[gbrown]

Using a web browser as an IDE just seems like a solution in search of a problem. Also, maybe I just never spent enough time working with Jupyter, but it seemed to me that it encourages a sort of exploratory workflow, really well suited to teaching programming and data science. It was less clear to me how it could be used well in production environments.

I use Rmarkdown and Sweave to write homeworks for my students in a very Jupyter way. I also use them to generate data driven static webpages, procedurally generate production quality and easily formatted PDF and HTML reports. I also use them as a templating system for auto-generated model diagnostic emails. Perhaps I need to return to Jupyter to see what I'm missing, but I don't really know what purpose it would serve, or what kind of work it would make easier.

[takluyver]

Doing the interface in the web browser has its ups and downs: it makes some things more awkward locally, but it's easy to deliver the same interface remotely - e.g. a university can run a JupyterHub instance for a course, and students visit a URL and login. nteract is an attempt to make a notebook interface as a local application.

[gbrown]

For sure, I totally see how it's useful as a teaching tool. I don't see how it fits into a proper production system.

[carreau]

Different system and requirement need different tools, if you prefer text editor. Depending on your preference look at Emacs IPython Notebook, the Jupyter VS-Code extension, or Atom Hydrogen. That will basically let you select chunk of code and execute in a kernel. You do not have to use the notebook format, or the browser based editor.

JupyterLab also allow Rmarkdown-like workflow where code-blocks in a markdown document can be executed to display graph.

I believe the important part is to allow interoperability between different ways people want to work. You can't have 1 size fits all, and there are still a lot of work that can be done to cover some use case.

[takluyver]

This may depend on what you mean by 'proper production system'. It's definitely meant to complement, not replace Python modules, scripts and so on. I wouldn't write a web app in a notebook. But the LIGO team that discovered gravitational waves published a notebook demonstrating their data analysis.

I see it as useful where illustrating and explaining some computational steps is at least as important as executing them. Teaching is one obvious use case, but it's also valuable for sharing scientific methods, documenting a library with runnable examples, or presentations at programming conferences.