search

aaronpeikert / reproducible-research

A Reproducible Data Analysis Workflow with R Markdown, Git, Make, and Docker
https://psyarxiv.com/8xzqy/
Creative Commons Attribution 4.0 International
122 stars 17 forks source link

add nutshell figure #29

Closed aaronpeikert closed 4 years ago

aaronpeikert commented 4 years ago

nutshell closes #26

aaronpeikert commented 4 years ago

Now ready for merge.

brandmaier commented 4 years ago

Very nice! Two things:

A solution to both problems would be to choose a different line type/color/width for the dependency arrows.

brandmaier commented 4 years ago

maybe long-dashes as line type for dependencies?

Otherwise, it's great.

brandmaier commented 4 years ago

one more thing: if we stay with renaming as suggested earlier, "clean.R" should become "process.R" and dependency should include "manuscript.Rmd".

aaronpeikert commented 4 years ago

one more thing: if we stay with renaming as suggested earlier, "clean.R" should become "process.R" and dependency should include "manuscript.Rmd".

@cjvanlissa uses the name prepare_data.R I like process more (because it is more generic) but I am in a mental pat, you have to decide...

aaronpeikert commented 4 years ago

Very nice! Two things:

  • the arrows for time (left-hand side across git versions) is the same as the dependency arrow. Let's make them distinct
  • The dependency arrow is percetually very similar to the git version frames. Can we make them stand out a little more?

A solution to both problems would be to choose a different line type/color/width for the dependency arrows.

I overlooked the dashed line advice for make dependencies, now it looks like: https://github.com/aaronpeikert/reproducible-research/tree/2f5632515619d850c503314ae4fce98362cc53e7

cjvanlissa commented 4 years ago

I think clean, process, and prepare are all fine - but for readability I'd add a noun, like "_data".

It's also important to consider that the clean_data file might not be the entry point to the project for secondary analyses, so I also add a run_me.R or simply the Rmd as primary entry point.

aaronpeikert commented 4 years ago

I think clean, process, and prepare are all fine - but for readability I'd add a noun, like "_data".

I agree. @brandmaier your call: clean_data, preprocess_data or prepare_data?

brandmaier commented 4 years ago

great work, @aaronpeikert! Let's stay as close as possible to worcs, so prepare_data it is.