TheKnarf

Literate Markdown and Datalog-based Dataview

The Obsidian Dataview plugin seems really useful. It let's you write queries over you Obsidian vault, treating markdown documents as a database, and embedding those queries in markdown files themself. Making it a bit closer to a "Notebook" without tying it to a programming language.

I would like something like that, but:

  • I wish it was not tied to Obsidian

    • I wish it could work standalone as a CLI tool

    • I wish it was a re-usable library that one could build on top off

    • I wish it could work both in Obsidian and Vim

  • I wish it could query more than just yaml frontmatter

    • Ex. query arbitary TODO's, or table data in a markdown file

  • I wish it could be used to build UI's like Kanban board's, and for changes to the query result to propegate back to the source.

    • Queries may need to function as lenses into the Markdown AST

  • Maybe it could work with Obsidian Freeform?

  • Based on Datalog instead of SQL-like queries

Use-cases

  • Collect TODO's across projects into an overview board

  • Mallable software, build simple tooling with Markdown and queries

  • An alternative to Notion databases

  • Sync Github issues to Markdown files and build your own Kanban-board materialized views into the data

  • Sync calender events to Markdown files, build Calender-blocking TODO lists (as materialized views)

Interesting tooling

  • nb - nb is a CLI tool for note-taking, bookmarking, archiving and knowledge base. It's doing a lot of what I'd like to build myself on top of my tooling.

Literate programming

Literate programming is the idea of mixing long form writing with code. This can be to produce books, articles, documentation that doubles as runnable programs. That way you ensure that the code examples in your blog post or book work, or you have a natural way of writing more long form documentation.

Interesting tooling:

Memex & Archiving