Comment by rgoulter
1 day ago
> And I don't think org-mode's babel features really exist in Markdown
I think the neatest part of org-babel is the source code block execution, & the various ways it supports for configuring output. This allows for org files to be "plaintext notebooks" (like jupyter in plaintext"). -- It's really surprising that this part is not more common.
More niche is the "babel" part of that: because the code blocks can take variables as inputs, and output values, this allows a polyglot notebook where values from e.g. Python get passed to R and plotted or so. -- Cute idea, although I've never found it too useful. The supported types are (unsurprisingly) limited, and the language support for code blocks is held together by duct tape.
(Even more niche is the noweb syntax for proper "literate programming". Which is mostly discussed about how awful it is to use in practice?)
Of course, org also has a long tail of neat features (like how each heading can have properties attached to it, as well as tags, and the task management that relates to this).
> It's really surprising that this part is not more common.
I think it is because of low Emacs adoption and other editors not having enough support.
The problem with polyglot notebook workflow is probably, that you can only use it well for small data, or at least not big data, because who wants to have a million lines of output suddenly appear in the buffer, only to then read them as input for the next language ... That would be a tremendous amount of computational overhead. And if we didn't have that, we would need a way to pass a proper value from one language to the other.
What I also like is, that you can define code blocks that are used as formulas for spreadsheets (tables) inside the document. That's quite powerful too.
> (Even more niche is the noweb syntax for proper "literate programming". Which is mostly discussed about how awful it is to use in practice?)
I don't find it very awful to use. I have used that for working through computer programming books and it was fabulous.
I agree about the duct tape, which I also use often around the house, so maybe that's why I like org-babel :)
Just wanted to say that I share data between different blocks in different languages through files and env variables (I add :session shared to the src blocks that need to access this). That is useful also to have src blocks you can execute repeatedly and that depend on something like an aws identity being assumed (you just assume it in the first block that shares the session).
I agree it's messy, it's just a mess that works for me.
I think this is the crux of the issue, use like this is like a real program, just built up incrementally in a notebook rather than a repl or shell-with-pipes, and with manual error handling. The STEPS project was all about this- a way of incrementally building blocks that can be composed.
With org mode in mind, ideally you would have language support for this ie. Comments are scoped metadata that can be formatted, tested, linked etc.
You need a well defined spec like djot as a DSL for this to work, so that parsers can be easily written for it. This level of language support allows many different views onto the source code. We’re not there yet.