[codeptualize]

This is awesome! Starting it up from the terminal and going to the url was not super convenient for quick tasks and scripts, this is much better.

Edit: one thing that would make it even nicer is if it would remember the folder and/or recently opened notebooks.

[hugobarauna]

Recently opened notebooks are on the roadmap. ;)

[codeptualize]

Amazing! Can't wait.

[redrobein]

Why would you use a notebook for quick scripts though

[codeptualize]

I often need to inspect/transform some JSON/CSV/text/whatever data, move around/organize a bunch files, and similar things. I find that notebooks work great for that as you don't need to set up an environment, you can easily collect and rerun your scripts if you do need them again, and the output options are great for debugging and looking at data.

[powersurge360]

I know this is a Livebook love-fest, but you may also consider setting up org-mode sometime. There's a lot to love about org-mode but to stay on subject, you can write your documents in a plain text syntax akin to markdown and any of the code blocks that you add in as examples are _executable_. And you can take the output from one and pipe to another.

At work, for example, we have to hit a remote API and the secrets are stored in AWS secrets manager. I have some documentation that has a block describing how to get the secrets and some other documentation elsewhere describing how to issue a post request and I can pipe the output from one block into another and receive the output either in the same file or put it somewhere else.

The one downside with this extreme flexibility is I sometimes catch myself squirreling away useful functionality in my own notes instead of adding it in a utility CLI or similar where it can be shared with the team.

If having access to a scripting language is duct tape, piping output from one script to another in this way in a single file is like having inter-dimensional duct tape.

[heresie-dabord]

Visidata† and Datasette†† are both excellent for rapid exploration. Both are available via Python pip.

† _ https://www.visidata.org/

†† _ https://datasette.io/