I hope that is not a duplicate. Probably most of you have seen questions in which the OP presents a code which is very likely not written by the OP, but the OP does not provide a link to the original code. While I do not really care about some "implicit" copyrights for answers, I really feel that one could give better and faster answers if the source of the code was available. Then one would have many or most of the potentially relevant posts available right away, and not discover that one just reinvented a solution someone else found earlier. We all know that the searching tools on the main site have some considerable room for improvement, the selection of related posts is a bit obscure (e.g. for pretty much all TikZ posts "Is plotting exponential graphs a known source of bugs in tikz?" gets identified as related). So given that the links provided at the main site are not optimal, I have the following

Question: Is there a way to politely convince users to disclose the source of their posts?

What is the purpose of this question? Apart from learning some nice strategies to convince the OPs to reveal the source of "their" codes, I hope to collect in the answer(s) good reasons why doing so beneficial for them. One may use the answer(s) then to tell the OP why they should add the link to the source.

  • 10
    The problem with convincing the OP to disclose the source is that in many cases the OP simply does not remember the source and the most precise information one can get is "from somewhere on the internet" – samcarter Jun 13 at 20:34
  • 3
    @samcarter Sure, there are cases of that sort. But I also have the impression that some just copy something from somewhere, often the main site, try to quickly adjust it to their needs, fail, and ask a question. If we could convince those, this would be quite something, I believe. – marmot Jun 13 at 20:37
  • 8
    If I try to find the source of some code: google for a unusual line from the code with "..." site:tex.stackexchange.com – samcarter Jun 13 at 20:37
  • And I don't doubt that having the source of the code would be very helpful in many cases. – samcarter Jun 13 at 20:38
  • 1
    @samcarter Sure, I am doing the same. But I got sick of it. And sometimes it does not immediately give me the relevant post, at least not among the top search results. – marmot Jun 13 at 20:39
  • 2
    I think the one good way to prevent the problem would be if stackexchange would provide a button to copy code blocks and a link to the post would automatically be included as comment... – samcarter Jun 13 at 20:41
  • 1
    Ask them about the source of the code? – percusse Jun 13 at 23:41
  • @percusse According to my experience, this is not the best strategy, i.e. these requests get often ignored. Therefore I'd like to collect some convincing arguments. – marmot Jun 13 at 23:43
  • If even that is ignored then you are done as far as I am concerned. – percusse Jun 14 at 0:24
  • 2
    @percusse I do not want to educate people, my aim is simply to avoid redundancies and to provide a reasonable answer. And if they see some arguments that (even) they find convincing, that may help us reaching that goal. – marmot Jun 14 at 0:35
  • 4
    This happens a lot with PGF/TikZ questions where people copy code without, say, copying style definitions, which then obviously can't be used as-is and often can't be completed effectively. (By the way, I'm not sure copyright is the concern, even if your concern is of that kind, which I know it isn't. More a question of attribution/acknowledgement/non-plagiarism/courtesy etc.) – cfr Jun 14 at 3:15
  • 2
    Sometimes, one reason is that code is now broken. Where that's the case, a link enables somebody to update the code (if it is from this site), rather than leaving the original answer broken for future users. That's not an OP-centric reason, but it might work better in some cases just because of that fact. – cfr Jun 14 at 3:18
  • You are right, I should have put the quotes around "copyright". What I wanted to say that some people do not like seeing their code being declared the OP's invention, but this is not the reason why I am raising this issue. – marmot Jun 14 at 3:46
  • 2
    As with other clarification requests ('please provide an MWE', 'what is your desired result', 'which compiler do you use' etc.) you can always decide, if the OP ignores the request, that the question is not sufficiently clear and you should put no further effort in trying to answer it. – Marijn Jun 14 at 14:23
  • @Marijn Well, of course, I could also say "I don't care about other people's LaTeX problems." This question is under the assumption that I am willing to give a good answer, and do not want to waste my time with googling for some code. A different example of this sort is that we are forced to list at least one tag. In principle you could say you don't care if someone does not tag her or his question, you are just going to ignore it, but you may agree that the requirement to have the tag makes the quality of the questions better. – marmot Jun 14 at 22:35
up vote 17 down vote accepted

There are several arguments one could try. I'll let you be the judge of their strength and effectiveness.

Politeness

It is polite to acknowledge other people when you copy or build upon their work.

Help for other people

Revealing the origins of code may help other people with a similar problem who could for some reason (bad tagging, not the right search terms, ...) not find the original question but only the new question. The original answer may contain more details, explanations or alternative solutions and would thus be of interest. Vice versa, having the new question link to the old one will make it appear in the 'linked' section on the right and can help people find the new question from the old, which is particularly useful if the new question is a natural extension of the old one.

Code maintenance/bug reporting

If it turns out that there are problems with the code, knowing where it came from and where to complain/report/fix the bug helps improve the code for everyone.

Context

The original code and answer (and also the question) can help investigate the current problem. Knowing where the code comes from may help to establish context which in turn helps to understand the question better and more quickly. Maybe the undesired behaviour happened because the code was not copied in its entirety or modified in some way. Maybe the original answer mentions limitations or possible useful extensions of the approach.

So you can appeal to manners, altruism and egoism.


That said, I doubt people need to be convinced to disclose the source of their code. If you ask them in the comments I suspect most people will happily point to the source of their code if they know where it came from. Like samcarter I assume that many people simply can't remember where the code came from.

You must log in to answer this question.

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