Ditch Excel and Use Julia Data Frames

Julia and excel are fundamentally different.

In Excel, the logic/analysis part is on the same level as the data. The data and operations done on it are a single file. A datum is in a cell, and a formula is in a cell. You can move formulae around like you can move data around. You are viewing and operating on the data/logic in the same context. When you open an excel document, what you see right away is the culmination of all analyses done on the data.

With julia, the data and logic are separate entities. You have to first import data[1]. The result dataframe is created by explicitly running cells/scripts on the import. When I view the dataframe, I cannot just point-click-edit it. Any edit I make is a separate line of code to run. Then the whole script is run again over the whole dataframe to get the new result. This is opposed to excel which creates dependencies on a cell-level and live-updates the spreadsheet with each edit, but without processing the whole spreadsheet.

Since logic is its own entity (script) in julia, it is easier to write long and more complex processes: things you cannot easily do in a single-line formula parallelized over cells in excel. Also, since the processing and editing are explicitly spelled out in a script, it is easier to keep track of changes and reason about logic. So they both have their advantages.

[1]: You can even do that in excel. Import csvs, databases etc.

In all honesty, no one (at least, no one in the real actual world) wants alternatives to things that already work. Give me a standard that's good enough and we'll see how to deal with edge-cases. That's what internet is for, dammit.

Yes, I know a million different better pieces of software exists for my peculiar task. No, i don't care. I have other things to do in my life than babysitting a damn computer.

Are there any existing FOSS tools that let you do a similar thing to what Excel does in the data layer but with Julia (or another programming language)?

i.e. I’m in my cell and want to type my Julia right in that cell and watch it live update when I hit enter, similar to excel

Yes, I think the best approach is to combine tools. But what I think is a big problem is that people stretch the practical application of tools like Excel too far. They insist on using Excel far beyond its practical application.

Personally I never use spread sheets, other than some simple plotting. I often prefer using a simple CSV viewer instead. I find it more practical to use tools which work natively with CSV, as that is the format one will be using in Julia, Python, R or whatever programming tools you use.

Julia is just as easy to install as Excel. There is literally no difference. If you cannot do a one click install, then how are you supposed to figure out how to make a Pivot table?

Of course learning to code is a higher barrier. I don't think anybody challenges that. But as you begin using Excel much beyond the trivial I will argue it is actually easier to learn how to code.

Code has the advantage, that it is very easy to document and explain with text. Explaining how to do things in a Excel through text is not easy. You will need images, videos etc which are not searchable. You cannot copy-paste and image to repeat an operation from Excel. With code in contrast you can copy and paste in code you find in blogs, help documentation or anywhere else.

Basically is much easier to reuse and modify. Once you learn how to code you can advance your capabilities much quicker.