What is the culture of the lab / group around feedback and code collaboration?
- Different standards will apply if code review is a regular routine vs. multiple folks working together on a coding project vs. PI or other mentor being asked to review code written solely by one person prior to submission.
What are the use cases?
- Is the code being used for one-off purposes, or will it be re-used?
- Who is going to maintain it?
- Is it being included with a paper or developed as software for the community?
- This matters in terms of training and time as well as what coding standards are expected by the community.
What are some practices that PIs can implement / adopt?
- training and clear expectations for lab members
- guidelines for code to be modular and testable
- guidelines for code reviews and collaborative work (which GitHub workflow are you going to use?)
- continuous integration and linting to automate tests (and also allow self-code-review and quick feedback)
- handling author/code contributions and whether to use public/private repos