Packaging research software and its dependencies

[rbirkelbach]

What should/will the learning objective(s) be?

Learn about packaging R/python software and its dependencies using some of the following tools:

• packrat
• miniCRAN
• conda
• docker

What do participants need to know / install / bring:

Install the tools mentioned above and bring a notebook.

In order to be accepted as an event, each suggestion needs to:

• receive 3 or more positive reactions
• be assigned to someone who will lead the event

then, a PR can be started against the _posts folder to include a YYYY-MM-DD-event-title.markdown file to actually schedule event.

[katrinleinweber]

Would this mostly be about dependencies? If so, we can keep #28 separate.

[jkreklow]

Sounds to me as if it's quite the same thing. I agree to @rbirkelbach that it would be okay to make an event showing both the R and Python version for this as the concepts are pretty much the same. What do you think?

[david-morris]

Maybe we should focus on one packaging tool per event. The differing philosophies between different packaging systems seems like they create different concepts to master and skills to learn.

[rbirkelbach]

I think it's good to keep this issue separate from issue #28. Maybe it's good to make it into two or three issues. Either teach packrat/miniCRAN and conda together and have another event with docker or have a session with packrat/miniCRAN, one with conda and one with docker. What do you think?