Monday, July 27, 2020

the zettelkasten / the zeitgeist

Discussed: The idea of a Zettelkasten, note-taking, index cards, wikis, How to Take Smart Notes by Sönke Ahrens.

This post roughly continues a thread that goes something like:


For the unfamiliar: “Zettelkasten” is German for “slip box”. It refers to a note-taking method where ideas and bibliographic references are stored on index cards or slips of paper.

There’s a decent chance my first exposure to the word was on a blog by Manfred Kuehn called Taking note, which started publishing in 2007 with an entry about Niklas Luhmann’s Zettelkasten:

One of the more interesting systems for keeping such index cards was developed by the German sociologist Niklas Luhmann (1927-1998). […] Luhmann claimed that his file was something of a collaborator in his work, a largely independent partner in his research and writing. It might have started out as a mere apprentice when Luhmann was still studying himself (in 1951), but after thirty years of having been fed information by the human collaborator it had acquired the ability of surprising him again an again. Since the ability of genuinely surprising one another is an essential characteristic of genuine communication, he argued that there was actually communication going on between himself and his partner in theory.

By the time I read that, I’d already spent time thinking about index cards as a way to organize knowledge, and experimented with a card box that might have become a full-fledged paper Zettelkasten if I’d kept at it. I think these ideas were on my mind because of C2’s stuff about index cards in software development, the notion of the Hipster PDA, and my friend Brent’s fixation on David Allen’s Getting Things Done.

Hypertext had been a preoccupation of mine for quite a while by the time I heard of Niklas Luhmann: HyperCard in the early 90s, the web, the wiki (with its roots in a HyperCard stack), Ted Nelson’s Computer Lib/Dream Machines. Apart from introducing me to Ward’s Wiki, Extreme Programming, Agile, and GTD, Brent Newhall wrote a simple filesystem-backed wiki in Perl with some unique features. I wound up maintaining that code for years, and used it to keep a personal wiki on this site for at least a decade. (Any readers I retain from back then might remember that it functioned as a comment / “marginal notes” / linkblogging system here for much of that time.)

Luhmann’s Zettelkasten was a kind of paper hypertext. He numbered individual cards/slips in such a way that related things could be found in physical proximity, and made links between cards by referencing those identifiers.

So now it’s 2020 and the Zettelkasten is having a moment. Sort of a nested moment, inside of a larger one about note-taking and personal knowledge systems. I haven’t really traced out the web of influence here, but there’s been an escalating flurry of pieces like these:

There seems to be a thread of interest in the rationalist / LessWrong scene. Apart from that, I’d guess much of this is due to the work of Christian Tietze and Sascha Fast, who maintain a long-running blog and forum at, sell note-taking software for the Mac, and have recently begun promoting an online video course on the method. (I believe there’s also a book in the mix somewhere, albeit one not yet translated to English.)

Unsurprisingly, the community at is the most direct place to watch an entire ideological complex, complete with in-group vocabulary and evangelical fervor, crystallize around the core idea. That said, it feels like it’s spreading and mutating in the wild by now, and would probably continue to do so independent of any particular guru figure or canonical text.

how to take smart notes

If there were a canonical text in English, at the moment it would probably be How to Take Smart Notes, by Sönke Ahrens. That’s the book that gets mentioned over and over again. I bought a copy back in February, after skimming the first chapter and reading a bunch of blog material like the stuff linked above.

I decided to write up my notes here after I recommended reading it to a friend who turned out to thoroughly hate it, and seeing similar reactions elsewhere. Although it fails to make as strong a case for its ideas as it intends, I’ve personally found it helpful for thinking about my habits.

This is a short book - 170 pages with bibliography and a very brief index in this edition. It’s also substantially longer than it needs to be, which isn’t unusual for this sort of self-help nonfiction. To its credit, it’s fairly dense, but it veers into evangelism and salesmanship often enough to be frustrating, and makes claims that some readers will find questionable, if not off-putting. It also comes with a dose of pop-psych material.

Construed strictly, Ahrens' idea that “nothing else counts than writing” is too narrow a conception of work for most people. It’s simply not true for programmers, engineers, designers, customer service reps, or project managers — let alone general contractors, farmers, or electricians. Most people who could benefit from note-taking habits aren’t chiefly concerned with writing documents even when documents are integral to their work. Where the exhortation that writing is the only thing does ring true is when your goal is to produce written artifacts, e.g. to turn your reading into research output.

Smart Notes as a whole tends that way: It’s explicitly aimed at students, professional academics, and nonfiction writers. While I occasionally qualify as that last, none of those roles map to the scope of my note-taking. Accordingly, this is a book I read selectively and with a critical eye, gleaning what I could and generalizing where useful. I’d suggest other readers approach it the same, particularly if, like me:

  • You don’t work in an academic field.
  • You aren’t much concerned with writing papers.
  • You rely on your notes to archive collections of specific facts and remember sequences of events as much as to connect and synthesize ideas.

I do think it’s a useful read if you’re interested in the mechanics of a Zettelkasten and haven’t found what you’re looking for in other writeups, or if you’re just looking to yak-shave a personal knowledge system.

I don’t, strictly speaking, keep a Zettelkasten. I have, however, been borrowing ideas from people who do. After finishing How to Take Smart Notes, here’s some of what I think I’ve taken away from it and related sources:

  • Your notes can be:
    • An extension of your long-term memory.
    • A living system.
    • Capable of surprising you with new connections, forgotten ideas, and emergent patterns.
  • Writing is a means of thinking.
  • Read (or work) with a notebook to hand. Jot stuff down as you go.
    • Using the same notebook for everything will save you thinking about which one to write in.
    • The notebook can function like an inbox. Process things from there into permanent note storage, be that in electronic form or on index cards.
  • Track citations / bookmarks / bibliographical references.
    • Luhmann’s paper Zettelkasten seems to have used a dedicated card file for this. Ahrens recommends tooling like Zotero.
  • Work in small units.
  • Summarize/restate ideas instead of just quoting or excerpting things. Link them to other ideas already in your notes.
    • Just reading a text isn’t the same as understanding it. Restating an author’s ideas and integrating them with your existing knowledge is a kind of self-test, and facilitates learning.
  • Add stuff to your notes if:
    • It connects to something already in the notes.
    • It’s open to future connections.
  • You might understand something if you can effectively teach it.
  • Hierarchy is likely to get in your way. Draw connections within the whole space of ideas, without being limited to the current level/tier/box/rank.
  • “To get a good paper written, you only have to rewrite a good draft;” for a draft, a series of notes, for a series of notes, rearranging what’s already in the slipbox, which you’ve written as you go. “All you really have to do is have a pen in your hand when you read.”

That last one cuts pretty close to the heart of the method the book espouses. It’s focused on writing an academic paper, but if you fuzz it out a little I think it gestures at something more generally useful.

Most of the work of understanding things is incremental and piecemeal: Refining and tending a fragmentary web of memories, perspectives, practices, states, and relationships. Notes are a technology for accumulating that work and extending its durability outside of our skulls. Used well, they’re a foundation for making new things and a solid place to stand when faced with recurring problems.

further research or whatever

Anyhow, while I find the Zettelkasten thing interesting as a cultural happening, I’m not concerned with replicating it.

In the broad outlines, the notes I keep in VimWiki look a lot like an electronic slipbox. There’s a bunch of stuff in the Luhmann / Kuehn / Ahrens / trains of thought that seems useful to borrow, and lines up well with things I’ve already learned working with wikis, version control systems, bookmarks, and a couple decades of paper notebooks. On the other hand, there’s a lot in how I model the world and how I think in writing that doesn’t fit.

I often need to think in terms of when very specific things happened: State changes to complicated systems, what happened when I ran some technical procedure, when I planted a bed of onions. While restating ideas and situations in my own words is a good way to get a handle on various things, I also find it useful to archive verbatim fragments of conversation, specific texts, chunks of code, and long transcripts of program output. Some of my “notes” are really executable scripts, and a lot of my external memory lives in source code repositories, wikis, README files, command-line histories, and issue tracking systems.

All of that’s led me to thinking in terms of logs and journals, and roughing out some tools for a 2-axis time vs. topic approach that I’ll elaborate on one of these days. I’d also like to make more room in my system for integrating drawings, photos, and structured data, though I’m not entirely sure how to go about it.

In the meanwhile, I’ve been thinking a lot about various collections of public notes (some more Zettelkasten-adjacent than others), stuff like:

p1k3 / 2020 / 7 / 27
tags: topics/notebooks, topics/notes, topics/reading, topics/zettelkasten