search

nsidc / qgreenland

Source code for generating the QGreenland package hosted at https://qgreenland.org/
https://qgreenland.readthedocs.io
Other
36 stars 9 forks source link

Distinguish code/software and data citations and update authorship #702

Closed MattF-NSIDC closed 1 year ago

MattF-NSIDC commented 1 year ago

Follow-up on conversation here: https://github.com/nsidc/qgreenland/pull/701#discussion_r1279993889

How should we cite our work products under this grant?

Twila is the PI and long-term responsible party for the work under this grant and cited as the first author of QGreenland on our doc's citation page. How should we communicate the authorship of the code? Should we do that in the DOI entry (CITATION.cff)?

Or should we treat QGreenland the package as "data", citing Twila as first author, and the source code (this repo) as "software", citing Trey and myself as co-equal first authors? This seems fairly in line with other projects I've worked on.

@twilamoon-science @trey-stafford

MattF-NSIDC commented 1 year ago

We can create a Zenodo DOI for the package separately from the software, too. The upload limit is 50GB.

twilamoon-science commented 1 year ago

This item has been on my mind. What steps have you already taken to finalize this? In particular, I think that the software citation really should have @MattF-NSIDC and @trey-stafford as first authors (we can add an asterisk so that it is a shared first authorship). I'm not sure that we functionally have more than 1 think to cite with v3, so this would be the v3 citation. When we someday write a description paper for QGreenland, I can lead author that.

MattF-NSIDC commented 1 year ago

@twilamoon-science thanks for sharing your thoughts! We've not taken any concrete steps, have just been chatting so far. Here are my thoughts coming out of those chats:

I think our situation is not identical, but is analogous to a project which produces software that produces a dataset: QGreenland software produces the QGreeenland Core dataset (and eventually, others). Often the dataset citation for those other projects will have PI as first author, and the software citation will be the relevant developers.

QGreenland Core is like our dataset, so I think you should be lead author on that. And I think it makes sense for the dataset citation to be the primary citation we display on ReadTheDocs, which is primarily aimed at users. The real final product of this work (so far) has been the dataset, and the software is a means to get there.

The software citation I feel can/should live only on GitHub/Zenodo. Or we could display a citation for QGreenland Core and QGreenland software both together on ReadTheDocs.

What do you think?

MattF-NSIDC commented 1 year ago

Notes from meeting:

MattF-NSIDC commented 1 year ago

Shouldn't close this until QGreenland the dataset has a DOI. WIP.

MattF-NSIDC commented 1 year ago

https://zenodo.org/record/8247896