uproot5

uproot5 copied to clipboard

If there is interest, it would be nice to add a CITATION.cff especially as there is already a Zenodo archive for uproot. I would be happy to make the PR.

It has been a goal of mine to get a CITATION.cff for Uproot. The only thing that has been keeping me from it is the process of defining the author list. Just as in awkward/CITATION.cff, I wanted to highlight the main authors and get full information for them. The list of GitHub userids that Zenodo pulls in doesn't include institutions, email addresses, or ORCIDs.

Do you have an idea in mind for curating the list of authors? I'm thinking I should acknowledge the fact that I got a lot of help on the remote backends part.

Also desirable for CITATION.cff (across projects, in my opinion): it should reference the versionless DOI for the repo. For instance, in Awkward Array, it's

https://github.com/scikit-hep/awkward/blob/8a4ea2b8c1c0cb2c92dab3565511bf5a717b7703/CITATION.cff#L4

which always goes to the latest version, whatever that is. It's therefore a stable identifier across time. For Uproot, this would be 10.5281/zenodo.4340632.

It has been a goal of mine to get a CITATION.cff for Uproot. The only thing that has been keeping me from it is the process of defining the author list. Just as in awkward/CITATION.cff, I wanted to highlight the main authors and get full information for them. The list of GitHub userids that Zenodo pulls in doesn't include institutions, email addresses, or ORCIDs.

Do you have an idea in mind for curating the list of authors? I'm thinking I should acknowledge the fact that I got a lot of help on the remote backends part.

Not really anything rigorous. I think on a smaller project like pyhf it is easier to define who the "development" team is and treat them as the authors. So I think that you'd either need to create an authorship criteria that seems fair to you, or just make selections on who has historically provided the help on the backends and then put that together. I assume that you mean contributions from across https://github.com/scikit-hep/uproot3 and https://github.com/scikit-hep/uproot4?

Also desirable for CITATION.cff (across projects, in my opinion): it should reference the versionless DOI for the repo. For instance, in Awkward Array, it's

https://github.com/scikit-hep/awkward/blob/8a4ea2b8c1c0cb2c92dab3565511bf5a717b7703/CITATION.cff#L4

which always goes to the latest version, whatever that is. It's therefore a stable identifier across time. For Uproot, this would be 10.5281/zenodo.4340632.

Yup. That's what I linked to above in

especially as there is already a Zenodo archive for uproot

and we do the same for pyhf.

Perhaps related to this issue: is there a recommended list of proceedings/papers to cite for uproot (also for awkward)? Should it be somewhere in the README?

Uproot doesn't have one yet (I have to draw up an author list), but Awkward Array does:

https://github.com/scikit-hep/awkward/blob/main/CITATION.cff

Perhaps related to this issue: is there a recommended list of proceedings/papers to cite for uproot (also for awkward)? Should it be somewhere in the README?

@jpivarski I think @jpata is referring to something along the lines of how pyhf's CITATION.cff includes other papers to cite in the references section and then we also ask for people to cite both the software and the JOSS paper in the README and the docs.

In that case, no, I don't think there's a good paper that would be an overview of Uproot or Awkward Array in general. Each is covering something specific, or was written at a time when the idea wasn't fully formed. I would just cite the software, not any accompanying paper.

This is done.