Emily - Chapter 4

[RohanAlexander]

Chapter 4

Theory

• “Workflows exist within a cultural and social context, which imposes an additional reason for the need for them to be reproducible” – this is a beautiful sentence
• ‘But it can be alternatively considered as: “Think an awful lot, mostly read and write, sometimes code”.’ – another absolutely beautiful sentence
• I think the discussion of reproducibility here is excellent for science and academia. However, something (in fact, part of my ungiven Toronto Workshop on Reproducibility talk!) that I think could use more emphasis is the practical benefits of reproducibility – namely, reuse. Students may be less inspired if they think reproducibility is to “help other people” but understanding how these same skills help “future you” work more effectively might be more of a hook

Technology

• Very nicely curated set of tools to promote reproducibility. I know you don’t want to overwhelm beginners but I wonder if there’s room for an optional section at the end of the chapter or in the Appendix for taking things one-step further by discussing some less tools-based aspects of reproducibility like, e.g.:
• More advice on good documentation as part of reproducibility
• Writing robust code (e.g. parameters vs hardcoding)
• Data validation / stating data assumptions?
• For Code Review tutorial, I wonder if two other links might provide useful context:
• Note that it’s possible to conduct code review in GitHub with an example link. I really think this is a place GitHub shines as a platform and something students might enjoy
• Note that we can do “rubber duck debugging” and try to review our own code as if we are a neutral third party