Submitting astropy core lib for pyOpenSci review

[Pey Lian Lim]

To whom it may concern,

As you might have known, https://github.com/astropy/astropy-APEs/blob/main/APE22.rst (Astropy Affiliated Packages with pyOpenSci) has started a process where new Astropy Affiliated Package submissions would go through pyOpenSci review process.

Furthermore, Sunpy also went down a similar path and they even submitted sunpy package itself for review (ref: https://sunpy.org/posts/2024/2024-01-24-pyopensci/).

Along that vein, I thought it might be uncontroversial to have astropy core lib reviewed there, which would also serve as a good example if we ever want to shepherd (note: we should never mandate) Astropy Coordinated packages for reviews as well. Thus, I opened https://github.com/pyOpenSci/software-submission/issues/251 , which I admit was a little too haste and I apologize for it.

Affiliated Package Editors requested that I also post here to collect more feedback on whether we should actually proceed or not. Options:

1. Address remaining todo on https://github.com/pyOpenSci/software-submission/issues/251 and make it ready for review.

2. Close the issue without further action.

Please let us know what you think.

Thank you,

Pey Lian

[Erik Tollerud]

The coordination committee discussed this in light of the Slack discussion (which seemed to be stalled without a consensus being reached), and are thinking to go ahead with the review with the caveat that we want to be sure it's something like the detail level Sunpy got, which according to the Sunpy devs was helpful but did not create any undue burden on the maintainers.  As long as the same is true here, we will proceed.

---

Erik T

--
You received this message because you are subscribed to the Google Groups "astropy-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to astropy-dev...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/astropy-dev/3b5a5639-2519-4ba8-85de-af2d75821988n%40googlegroups.com.

[Pey Lian Lim]

Hello again,

The pyOpenSci review has been going well, which not much overhead for us. Now I have a question for you, should we add a "paper.md" to the astropy core repository to also submit to JOSS (it comes "free" with pyOpenSci when you have that extra file). I couldn't emulate sunpy (https://github.com/pyOpenSci/software-submission/issues/147) because it already was accepted to JOSS (https://joss.theoj.org/papers/10.21105/joss.01832) by the time it did pyOpenSci submission.

• Do we really need JOSS listing?
• Even if we don't, perhaps adding a "paper.md" is low enough barrier anyway that we should just do it?

Any opinions? Thanks!

[Brigitta Sipőcz]

Hi,

If you must, at least call it joss_paper.md as opposed to "paper.md" as what that suggests is that's the one and only paper we have. IMO we already have 3 actual papers, zenodo DOI for each releases, and a citation policy all of which should take priority. 

Cheers,

 Brigitta

To view this discussion visit https://groups.google.com/d/msgid/astropy-dev/ff5c923d-989d-4d91-95b0-a6e26996b0ebn%40googlegroups.com.

[hamogu]

We don't need a JOSS paper - we published astropy  in other ways several times are Brigitta listed.

And it's by no means required by PyOpenSci - it's an offer to the community since JOSS will accept the PyOpenSci review and not review again, saving the authors and the community (so a package doesn't have to be reviewed multiple times) work for packages that want both a PyOpenSci listing and a JOSS publication. I don't think that applies to astropy, so we should simply untick the box "JOSS submission" and move on.

Moritz

[Pey Lian Lim]

Hi and thanks for your inputs! I will remove JOSS request from the submission and let them know.