Submission to rOpenSci?

[iqis]

Is submitting to rOpenSci a good idea? The platform has hosted many scientific packages through a rigorous peer-review process, example here. The dev guide is very thorough and helpful.

psyphr can fall into the data munging category.

[wendtke]

Yes, I think so! My previous R professor, Brooke Anderson, is an Associate Editor of rOpenSci. I see her for a dinner next Tuesday and will ask her about this.

[iqis]

Wow! If she could provide some guidance in the package development from early on, it will be awesome.

Also I've recently realized that many many wonderful R users and developers are women; this is something that's quite unique about R, as a computer language; and all R conferences or meetups I've been to has a healthy gender mix, a contrast to even those of JavaScript. I've also just stumbled across this article today, which did a really good analysis on the phenomenon.

[wendtke]

Yes! Let me know if you have any specific questions you would like me to pass on to Brooke tomorrow night.

Brooke is actually the reason that 1) I know what I know about R; 2) I love R and am so eager to learn; and 3) I attended the Uncoast Unconf. I owe so much to her. She is a role model to me, and I hope to do the same for other students, particularly women and psychologists/social scientists, in the future. I never want to be considered excellent in a certain skill (e.g., R programming) because I am a woman, but, rather, be an R programmer who happens to be a woman.

[wendtke]

Roger Peng's blog post about publishing software

[wendtke]

Fast-track review for JOSS

[wendtke]

G.B.A. (06/23/2019) on rOpenSci review timeline and connection to CRAN:

"The [peer review] timeline definitely varies from package to package, and some depends on how long it takes to find reviewers and how long it takes your group to make changes in response to the reviewers. For most, peer reviews are in within about two months of the submission, I think. Usually then there's a more interactive process of responding to those reviews and then the reviewers commenting back on those responses. The whole process is open, so the best way to get an idea is to look through some previous or in-progress reviewers: https://github.com/ropensci/software-review/issues

Yes, ROpenSci packages can definitely also go on CRAN. I think the official recommendation is to finish the ROpenSci process before submitting to CRAN, in case you make a lot of changes based on the review process. Once you get something on CRAN, you'll want to avoid making a lot of changes to things that affect the users' interface, so you won't break other people's code (e.g., function names, arguments to functions)."

[iqis]

Thanks @wendtke. Great information. I think this has cleared a lot of cloud.

[wendtke]

Update from discussion with @geanders on 20190701.

• Provide strong justification for why psyphr fits within scope of rOpenSci, specifically in the data munging category.
• "Complete" submission might include support for full MW 3.2 output suite; exploratory analysis and visualization options; full vignettes + tutorial via packagedocs.
• @geanders will recuse self from peer review process and be listed as author or contributor

[wendtke]

See #47 for full timeline; refer back here if needed