I am concerned about a possible tension between the Don't Repeat Yourself rule and Reproducible research, and wanted to hear some opinions from this community. In particular, I want to know what people think about the use of personal github repos/packages to satisfy the DRY rule. I'm thinking in particular about packaging code used for common data wrangling needs, not necessarily to train a model, and then code used to describe the analysis in Rmarkdown, for example, code to display session info using something like kable rather than just print. More directly, should code submitted to a journal include only CRAN packages? And, is it impolite or improper to require other scientists to install personal packages from github or elsewhere?
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.