[xj9]

i think spreadsheets could represent an interesting future programming paradigm, but excel is not the tool that'll do it.

[peterburkimsher]

I feel like the next major advancement in spreadsheets is nested hierarchy.

Specifically, iTunes (especially the older interface from iTunes 5 to 10) allowed for Playlist Folders and Smart Playlists.

Workbooks can have several sheets, but not in a nested structure. Same with SQL databases and tables. If I want to organise data (e.g. Country > Social group > Friend name/email/phone number) then it's easy to do in iTunes but difficult with typical Contacts apps.

Web browser bookmarks are pretty good, but the end object has only 2 fields (name and URL). So what I do is I have a fake-radio track object in iTunes, with all the extra tags. Double-clicking it loads a page on localhost, which launches a PHP script that then opens a URL in my browser (for the friend list) or document.

[goatlover]

Notebooks are the next-gen spreadsheets, except they use programming languages instead of formulas.

[hobs]

And they require so much additional setup and processing power that they will never supplant spreadsheets.

[CJefferson]

Jupyter is a joke compared to spreadsheets, it doesn't auto-recalculate when you edit cells.

[wibblewobble123]

Observable does. So does Julia’s Pluto.jl.

You’re not paying attention to the space but recalculating notebooks are gaining popularity.

[CJefferson]

Things are getting better. I was aware of them, but jupyter is the 600 pound gorilla in this area.

I wouldn't use observable, as you are tied to their web service, and I've had too many companies close to use such a service for important data. Pluto is promising, but (I believe?) Only supports Julia, which is growing but still not that popular, or known.

[goatlover]

True, but you have access to whatever Python libraries are installed, and your output is web-based. So you can potentially output anything a browser can display, including interactive 3D graphics and widgets. You also have access to the command line in Jupyter, so you can run whatever shell commands you like.

[FabHK]

Not only that, it doesn't even keep track of the calculation graph, so it doesn't know what needs calculating, or in which order to calculate.

[TremendousJudge]

yeah, and also it's 1-dimensional; you can only go up or down, never to the right

[goatlover]

Because the data is inside a data structure like a table or dictionary that you can output, iterate and map over however you like with the various Python and library methods or functions.