Related resources

[sdruskat]

An issue to collect related resources that we may or may not find a better place for

• Mentioned by @danielskatz: https://eos.org/opinions/credit-where-credit-is-due

• Authorship in Open Source (Google Open Source Casebook)

• Authorship definition (partly based on copyright) by U Louisiana: https://vpresearch.louisiana.edu/faq/who-author-software/177

• @danielskatz' blog post "Contribution vs credit vs authorship for software

• Related academia.stackexchange discussion

• Funding codes of conduct/guidelines may include authorship definitions, e.g., on p. 18 of the DFG guidelines.

• Registries paper has a section on establishing authorship policies: https://peerj.com/articles/cs-1023/#p-45

• Milewicz, Reed; Pinto, Gustavo; Rodeghero, Paige (2019): Characterizing the Roles of Contributors in Open-Source Scientific Software Projects. 2019 IEEE/ACM 16th International Conference on Mining Software Repositories (MSR), 421–32. IEEE.

• Authorship analysis (detection, sort of random)

  • https://doi.org/10.3390/math10152572
  • https://doi.org/10.1016/S0167-4048(97)00005-9
  • http://hdl.handle.net/10292/3496
  • https://doi.org/10.48550/arXiv.2001.11593

[danielskatz]

The ESIP group that wrote this has been thinking about what roles should be reflected by authorship and which by other means

[jcolomb]

Some earlier discussion at FORCE 11: https://demo.hedgedoc.org/WWA2OwbbSeiVXkTkLSwadA.

Additional links from that document: JATSxml implementation: https://jats4r.org/credit-taxonomy/

Brainhack example: http://brainhack.org/brainhack_jupyter_book/contributors.html

other example with values 1 to 3 to code how much work was done: https://raw.githubusercontent.com/int-brain-lab/paper-behavior/master/authorship/contributions.png codemetada discussion: https://github.com/codemeta/codemeta/issues/240

first guide on Skilling in general for CitSci projects and data - https://cs4rl.github.io/guide/#/

[sdruskat]

Thanks, @jcolomb! Very useful links!

[jcolomb]

you are welcome. There is another thing that could be considered: https://www.nature.com/articles/s41467-023-37039-1

Here the idea is to simply "say" what people did, in plain English. It is not very computer readable, but it has the advantage to take the uniqueness of each project into account.

(I personally have not yet made my mind about this strategy)

[sdruskat]

Here the idea is to simply "say" what people did, in plain English. It is not very computer readable, but it has the advantage to take the uniqueness of each project into account.

I think this could be a good fallback/additional info, where a taxonomy doesn't work or isn't precise enough.

[sdruskat]

From the meeting notes:

• "FAIR Impact project is producing framework for software metrics that is defining Community (effectively a group defined by the goal, e.g. software creation team)/Actor/Roles for research software - we could crosswalk this to the Contribution role taxonomy"

[sdruskat]

This is one interesting case (authorship of Linux kernel patches) that we could use to test our work against. Could our definition and roles help solve this/suggest a way this should have gone?

https://ariel-miculas.github.io/How-I-got-robbed-of-my-first-kernel-contribution/