A Zettelkasten (German for “slip box”) is a method of note-taking and knowledge management which encourages linking between notes to allow emergent structure to develop. By slowly building a collection of densely linked notes over a long period of time, it becomes possible to see interesting links between things that were not initially apparent.
The idea, which has existed for centuries, was pioneered in modern times by the German social scientist Niklas Luhmann, who credited his astoundingly prolific writing output to his physical Zettelkasten, a labyrinthine system of index cards linked with an intricate ID system (predating hyperlinks by twenty years). There has a been a resurgence in interest in the idea in the late 2010s, related to the GTD movement and the idea of personal knowledge management or building a second brain. Roam Research popularised it even further with their tool and its deep embrace of the idea of backlinks.
This site is built from a sample of my own Zettelkasten, which contains all the notes available on this site together with a much larger collection of private notes, journal entries and memories. I edit all these files in emacs org-mode and use org-roam, an open-source Roam clone implemented in org mode, to generate backlinks.
Was reading Alex Kehayias' reflection on writing five hundred notes in his excellent Zettelkasten. Got me thinking about mine and how a lot of my notes here are “stubs” and whether that’s OK. But what should be the “generic format” of a note. I like that he has simple descriptions at the start of every note that give a nice context. Will start to adopt this.
Today I wrote a couple of tweaks to my scripts to also generate a local version of the full Zettelkasten (i.e. not just those files that are labelled public). For some reason it runs really slowly; need to understand why. Another thing I definitely notice is that when the backlinks are to date entries, the lack of context around the backlinks makes them completely useless.
Adding new atomic thoughts under a note: it’s good to use dates. A few format / guideline changes I should make throughout before it gets too big to handle.
Good public Zettelkastens #
- Karli Coss
- Tom Critchlow
- Phil Jones
- Alex Kehayias
- Jethro Kuan
- Andy Matuschak
- Manuel Odendahl
- Bill Seitz
- Mike Tannenbaum
- Harry Taussig
- Cosma Shalizi
- Nikita Voloboev
Links and resources #
- A repository of 10,000 Markdown files with links for testing note-taking / Zettelkasten tools.
- An interesting knowledge base for maths and physics, which is clearly influenced by Zettelkasten-style tools. Not a lot of content here but definitely unique.
- The man himself, Niklas Luhmann, describing his physical index card-based method in full. It includes the wonderful quote:
As a result of extensive work with this technique a kind of secondary memory will arise, an alter ego with who we can constantly communicate. It proves to be similar to our own memory in that it does not have a thoroughly constructed order of its entirety, not hierarchy, and most certainly no linear structure like a book. Just because of this, it gets its own life, independent of its author.
- A nice list of public Zettelkasten (which includes Mt. Solitary).
- The Ross Ashby digital archive, digitised versions of Ashby’s notes made available by his estate, organised as a Zettelkasten.
- A great Atlantic article on the history of index cards, which have always had some weird appeal to me.
- A nice summary of Mingyang Li’s Zettelkasten focussing on clusters in the generated knowledge graph (implemented in Obsidian).
- Zetteldesk, an interesting package for arranging org roam nodes on a “desktop” scratch buffer to help connections emerge.