Mirror, mirror on the wall, which episode order is the FAIRest of them all?

[katrinleinweber]

I don't think it's feasible to reorder the lesson now, but for a second version of this lesson, I have a feeling that a FAIRer and/or more scientifically valid order of work could be taught:

1. creating a barebone package skeleton (currently episode 4),
2. packaging raw data, cleaning scripts and tidy data (see Tidy the inflammation dataset #5),
3. functionalising & packaging code for the analysis (currently episode 3) of that tidied data, and
4. writing vignettes (potentially another lesson, or referring to r-novice-inflammation/07-knitr-R.Rmd, rr-literate-programming, or r-novice-gapminder/15-knitr-markdown, etc.)

[lwjohnst86]

Yes, that order makes a lot more sense...

[katrinleinweber]

Another idea is being explored in #19: add testthat::expect_ations immediately after writing function code & separating it later when packaging.