I used many open-source packages in my research, such as matplotlib, numpy, pandas, scipy, etc. And I found this on the website of SciPy: Citing packages in the SciPy ecosystem. Should I cite every one of them in my paper? (I'd like to cite them but there are so many of them which will make my paper even half or one page longer)

  • 4
    Perhaps related: Should I cite all R packages I used?
    – GoodDeeds
    Aug 20, 2020 at 11:15
  • 2
    There is a more general related question too: What software to cite in a manuscript, and how?
    – Anyon
    Aug 20, 2020 at 13:18
  • Is your paper describing your research or your software? Aug 20, 2020 at 16:38
  • @Azor Ahai -- he him Well, I just used the packages for data extraction and analysis, that's why I felt confused. I'm not writing a paper to introduce my new software. I just wrote some scripts to process my data and draw some pics.
    – C.K.
    Aug 21, 2020 at 3:45

4 Answers 4


Presumably your research uses your software, which in turn uses and acknowledges open-source packages. Your papers need only cite the software used directly, which is perhaps just your software. You could also mention your software is built from open-source packages, but I don't think that's strictly necessary, since acknowledgements appear in your software.


I would apply the same prioritization as I would when referencing academic papers: Cite the packages you directly use, but not dependencies, as these are implied.

If you would like to be complete but are under a space or reference constraint in the main text of your publication (typically the case in big-name printed journals), you could include a list of references to all used packages as a supplementary table.

  • Thanks! One more question: should I cite programming languages such as Python/MatLab/R while almost everyone uses them?
    – C.K.
    Aug 21, 2020 at 12:32
  • 1
    If I was using solely base R, I would cite it. However, if I was using packages on top of R or Python, I would instead cite the packages, as the use of R is implied and the package citations will themselves cite R. I would not give a citation for Matlab as it is a commercial product, unless a citation was specifically needed to explain a method or function that I used in the paper. Whether you include a citation or not, you still need to say what software you used in your methods section. Oct 5, 2020 at 16:33

The correct answer is yes, it is because they are open source specialized packages. The answer could be be yes or no depending software developer. Best course of action when in doubt is to include the citations as if you were citing any other individuals source of work, to avoid the plagiarism snafu. Whether you do are not, I would highly suggest you bring this up with your instructor for direction to cover your bases.


In the "acknowledgments" section of your manuscript, you could say something like

The authors thank the developers of SciPy (Virtanen et al., 2020) for making their > code available on a free and open-source basis.

Depending on the circumstances, that may or may not add less extra length to your manuscript than trying to shoehorn a citation into the main text.

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