Authorship and direct/indirect contributions

[anniequinn]

Hello!

I work on a research project that encourages a highly collaborative workspace. As part of the project, two pieces of software in R have been developed, with varying involvement from different team members. With the publication of software being something that is relatively new to us, we were hoping for some guidance on authorship.

The ‘Submitting a paper to JOSS’ guidance indicates that it is the authors who assume responsibility for deciding who is credit as a co-author. Would it be possible to clarify what you mean by an author in this instance?  It is our understanding, from the journal website, and from articles written by editors such as Daniel Katz that, in this instance, an author is someone who has written the software? i.e. the software author.

The submission guidelines further say that non-code contributors may be included as authors/co-authors. Does this mean it is within the guidelines for someone to be included as an author/co-author in the JOSS paper/citation, if they have not contributed directly:

• written the software code
• provided feedback or direction on the software code itself

But have contributed indirectly to one (or more) of the following:

• given feedback on or edited the JOSS summary manuscript

• been involved in the conceptualisation of the vignettes, for example:

  • provided sample data/inputs used in vignettes
  • provided scripts applying the software
  • reviewed the vignettes

• used the code

• provided feedback or direction on the wider research project that the software is a product of

In our case, our Zenodo archive, for one of these pieces of software, credits some software authors who are GitHub Contributors (direct contribution as per the above definition), but also a project leader and a researcher who are not GitHub Contributors (indirect contribution as per the above definition).

If possible, we would like to list everyone as an author on the JOSS paper/citation, giving appropriate accreditation for everyone’s role. I am making this enquiry to establish the feasibility of this? i.e. whether JOSS requires authors to be direct contributors or if indirect contributions are sufficient? Clarification would be much appreciated for our team to understand software authorship and attribution, compared to more ‘traditional’ research paper authorship and attribution.

Thank you and kind regards,
Annie Visser-Quinn (on behalf of the Water Resilient Cities project team)

[danielskatz]

In my opinion, all of the people who contributed to the work to be published (the software project and the paper) should agree on who should be authors.

[danielskatz]

In terms of your indirect contributions, my personal opinion (not a JOSS opinion) is that, by itself:

• giving feedback on the manuscript may not be sufficient to be an author
• editing the manuscript probably is, as long as it's not just fixing typos
• being involved in the conceptualization of the vignettes is sufficient
• using the code is not sufficient to be an author - all users cannot be authors
• provided feedback on the wider project is not sufficient unless the feedback is contributing to the software product

Again, these are just my personal opinions

[arfon]

In my opinion, all of the people who contributed to the work to be published (the software project and the paper) should agree on who should be authors.

👍 agree.

I also agree that some of the indirect contributions you suggest could warrant authorship but being a user of the software doesn't seem appropriate?

[nastaran62]

Hello, Is the supervisor of the main project who has provided all facilities and supervised the main research and gave feedback for the software, but not directly contributed to the software, should be in the author list?

[danielskatz]

In my opinion, yes, since you said "supervised the main research" and "gave feedback for the software"

[platipodium]

The ideal place to do this would be a CRediT (Contributor Roles Taxonomy) author statement, clarifying the roles of each contributor. https://credit.niso.org

JOSS could consider enforcing this, as many other open access journals do or encourage.

[danielskatz]

As someone involved in CRediT from near the start, I don't believe the 14 roles map very well to software project activities.