in docs/notebooks/ factor out shared data download steps into separate notebook? Or .py?

[mccalluc]

Would it make sense to separate out the data directory creation and downloads to a separate notebook, if multiple notebooks use the same sample data? And then just have a link in the main notebooks to the prerequisite notebooks? Right now some notebooks are missing steps, and they can't just be run from the top to the bottom.

... Alternatively, the shared download code could be handled as just a plain .py file in this directory, and it could actually be run from each of the examples.

• Plus: the notebooks could actually be run from top to bottom. (I know I don't always actually real the instructions.)
• Minus: in terms of explaining the steps, we'd be hiding some of the details.

---

... but if the primary purpose of these notebooks is actually to support https://vitessce.github.io/vitessce-python/widget_examples.html, then we don't want references to files that only make sense when running the notebook: We really need everything to be right there, in each notebook.

There are ways to run one notebook from another, but what I think we'd really like is some kind of transclusion of the content, and I don't see that.

[keller-mark]

My vote would be to keep the downloading in the notebooks, rather than hiding things in .py files

[mccalluc]

My vote would be to keep the downloading in the notebooks, rather than hiding things in .py files

@keller-mark : Do you also think it makes more sense for each notebook just to be self contained (ie: no data-download notebook that would need to be run first) rather than having a separate download-files.ipynb? Shouldn't be a huge task just to fix the problems in the nodebooks.

[keller-mark]

I would prefer the notebooks to be self-contained so that it is easy to point people to a specific one