matt-dray / drake-egg-rap

:duck::egg: Demo/talk: {drake} for making a reproducible analytical pipeline for a stats report

LARPing as a data scientist in the public sector.

Visit my personal site, GitHub repos or Mastodon profile.

The latest realised showerthoughts on my blog are:

• The Aesthetics Wiki: an R Addendum
• Stop Jenny committing arson
• Assign abstracts to sifters
• Shtty R help from shtty AI
• Ellipses, eclipses and bulletses

The thought technologies presented here have erupted from my own brain.

drake-egg-rap's Issues

Questioning the structure of the repo

Is it confusing that the docs/ folder contains objects before you even run anything? They're there so that they can be served by github.io. Maybe they can stay there but the RMarkdown output can be generated into another folder.

Other files and folders like .binder and DESCRIPTION could be confusing. perhaps more emphasis on the doc tree in the readme. The additional files could be explained as well though.

Improve the README

Explain that it's a very basic example and point to links for other introductory material (e.g. Garrick Aden-Buie's latest talk).

Small code clean-up tasks

• library(dplyr) twice in functions.R
• The title in create_plot() is the result of a keyboard smash. This is a feature if people make the plan and then realise the plot title is wrong and then re-make it. Except I never mention this anywhere. Could update readme (see #1) and blog post.
• Don't need the namespace for render() in the plan