[sitkack]

This is for the room. You understand all of this stuff.

> Our goal isn't to replace production ETL pipelines etc.

Why not? The OP is conflating accidental complexity in current spreadsheet design, there is no reason that you couldn't get a graph of code. Everything becomes production, or mission critical at some point. There is no difference. Production ETL pipelines exist because spreadsheets have historically only run a single machine, hence the whole "big data" definition.

I am absolutely sure you are familiar with https://en.wikipedia.org/wiki/Felienne_Hermans and ZUIs, your work is the intersection of both.

I kinda hope you are pandering to the HN gestalt against spreadsheets. But yes, please do introduce fork/merge/diff (you already have-to to support multiple simultaneous users). As well as property testing.

Aside, I am kinda losing my _, that you pulled all of this off. I absolutely love it. Amazing, hopefully we have a future to thank you.