Joss paper

[rsh52]

Description

This PR provides a draft for a submission to the Journal of Statistical Software (JOSS).

Areas for improvement:

• Design Image, the current one is a little busy and I found it difficult to generalize what we're trying to achieve without getting too in the weeds or losing context
• Comparison table with other available packages
• Acknowledgments
• Confirmation of authors in YAML
• Conflict of interest section is a direct copy from the Funding section of the README

At present we are just over 800 words, with a max of 1000 per the submission criteria.

Recommend viewing the paper.md file outside of this PR so it renders in the GitHub display.

Useful References

• Submission Criteria
• Similar submission for secuTrialR

Proposed Changes

List changes below in bullet format:

• Draft paper.md and paper.bib under the new paper/ folder
• Add paper/ to .Rbuildignore

[skadauke]

Looks good! Is the paper ready for my review?

[rsh52]

Looks good! Is the paper ready for my review?

I believe so, at least a first pass barring the "Areas for Improvement" I mentioned.

[skadauke]

Hi, I agree with those Areas of Improvement. We should also work in the following:

• Comprehensive documentation in the form of vignettes and articles that explain in detail the data structure and design decisions of the REDCapTidieR package
• Meant to be deployed in production environments - so we built an extensive automated test suite etc. (I guess this could be added to the table - only REDCapR and REDCapTidieR apply?)
• This effort has earned us OpenSSF Best Practice award
• Support for labeled data
• Utility functions for common workflows such as Export to Excel

[rsh52]

Hi, I agree with those Areas of Improvement. We should also work in the following:

• Comprehensive documentation in the form of vignettes and articles that explain in detail the data structure and design decisions of the REDCapTidieR package
• Meant to be deployed in production environments - so we built an extensive automated test suite etc. (I guess this could be added to the table - only REDCapR and REDCapTidieR apply?)
• This effort has earned us OpenSSF Best Practice award
• Support for labeled data
• Utility functions for common workflows such as Export to Excel

@skadauke Awesome thanks! Let me know if the recent changes address these additions.

[ezraporter]

This looks great! Added some comments to particular spots but broadly:

• We think about how well-defined the terms in the comparison table are
• We should emphasize even more how life saving REDCapTidieR is for working with longitudinal and repeating events

[rsh52]

@skadauke Taking another read through the submission criteria just two things that jumped out to me.

We may want to convert the Figure 1 caption to text or think of a way to consolidate it so it fits into the link text/brackets:

https://joss.readthedocs.io/en/latest/submitting.html#figures-and-images

I didn't see any recommendations on captions for tables, so I think we can keep that as is.

It's also possible, but I don't personally think necessary, to include a GitHub action for rendering the paper.

[rsh52]

Proposed change for production readiness, at the very least we can lean on these numbers/metrics for the wow-factor.

Current:

REDCapTidieR is built with production readiness in mind. It builds upon REDCapR, which contains an excellent test suite, to make API calls, and includes an extensive automated test suite and ample documentation through a pkgdown site [@redcaptidier_pkgdown_cit]. It meets the rigorous requirements of the OpenSSF Best Practices Badge [@openssf_cit], which certifies open-source projects that adhere to criteria for delivering high-quality, robust, and secure software.

Proposed:

REDCapTidieR is built with production readiness in mind. In addition to an extensive 98% coverage test suite, REDCapTidieR execution is evaluated against 15 test databases that cover many complex configuration scenarios. It also provides ample documentation through a pkgdown site [@redcaptidier_pkgdown_cit]. It is also built on top of REDCapR, which contains its own extensive test suite, and evaluated against an additional 26 test databases. REDCapTidieR meets the rigorous requirements of the OpenSSF Best Practices Badge [@openssf_cit], which certifies open-source projects that adhere to criteria for delivering high-quality, robust, and secure software.