I'm building a public open source package that uses, as part of the library, a modified script from an already modified code originally under the GPL v3 license.

I want to allow optional academic citations in my package. E.g. in the readme of geobr the citation suggestion reads

Pereira, R.H.M.; Gonçalves, C.N.; et. all (2019) geobr: Loads Shapefiles of Official Spatial Data Sets of Brazil. GitHub repository - https://github.com/ipeaGIT/geobr.

How should I suggest the citation to be? Can I put only my name? Or should I put the previous authors, even without their consent? I've cited the previous authors in the modified script's docstring.

  • 1
    Is the code or the license modified? What do you mean with use and what kind of citations? Please edit the question for clarity Jul 24, 2020 at 3:12
  • Academic citations, like in the end of this README.md github.com/ipeaGIT/geobr . Just edited the first line to improve clarity @planetmaker Jul 24, 2020 at 3:18
  • 1
    @PatrickNasser I think confusion may be arising as to whether the readme you link to is the readme of the package you've modified, or simply an example of package-based citation style. Are you asking how to modify a pre-existing citation recommendation in a package you've modified (see link to actual package), or are you asking how to introduce a citation recommendation in a package of your own (see link for example of someone else doing it)?
    – MadHatter
    Jul 24, 2020 at 5:47
  • @MadHatter it was just an example of package-based citation style. So it's the second one! Jul 24, 2020 at 13:26

1 Answer 1


A section for "how to cite" is legally speaking a nice-to-have part often found in repositories used for academic research. It is not a requirement by the licenses which require that proper attribution is given or retained within the repository as to not mis-represent who contributed and what is the license.

The suggestion on how to cite generally should be treated with similar measure like author information for license considerations though. While you have the right to remove the suggestion and replace it exclusively by your own, doing so might not be the wisest choice - but depends on how much you modified the content of the repository and the code's functionality.

The likely best approach is to amend the quote suggestion such that you either add an additional line which refers to your repository or to modify the existing in a way that both you and the original authors get credit (but possibly only linking to your repo).

A suggestion following the suggested quotation from the link you provided could be

Name, Your (2020); Pereira, R.H.M.; Gonçalves, C.N.; et. al (2019) geobr: Loads Shapefiles of Official Spatial Data Sets of Brazil. GitHub repository - https://example.com/YourRepositoryLink.

  • 2
    I haven't been an academic for some years, but trying to combine all the contributors into a single citation, when none of them have actually collaborated with me, wouldn't have been something I was happy about, way back when. Maybe someone with more current academic "chops" can come along and give us some guidelines into current practice. Also useful for clarification: if in my (primary) work I cite a work on which I have explicitly relied, but that second work in turn was taken in large part from a third work, should I in the primary cite only the secondary, or both secondary and tertiary?
    – MadHatter
    Jul 24, 2020 at 5:55
  • 1
    That's why I made it - IMHO - clear that it's a sequence of contributions by giving the different year. The citation here does NOT cite a paper but only a git repository - that's why I made it one citation as the code cited only has one source with several authors: the repo. It would look different, if it were citing technical papers published somewhere else - then I'd cite both separately. Jul 24, 2020 at 6:01
  • Should the quotated citation suggestion actually be a citation suggestion for an existing paper, it would be a bad one as it is missing the journal. Jul 24, 2020 at 6:05
  • Fair enough. I guess I don't understand exactly what the OP's trying to do - see my comment on the question - and I'm not entirely sure I understand why. If the citation style fulfils neither copyright licence requirements (which as you point out, it doesn't) nor academic citation requirements (as I don't think it does), I'm not quite sure what it's supposed to be for. The question title certainly suggests an academic purpose is intended; if this is so, I stand by my objection.
    – MadHatter
    Jul 24, 2020 at 6:53
  • I read it as the desire to cite a code base in a scientific article where the code used is not published elsewhere. One doesn't write a paper about each code. But it is good practise to still make it available and give credits where credits are due Jul 24, 2020 at 7:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.