expectations of propriety. –  tchrist Sep 3 at 17:48
2  
@tchrist: I did flag it. But supposedly attribution in the image's ALT tag was deemed OK. How that is OK, but a link target is not is beyond me. Anyway, it seems that at least some moderators will, contrary to what we have read, not delete on sight, but rather give the poster some to include proper attribution. As you have been the most outspoken about what you consider proper attribution, I suggest you take it upon yourself to give the correct example and explicitly identify your links as Wikipedia links. They are just underlined animal names now without sources. –  oerkelens Sep 3 at 17:54
    
@Frank — but if such a source can be found (and it often is when an on-line dictionary is copied!), I fully understand the hard-line approach towards the plagiarising poster. However, people are now called out for "plagiarism" while they do the same they have been doing before, because of a hard-line interpretation of one added line somewhere in the help centre... I feel almost afraid to post any link and I have already decided not to answer several questions today in order to avoid such crimes as thinking that a link to Wikipedia would indicate that I am quoting Wikipedia... –  oerkelens Sep 3 at 17:58
    
We'll get into trouble for 'chatting' now but it's meta advice. I use the Wikipedia 'cite' tool and any other 'cite' tool from dictionaries and such like. Just copy and paste the citation somewhere in your answer and you're covered attribution-wise; there's no mention in the rule that the cite must be next to the copied works if you don't want to clutter up the body of an answer with citations. (but now I've mentioned it ...) –  Frank Sep 3 at 18:10
    
Give @tchrist his due. I pointed out to him that he had failed to attribute his images properly and he did add attribution to Wikipedia in plain text (for all images) which I believe is all that is required. I then pointed out that one image required the author to be named in the attribution and he changed the image to one with a less demanding attribution. I think that's fair enough. –  Frank Sep 3 at 18:19
    
@Frank I'm not referring to the images. If "a link is not an attribution", the links to the animals on Wikipedia are not attributions. But ok, if linking to Wikipedia is allowed like this whereas linking in the exact same way to dictionaries is a crime, maybe you can explain the difference to me? I'm not after anyone in particular, I am just really having a hard time understanding. –  oerkelens Sep 3 at 18:28
1  
You seem to be misunderstanding "delete on sight". This refers to unattributed content only and not to an entire post. It is also extremely unlikely that content will be removed without giving authors the opportunity to remediate. It is far more likely that attributions will simply be added to content, since our preference is to improve posts rather than remove them. –  KitFox Sep 3 at 19:21
1  
Also, I'm not sure what your expectation with the Wikipedia attribution is. There are a series of pictures with embedded link captions and then at the bottom, it says they are all from Wikipedia. That's fine. I'm not sure what you mean by "there are no proper wikipedia links". How are they not proper links? –  KitFox Sep 3 at 19:24
2  
@oerkelens I think you're misunderstanding what requires attribution. The links to wikipedia are just links. Links are not plagiarism. tchrist didn't quote those wikipedia articles, so there is nothing to attribute. –  Mr. Shiny and New 安宇 Sep 3 at 20:33

TL;DR

I believe providing links to quotes can be enough to be attribution. Asking for the source to be apparent from the post's text is more likely to prompt inconclusive debates about style rather than good attribution practices. Three Ugly questions are examined with the intent of supporting that claim.

Intro

Growing up I remember a debate among fellow players in a sport about how focused our rules should be. One side said they ought to clearly delineate what is and is not allowed. The other side agreed in principle but cautioned that such a task was difficult (if not impossible) to do -- particularly without coming across as legalistic. Some even thought it might promote shifty behavior, intentionally finding loopholes, perhaps to use as tactical advantage. And so those who said nay to further clarification took to using a phrase that I would like to invoke here: the spirit of the rules.

A Tale of Two Citations

The spirit of ELU's rules about citation, as I understand them, is primarily about academic honesty: giving credit where credit is due, or (more bluntly) not implying the work of another is one's own. These reasons motivate the opening and closing thoughts of the ELU Help Center’s page on How to reference material written by others:

Plagiarism - posting the work of others with no indication that it is not your own [emphases added] - is frowned on by our community, and may result in your answer being down-voted or deleted.

...

And always give proper credit to the author and site where you found the text, including a direct link to it.

Not mentioned, but just as important a hallmark of academic honesty, is to try to present cited material in the way its original author intended. This Meta question covers a serious corruption of the first hallmark, plagiarism. And it has not even touched on the corruption of the second, misinformation:

false or inaccurate information that is spread unintentionally. It is distinguished from disinformation, which is intended to mislead.

I have structured the above citations to display a point. The quote from the Help page goes above-and-beyond the thoroughness I think necessary. In it I:

  • Maintain the cited work's hyperlinks and emphases, regardless of how irrelevant I think doing so may be to the topic at hand. The cited work I thought was important thought they were important, they're related (or at least not spammy), and I know how to include them. So why not?
  • Use ellipses to indicate omitted information, indicating to the reader implicitly that:
    1. The given information may not comprise all the link's material on the topic
    2. Any loss-of-flow in the cited passage is my fault and not the original author's
  • Indicate when I have tweaked the source, using the plural emphases to indicate that both the bold and italics are not necessarily the original author's.

And yet this first example has not even met the standards I was taught in academia! Because of the ephemeral nature of online information, I was asked to provide the times I accessed a webpage when learning the MLA citation format. But posting a timestamp "7 July 2014 20:30:16 MST" to all (or even to the most-likely-to-change) webpage accesses, while official and precise, is not something I would ask every poster to take time to do. (For my purposes, the "upper bound" to when the poster could have accessed the information, automatically provided by the edit/answer capabilities of SE, is more than sufficient.)

I posit that the second citation, while not as rigorously thought-out as the first, still ought to be regarded as fine, even though:

  • It does not include the links to the False, Information and Disinformation wikipages that the original citation now links to.
  • Not including ellipses or words in the beginning may be seen as a misattribution of source. The wikipages don't go around starting fragmented sentences with lowercase letters like that! :)
  • The added emphasis on unintentionally is not explicitly noted.

I think these oversights should be pardoned because they adhere to the spirit of the rules. By virtue of having a hyperlink on the word misinformation that links to a page that reasonably could have contained the information the poster claims it does and by using the site's "< quote" markdown, I think the poster has put enough effort toward academic honesty, that a link alone can be considered attribution -- the poster is pointing to where the credit should go.

But is it properly used attribution or is it misinformation? In this case I would say "used properly enough." I think the poster's emphasized text is consistent with the article's point, so it doesn't matter to me if it's notated because it's being used as a tool to highlight the poster's point.

I agree that posts that present a source's material verbatim without crediting the source ought to be handled, preferably edited to include the source. "Bad" and "Ugly" posts that can include these changes will likely be improved. But I think it's more likely enforcing these changes will invite debates about style. To demonstrate, here are three cases from the Ugly section above.

Case #1 (Ugly #5)

Poster answers Is there a word for telling the truth (technically) in order to misguide? with

Equivocate: To make a statement that is capable of being taken in more than one way, with the aim of exploiting the ambiguity.

The answer does not appear to have been edited since its posting Nov 14 '11 at 15:40. The link on Equivocate leads to an Answers.com page that after two years still has the cited text, attributed to The Oxford Dictionary of Philosophy.

Admittedly, it would have been better to draw the word from a publicly available or hard copy. But I see little difference between trusting a web service's citation of such a source versus an ELU member citing a hard-copy. Though reputation of either helps me intuit what the situation is in both cases, I wouldn't count either as invalid until I had seen the hard copy or disputing sources. Unless disproved, asking for the reference name here seems more like a stylistic choice than keeping with the spirit of proper attribution.

Case #2 (Ugly #7)

Poster answers Single-word synonym for a “pedantic rule-follower”? on Oct 3 '12 at 21:56 with

I've encountered a few people that you describe. Often, they were bureaucrats:

An official who is rigidly devoted to the details of administrative procedure.

The answer is later edited by another user on Jan 2 '13 at 21:32 for the given reasons of "readability (typography), footnote" to

I’ve encountered a few people that you describe. Often, they were bureaucrats:

An official who is rigidly devoted to the details of administrative procedure.¹

The footnote is easy to miss in the edit, but stylistically I agree with the decision. The page leads to the singular bureaucrat, so OP's decision of linking the more obvious bureaucrat (minus s) in the link makes sense. But the editor's change to a footnote is more aesthetically pleasing, as is the change of bold text to less-heavy italics. After about a year-and-a-half, the link to TFD in the original, which stayed through the edits, still contains the cited text but without emphases. I believe any changes to this answer's attribution would be stylistic.

Case #3 (Ugly #13)

Also answering Is there a word for telling the truth (technically) in order to misguide? but on Nov 14 '11 at 15:06:

I would use the word prevaricate:

to speak falsely or misleadingly; deliberately misstate or create an incorrect impression; lie.

Another commonly used word for this same behavior is to fudge, meaning to disingenuously avoid or talk around an issue.

Because prevaricate follows the patterns I defend above, I will focus on to fudge. This appears to be a paraphrase, because my Google search of the quoted phrase leads primarily to this answer.

I suspect OP meant to tag the answer currently above it, which neither cites nor defines its suggested word of obfuscate. A source will certainly improve this answer in my book. But if that's not the case, I think adding in that the cited definition for prevaricate comes from Dictionary.com is mostly a stylistic choice.

End

Because these three cases which I believe have a fair case for legitimate attribution, have been lumped in the Ugly category with the worst offenders because they don't meet a rule about providing the author's name, I think it's better to go with a looser spirit of the rules philosophy. Does it seem like a person is farming rep without attribution? That can't go on. But anything aside from verbatim-without-citation probably ought to be handled on a case-by-case basis.

share
    
Given how within a few scant hours of when you posted your own adamantine — but by no means discourteous whatsoever — position on the notion that due source citation should be a matter not of exigency but of courtesy, The Powers That Be have seen fit to update our Help Center’s section on how They expect us to properly cite text which has been elsewhence copypasted into an SE posting, I cannot help but wonder whether said update may have in part mediated and perhaps even mollified your adamance with regard to this sensitive matter which is now under fruitful discussion here on ELU’s Meta. –  tchrist Jul 8 at 17:58

I like the direction that ermanen and Erik Kowal are going in. I also agree with the comment Erik Kowal said about nitpicking. (His comment can be read under emanen's answer--linked above.)


I'm unfortunately on this list in a rather unflattering light (thankfully not Ugly, but The Bad #34). I've since corrected my answer, and hopefully it's better now. *Please note: I have screenshots in case linked-to points are tampered with.

If commenting is an action that should be taken, is there a comment that should be copy/pasted to stay uniform (and to make sure everyone is responded to fairly)?

I agree that flagging is not the answer. Flagging will cause mods to lose sight on their main purpose to SE. Deleting posts seems too rash. While they may not be appropriately cited, who is to say that their post isn't beneficial to SE? I think those posts should just be edited--either by another person or by the one to post the atrocity.

There is a blatant lack of consistency: OP has gone on a rampage and is rather unfair to some while being somewhat fair to others. Sure, they made it to The Bad list. However, the rhetoric isn't the same. (I find this to be more than relevant to the question. The question talks about posts not being uniform to the Help Center and a way to correct this. This response is relevant because if we're going to nitpick on consistency, then perhaps notifying people with vigilante comments should be consistent and polite.)

Not everyone is treated fairly when getting a nasty-gram about their questionable plagiarism or line-bordering plagiarism. There is a non-uniform rhetoric (sometimes with grammar issues). Then there is this case where it is overly friendly #26. Then there is a scenario where there is a blatant disregard for notification for the dastardly plagiarism, but instead a playful comment (listed on The Bad #7).


Also, consistency is lacking from OP. Where is your mention of the source for the link you provided? After all, quite a few of those on The Bad list did exactly what you did and, well, they are on your hit list.

Ironic hypocrisy


SUPER ULTRA POINT TO BE MADE HERE: OP NEEDS TO SEE THAT THIS HAS ALREADY BEEN ADDRESSED ON META SE. (link)

share
    
*To clarify, when talking about consistency, I'm referring to the plagiarism and how things need to be cited. It has been made a point, by OP, that everything must be cited in a precise manner. –  Souta Jul 9 at 2:42
1  
My name is Tom, not OP. I shall permit you to call me tchrist. But I will not permit you to call me a hypocrite, especially since the evidence to the contrary is staring you right in the face: I named who the author was for the citation and provided a link along with the text. Simmer down. –  tchrist Jul 9 at 2:53

You must log in to answer this question.

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