Content beyond the paper

[nuest]

I just came across an interesting and educational discussion on GitHub on how to best install Python for data science. Maybe we need to start a Wiki or website with all the stuff that cannot make it into the paper, so I'll collect that information here. Contributions welcome!

• installing Python for data science: Use miniforge to get python rocker-org/binder#39 (comment)

[vsoch]

@nuest I added a comment here rocker-org/binder#39 (comment) and I'm not sure if you were thinking about this, but it would be a fun project to actually do what I suggested, e.g.,

1. come up with list of general things people do in containers
2. find recipes in the wild that do them, reduce to dummy examples (e.g. "just installing pythong)
3. come up with some criteria to rank on (size, speed, usability, etc.)
4. have different categories and then best practices to go along

Warranted that things change, this still feels like it would be potentially useful or fun!