Rich FitzJohn

I think this is a fun thought experiment at least, and could be super useful. I wonder though if it's worth resisting the urge to make the language _more_ complex...

> Facilitate encryption for researchers working with restricted data, e.g. containing personally identifiable information With apologies for self promotion, I've been working with Jeroen's excellent packages to write some user-friendly...

My opinions are mostly variations on the theme that python packaging is irreparably broken. An option to make the vendored version skippable, or in the presence of a reproducible example...

It didn't work for us, with workflows we have. Currently it does. I am not incredibly inclined spend time explaining why we made choices that work for us by someone...

Yes! @nicolewhite might possibly be interested too with her port of some string handling things: https://github.com/nicolewhite/pystr

Though in terms of @vsbuffalo's original topic, I now endorse meaningless filenames backed by a lookup to a key-value store for the metadata.

Oh dear god yes. This bit is so hard to get right. Since I started on this I think there are a couple of other, more complete, attempts at static...

I think that @aammd had some ideas for this, too?

@rich-iannone; do you have a best-practices way of doing this?

Thanks Rich! Mostly for my own information later, this thread has some useful hints: https://github.com/ramnathv/htmlwidgets/issues/95 The phantomjs approach looks promising but heavy.