I'm working on an R project with a bunch of data and code files, with thousands of lines of code spread across a bunch of files. I've created a package to hold everything as that relieves the stress of keeping track of it all.
I'll admit I don't do this as often as I should. When I do, I document all the functions and use attributes to document the data series.
This is the kind of thing #rstats gives you that isn't appreciated or even understood by the critics.
Add comment