VSCode Adds Notebook Support (code.visualstudio.com)

Could just use Code - OSS that removes telemetry.

Emacs org-mode allows taking notes and having code blocks you can execute in the notes.

As someone who doesn’t use emacs though, I personally would love to see something like that with markdown/latex/notebooks. I guess you can already sorta do it with Jupyter and things like this.

Would be interesting for a more note-oriented tool to do this though.

Notebooks are for the most part markdown with embedded executable code snippets.

Outside of datascience they can be used for exploring API's, databases, prototyping libraries etc.

But for general note taking, markdown would be better.

Tried it, not particularly useful. Notebooks are code with embedded markdown blocks; it shines when you want to run the code in the notes. Otherwise, it has no more feature than a regular plain text file, only more formatting boilerplates.

I wrote a BASIC interpreter in literate org-mode Scheme once. I find it helpful to organize my thoughts in notes, add code, and turn that code into a working program via "tangling".

I've had an idea for a while to make a dashboard style notebook. Basically something that can query a few internal systems (bug tracker, code reviews, etc) and produce of dashboard that basically shows what need my attention.

The Github and REST notebooks in the article imply something like this is possible.

I used Notebooks for notes for a while. It's useful. The choice is essentially between Markdown with inline code, or Notebooks, which are Code with markdown inbetween. Markdown focusses on the text, and the code is incidental. You usually don't want to execute it. For Notebooks, executing the code is the focus.

I personally thought I would really value the ability to execute the code, but I found out that it wasn't useful to me. The kind of code I write in my notes is rarely executable. It misses library includes, boilerplate declarations, or sometimes isn't quite syntactically correct.

The overhead of trying to make each (or even most) cell in a Notebook executable takes away from the "whatever is most expressive" mindset I have for notes.

There's one big caveat though. Notebooks are much better for sharing with others. If the code is complete, and literally executable as-is, that is much more useful to others I show the notes. Most of my private notes don't have this constraint though.

I'd probably use notebooks if I was writing something I would possibly publish in the future. Much easier to share it with others to get feedback that way.

Yes, they are great for developing modular python scripts, like for CLIs, where you might want to prototype some feature without importing the entire app.

For example, I'm working on a cli for a client that ingests a bunch of spreadsheets and processes the data in an SQLite db. Pretty lightweight, but it takes a half hour to run. Using Jupyter notebooks to import some classes and iterate on SQL queries without needing the entire corpus of data is handy.

I did think they could work quite well as a test/documentation framework. So as you develop the tests you write documentation, so other developers can use this to better understand the codebase, alongside how best to use the library/application.

It’s also great for writing/developing tutorials because you can easily mix rich text, code and results of code.

I’d say they are great for any exploratory task, and doubly so if that exploration ends with a write up of the findings.

That’s very often needed in any kind of data analytics, which is why notebooks are spearheaded by data science use cases.