19

When reading the code of conduct of this site, I was a bit surprised not to see any explicit mention of "academic honesty" there. According to how I read this, you can shamelessly copy tricks or code from other posts as long as you are polite and friendly on the surface. The issue itself has, to some extent been discussed in this post. Personally I would define it slightly broader, i.e. go beyond the question whether or not a user has converted a code fragment to an MWE. In fact, personally I would wish that anyone learning a trick from another answer and using it in their own answer mentions the origin of the trick. Yet the purpose of this post is not to discuss these "details", rather to urge the "powers" to extend the code of conduct by the item "academic honesty".

This is the reply by the officials

Hi marmot,

Thank you for contacting us. I hate to send you off in a new direction, but the best way to get your idea seen is to post it on the community’s Meta. Stack Exchange is collaboratively built, maintained, and ultimately designed by the community. Right now, this email is somewhat walled up in a back room while most of our design is done collaboratively in public view. The best way to get ideas seen is to get it out there where the users and developers can vet your idea and discuss the benefits and any potential issues. Meta is the way to go.

Regards, Stack Overflow Team

I am writing this post because I got told to do that, and, more importantly, to have a basis for a discussion with the "authorities". The first step seems to be that we agree within the TeX community on whether or not we want such an addition. Any feedback, of course also critical feedback, is welcome!

Further clarification: There are very different levels of "academic honesty" one may think of. In the nice answer by Raoul Kessels, a very strict implementation is discussed. This is not exactly what I had in mind when bringing this up. I guess if one really was to go for this, we would have to amend a one-line answer by a lengthy list of references. What I had in mind is

  1. if there is already an answer or a comment that goes into the same direction, one may want to acknowledge this. (Sometimes it happens that one sees a question, knows how to answer it, and when one is ready with writing the code, a comment has popped up going in the same direction. I personally still write "As pointed out by ..., one can do ...", as this does not imply that I learned this from the comment, but just to give feedback that the two of us independently thought about something similar.)
  2. Of course, if one makes use of another post, which is in turn based on yet another post, I personally think one should only "cite" the more original one unless the later one adds something important. That's of course just my personal opinion.
  3. If one learns something from another post rather than a manual, say, and uses this much later in an answer, I personally do not think one has to quote the post as long as this does not go beyond the manual. This does not at all mean that answers that "only" point to a specific place of the manual are worth less. It just means that after some long time it may not be possible to recall where one got the information from, and if it is in a manual, then this might qualify as the "original" source.
  4. As for the question of enforceability: this is no different from the other things discussed in the code of conduct, and I never had "enforceability" in mind when writing this. As far as I understand, the code of conduct does not really get enforced. The reason why I am bringing this up is to say that the expected behavior is to give credit to sources of information.
27
  • 1
    How enforceable is this? And if not, what value does it add? Who it the target audience this is aimed at? (Mostly) members of the site, or people who copy content from here only, or both?
    – Werner Mod
    Dec 5, 2018 at 6:47
  • 2
    @Werner It is as enforceable as the other things mentioned in the code of conduct. And I do not think one has to restrict the target audience, but of course a proposal written here will be read, discussed and perhaps agreed on by users of this site.
    – user121799
    Dec 5, 2018 at 6:52
  • If I'm not mistaken, isn't content published here licensed under a "Creative Commons Attribution Share-alike" license?
    – gusbrs
    Dec 5, 2018 at 23:16
  • Mmh, there is this What is the license for the content I post?. But, to be sincere, when you actually click on the Terms of service linked there, things seem less clear cut than that... At first glance, this applies to the "Creative Commons Data Dump" which "from time to time, Stack Overflow may make available". I recalled the issue of the license because of the recent outcry due to the recent design changes.
    – gusbrs
    Dec 5, 2018 at 23:44
  • Ah, it is clear below, at "Subscriber content", for which "You agree that any and all content, including without limitation any and all text, graphics, logos, tools, photographs, images, illustrations, software or source code, audio and video, animations, and product feedback (collectively, "Content") that you provide to the public Network (collectively, "Subscriber Content"), is perpetually and irrevocably licensed to Stack Overflow on a worldwide, royalty-free, non-exclusive basis pursuant to Creative Commons licensing terms (CC-BY-SA) [... etc etc]"
    – gusbrs
    Dec 5, 2018 at 23:48
  • @gusbrs Yes, but this just means that I do not own these posts, right? That is, there are no legal consequences, right? I am actually not really interested in legal consequences, all I want to say is that being nice to each other implies IMHO also to give credit to others who did similar things before, in particular (but not only) when they did that in an answer to the same question.
    – user121799
    Dec 6, 2018 at 0:07
  • 1
    I understand the point of your question. I can relate with your yearning, but I don't think the proposal would change behavior much. My point is that it may somehow be implicit in the license. Are you acquainted with Creative Commons licenses? In the case of the license used here, it states you must "give appropriate credit" to use the content. See creativecommons.org/licenses/by-sa/4.0. However, I guess the formal way to read it is that one has to give StackOverflow (the copyright owner) credit for content reproduced outside the network.
    – gusbrs
    Dec 6, 2018 at 0:20
  • marmot, never mind the old grouch realist here. I didn't step in to antagonize, on the contrary, just to add the information. ;)
    – gusbrs
    Dec 6, 2018 at 0:29
  • @gusbrs Your comments are all well taken! Obviously I did not do a good job in writing this post. I hope someone more eloquent does posts some similar post elsewhere, and these things will make it to the code of conduct.
    – user121799
    Dec 6, 2018 at 0:39
  • 4
    @gusbrs Why do you think SE is the copyright owner? You've licensed it to them, is my understanding, and not given them copyright. If you gave them copyright, people couldn't dual licence code here, which they do. (There's a Meta question where people list the licences they release code here under, in addition to the CC one.)
    – cfr
    Dec 7, 2018 at 1:30
  • @cfr I think you are correct. This started as a vague recollection of the "attribution" part of the license. I never delved much into the terms of service, but it seems I did mix the "StackOverflow content" with the "Subscriber content" there. If so, in the end it seems the license does require attribution to the author after all.
    – gusbrs
    Dec 7, 2018 at 1:43
  • These stupid “Creative Commons 08.15 Attribution Share-alike Name-your-firstborn-after-the-author” licenses, are the reason why everything I post here is public domain. If you want proper credit for your work, you shouldn't post on this site. Dec 7, 2018 at 8:43
  • 1
    "Stack Exchange is collaboratively built, maintained, and ultimately designed by the community." Yeah, no. Recent events have proven this statement wrong.
    – Skillmon
    Dec 7, 2018 at 13:50
  • 3
    "I would wish that anyone learning a trick from another answer [...]" while in theory I do share your opinion, often one does learn a trick and then reuses it without remembering where he first learnt about it. I know that I didn't think of \if\relax\detokenize{#1}\relax as a very stable way of testing for an empty argument myself, but honestly have no idea where I learnt about it. Same is true for many other small TeX programming tricks one learns over the time.
    – Skillmon
    Dec 7, 2018 at 13:54
  • 1

5 Answers 5

13

Let me just throw around a few thoughts here.


The code of conduct in its current form (as found on https://tex.stackexchange.com/conduct) seems to focus mainly on being friendly and welcoming and tries to shut down offensive and offending behaviour. Many of these things are more concerned with how things are said and how they are received and perceived rather than with what is said. Academic honesty is about what is said and acknowledged in answers. I feel that it operates on a slightly different level than the focus of the current code of conduct. When the code of conduct was created, other patterns of behaviour were targeted than the factual contents of answers and their origin and background. That is not to say that I think it academic honesty can't or shouldn't be added to the code of conduct, but I believe that it explains why there is no mention of it in its present form.


On a technical level crediting other answers whenever one copies content is probably already required due to the license of the content. gusbrs mentioned that in the comments under the question already. As far as I understand you remain the copyright holder of your post and would thus have the right to pursue license violations. I'm not sure if you would have a case against the individual that posted the copyright violation or only against Stack Exchange (with which the terms of service seem to agree a separate license agreement "pursuant to Creative Commons licensing terms (CC-BY-SA)"). But license issues are terrible enough as they are and I can understand that one would not want to evoke them in this context. (After all I would assume that most posts that do refer back to other people's work on this network would not meet the formal requirements of CC BY-SA 3.0 as clarified in https://stackoverflow.blog/2009/06/25/attribution-required/. And most answers usually don't acknowledge the MWE for the question as starting point.)


If academic honesty is to be added to the code of conduct it should be clear which level of honesty we are talking about.

I'm pretty sure that (almost) everyone agrees that credits should be given if an integral or non-insignificant part of the answer or the idea of the answer is taken from a different post. Especially if the new post was created by copying and pasting larger blocks of code from elsewhere.

Other cases are less clear cut, you may have learned a certain trick somewhere on this site (Skillmon mentions \if\relax\detokenize{#1}\relax in the comments), but might have gotten so used to it that its application becomes normal. Most answers are collections of many such idioms and tricks and so it seems nearly impossible to credit all sources of inspiration. See also Raoul Kessels' answer. I'm not sure if the point 3 in the further clarifications is going to help a lot here: Naturally, if things are in a manual one could try to defend against a 'you didn't cite my answer' claim by pointing to the manual as possible source of the idea. But then why should there be no obligation to quote the manual? Would the manual of a random package really be a good source for a common or even uncommon trick if there is no further explanation of the trick and no indication that the author is the 'inventor' of the trick? What about small tricks that are not in a manual?


Re point 1 in the further clarifications: I don't think there should be a requirement to acknowledge previous similar work if you did not use it to come up with your idea (and especially if you were not aware of that work at the time of writing your own post). This community is generally quite good at avoiding writing multiple answers using the same methods and usually ideas from the comments are not usually 'snatched' by other people. TeX.SX is not like academia where novelty is important and a survey of related work is expected.


Academic honesty could be enforced via the same channels as the rest of the code of conduct and I believe that the mere fact that it would be part of the code of conduct would already make people aware of the fact that attribution is important and not just best practice or a nice gesture.

But I fear that it might be harder to enforce than offensive, unfriendly or unwelcoming behaviour. For the latter the community must accept (and I assume has accepted) that there are no universal objective standards and that at least some of it comes down to subjectivity and feelings. If I flag a comment as offensive it is implied that I think it is offensive or could offend some people. If a moderator then agrees with my assessment the comment is removed. There can be little disagreement about 'the facts' in those situations since it is accepted that feelings play an important role here (indeed the feelings are the fact). With academic honesty, on the other hand, there could be discussions about 'the facts': If you claim that I did not behave academically honestly by not crediting a certain answer, I could try to appeal starting a discussion about whether or not the answer is relevant, about whether or not the idea could have come from somewhere else (a manual that need not be credited?), about whether I had seen the answer before I posted mine etc. That could get messy...


On the subject of enforceability (see also the comments), I believe that a code of conduct does not live exclusively through being enforced, it also shows the will of the community (or the powers that be) and their position of what is and what isn't acceptable behaviour. But the code of conduct (at least in its current form) should be enforceable if need be. So the question of enforceability is not one we can completely ignore when we aim to add something to the code of conduct.


Personally, I didn't think a code of conduct was required: We are all decent human beings after all. But apparently it was needed. I didn't think academic honesty (at least in a the sense of citing other people's work if one's answer is based on it to a significant degree) would have to be added to the code of conduct, but I accept that I am very likely wrong about that as well.


Lastly, I thought that the code of conduct was a network-wide thing, so I'm not sure how much our thoughts matter here. At some point this would probably have to be discussed at https://meta.stackexchange.com/.

5
  • I agree with everything you are saying. My post must be really poorly written because I am not so much concerned about the enforce question. It is more to have something that says that the community agrees that just copying another answer, or key elements thereof, and selling that as one's own answer should not be the way users behave. Unfortunately, there are a few who do not even agree with this, and we could have some statement (certainly not my poor post) which is "official" that one can refer to when one comes across some of these things.
    – user121799
    Dec 8, 2018 at 12:51
  • 2
    @marmot Mhhh, while I posted under Alan's answer that I think that a code of conduct does not exclusively live through enforcement, I do think that at least in its current form everything in the code of conduct should be enforceable if need be. It is probably fair to say that when asked about academic honesty the community would unanimously come out in favour of it. I gather that you speak from experience when you say that people use significant portions of other people's answers without attribution, and I can't say that I have come across many answers where I realised that was the case. ...
    – moewe
    Dec 8, 2018 at 13:52
  • 1
    ... In those cases I take it that the tacit assumption of the community that academic honesty is a thing did not work for them, but then I'm not sure if a mere "official post" on meta would help if there is no repercussion if no attribution is given. Which means in the end you are back with the code of conduct and enforceability.
    – moewe
    Dec 8, 2018 at 13:54
  • Yes, "Many of these things are more concerned with how things are said and how they are received and perceived rather than with what is said. " is IMHO the only conceivable explanation of the absence of academic honesty in the CoC.
    – user121799
    Dec 18, 2018 at 19:56
  • @marmot I have a different theory, but that hardly matters for this discussion.
    – moewe
    Dec 19, 2018 at 6:52
14

In my opinion, while desirable, it is also very difficult to accomplish, even in the most honest way. The problems with all know-how, including code and all kinds of technology, is that once you know it, it stays.

This means that, many times, it is honestly impossible to give credit to whom gave you the idea. Even if it is exactly the same stuff. You know it, you use it and in the end it comes naturally. You just do not remember where it came from. Much worse if you have modified slightly some details or simply some command, from \zzz to \mycmd for instance, to keep track of what is happening.

Naturally, it is still worse when you see so many examples daily, as in this site. You copy something somewhere, use it, use it somewhere else and when you post an answer, you just do not know where the code came from.

When I wrote this question, I was already using the codes and had to do an explicit search to find all the authors of the different formats. I did it because it improves the question and obviously it was not my code in any case.

So, in my opinion, it desirable and it should be encoraged, but it might be impossible even being honest.

5
  • 1
    Thanks for your nice answer! As you say, there are different levels of "academic" honesty. The one you seem to imply, which seems to amount to keep track of where precisely one learns what, seems very strict, and may be, as you say, impossible to implement. Im my question I was referring to this post, which is about a much less extreme implementation: discouraging copying parts of another answer to the same question without giving credit, which may not be impossible. What do you think?
    – user121799
    Dec 5, 2018 at 14:34
  • In such a case, I think that the other answer should be mentioned. I see it sometimes in SE: "based on the answer by <user>". In any case, the original poster or anybody else can add a comment asking for an edit or even edit the answer to include the reference to the original answer. That said, I should not call it academic honesty, but just politeness. And I do not think it would yield more upvotes to the first poster. But you are absolutely right, the site is collaborative, but there is no collaboration between answers, only between Q and A; possibly because of the rating system Dec 5, 2018 at 15:43
  • I agree with all you're saying, but would prefer not to have to ask the user posting the second answer to give credit to the first answer if they use it. It is nice to see that we agree, but my post above is about my impression that we have a code of conduct that has several items, but academic honesty (or however you like to call it) is missing. Therefore, users may have the impression that it is perfectly fine to just copy code from others without crediting them, and all I want to achieve is (sort of official) statement by the community or authorities that it is not.
    – user121799
    Dec 5, 2018 at 15:49
  • "once you know it, it stays" - if only that were true... Obligatory xkcd: xkcd.com/1168
    – Marijn
    Dec 6, 2018 at 9:14
  • 2
    @Marijn Thx :-) Funny how there is an xkcd for everything Dec 6, 2018 at 12:24
5

Disclaimer: In the following I consider "Academic Honesty" as the "loose" definition from items 1.-4. in the question, not the very strict definition from Raoul Kessels answer

There is only one acceptable reason why academic honesty is not yet in the code of conduct: because it should be a givens.

Unfortunately some posts on the main site disprove my view. The problem is even broader than copying from other posts without mentioning the source, copy & pasting whole text passages from external websites without disclosing the source is also not unheard of.

So I strongly support that "academic honesty" should be added to the code of conduct.

In case we can't convince stackexchange that its addition is a must, we could think about finding an internal solution for tex.stackexchange. Maybe a meta post similar to Relicensing code from answers where people can enter their username if they support academic honesty.

3
  • 2
    I am a bit confused about all this "academic honesty". Isn't it a bit late at this point? My school teachers used to copy stuff from somewhere without giving a source. Images, data, texts, etc. Teaching (highschool) students how to work academically should be the way to go. Meaning you have to teach the teachers first.
    – Johannes_B
    Dec 8, 2018 at 7:17
  • 4
    @Johannes_B Just because something goes wrong in other areas, it does not mean it has to go wrong here as well. Dec 8, 2018 at 13:22
  • 1
    @Johannes_B It is never late. IMHO, It's only late, when it is never attempted for :) Dec 12, 2018 at 9:33
5

Whether something is stated explicitly in the code of conduct or not is (almost) irrelevant, to the extent that it really only can come into play if the SE "powers" are engaged in policing the behaviour by e.g., implementing suspensions or banning users.

In all other cases, behaviour on the site is by mutual understanding with (mild) "enforcement" of those standards by way of comments. For example TeX.se has a community practice of preferring to comment on unclear questions and waiting for responses from the OP before voting to close such questions. On other parts of the site, this is not the norm. It works for us, however, and we see it as a more friendly way of dealing with new users than closing and subsequent reopening when the question is clarified.

We also have a tradition of commenting on answers to improve them rather than simply posting a competing answer with almost the same content by with a minor change.

Most of us are also good at referring to answers on the site we use.

Does everyone do this? No. Should they? It would be nice, but it's unlikely to happen. So what can we do about the people who don't?

  • Post comments with links to answers that are copied from.
  • Edit answers to include links to copied content.

If an edit gets rolled back, then that behaviour (assuming it's unwarranted) can be policed if needed by our own moderators. Obviously we don't want to start edit wars, so posting a comment first would be the prudent first approach.

4
  • 2
    If academic honesty were added to the code of conduct one would have a stronger case in convincing other users to exercise it: As in "you must follow the CoC" instead of "please follow this uncodified, but established practice of the community". I'm not sure how useful that is, but I would argue that even without policing the CoC is not completely irrelevant.
    – moewe
    Dec 7, 2018 at 18:06
  • 2
    I usually go with the comment approach when I find copied code in questions (along with a short comment such as "parts of the code probably come from" or "does the code come from?"), but I feel that it could come about as passive-aggressive if not done carefully on answers.
    – moewe
    Dec 7, 2018 at 18:09
  • 3
    I agree with @moewe. If you, Alan, were right, the whole code of conduct would be pointless. And unfortunately I have come across users who said that they do not have to acknowledge posts by users that joined later than them, and similar stories. And I am not sure whom you refer to in "We also have a tradition of commenting on answers ... ", who is "we". Unfortunately my experience tells me that "we" does not include all users here. This sentence would IMHO a very nice thing to add to the code of conduct.
    – user121799
    Dec 8, 2018 at 6:24
  • 1
    @marmot Maybe it's just the kinds of questions you and I answer that makes the difference here. Or maybe we just react differently when it happens. But for the most part I don't see a big problem here with users in general. I'm not trying to downplay your experience of it, just pointing out that our experience is different.
    – Alan Munn
    Dec 8, 2018 at 13:42
4

Of course, I agree.

If someone takes your answers, makes some little changes and post it without mentioning the origin is undoubtedly unfair.

But I also agree with Raoul Kessels, sometimes it could be difficult to find the origin of the code.

For example, I answered here more or less copying from a previous answer of mine.

That previous answer was more or less copied from the link indicated in the question, which was copied from here.

Hence, eventually, I copied from this last link, but re-build the chain could not be easy.

1
  • I think it will just be an additional (but a very valuable) point in the CoC that explicitly asks the users to cite the answers/code/whatever they used (ofc, iff they referrred something), I think :) Dec 12, 2018 at 9:37

You must log in to answer this question.