Code And Cocktails

Notes on Zettelkasten, GTD and the Memex

| Comments

(Recently I started hearing about this note taking system called Zettelkasten. (Seems, as usual, I’m a bit late to the game as the Interwebs was talking about it more in 2020 and prior…) What follows is some notes and thoughts about it and how it connects with other thoughts.)


Zettelkasten (or ‘Slip Box’ / ‘Card Box’ in English) is a note taking system used (and perhaps invented) by Niklas Luhmann. He said it was the way he was able to so prolific in publishing academic papers. It involves writing notes/thoughts on small cards and organizing them in a tree structure, physically in a card box. Each card gets an index number (based upon where it is the tree) and references to other cards if applicable. Index numbers were made of letters and numbers alternating thus allowing for the branching. Cards “1” and “2” are not particularly related, but “1a1” is a note that continues from “1a”, while “1b” is a note continuing from “1” but not necessarily related to “1a”. What Luhmann thought was very important was that there was not any organization a priori. He felt that “accident” was very important and having an organization ahead of time would limit the possible thoughts one could have.

More details on this system and how to use it can be found in Luhmann’s paper and also in Sönke’s How to Take Smart Notes.


How does this connect with GTD? Luhmann’s process of reading and taking notes reminds me a little of the GTD’s separation of capture and processing.

Luhmann’s system had three types of notes: “fleeting”, “literature” and “permanent”. The only differences as far as I can really see between “fleeting” and “literature” notes is where they came from and how much supporting bibliographic material they may have. Both are written down at any time and collected together. Later (but not too much later) they are reviewed and either discarded or turned into permanent notes. Permanent notes are stored in the Zettelkasten. As notes from the inbox are processed one decides what one needs to do with them, where do they fit into the system.


In 1945 Vannevar Bush published an article in The Atlantic Monthly entitled “As We May Think”. In it he describes the “Memex” a sort of work desk / computer which could be used to search for and view documents (on microfilm) stored in the device. More data could be added to it - either personal notes / documents which would be scanned in; annotations on the documents; or even new documents bought or given to the user by others (you could make copies of documents to share with others…)

Basically if you squint at it just right he described a sort of World Wide Web. Documents available at your fingertips and documents were linked together. This availability and linking would allow people to do research on any topic quite easily, synthesizing existing information into new information.

One thing he mentioned that the Memex would have tat we do not have in the WWW is his idea of “trails”. With the Memex you would be able to annotate and save the “trail” of our research, that is you could save the fact that looking at document A led you to document B which lead you to a footnote of document E etc. etc.. These trails were another piece of data one could store and share with the Memex. A trail could be shared with a colleague thus not only sharing the documents you looked at, but how you looked at them.

I think it could be interesting if this idea of trails could be developed.


So how do these thoughts connect? For me the idea of the Zettelkasten and of the Memex are similar, how to store and access information; how to synthesize information into new information. That alone is a good thing - but how to deal with all the information you may be putting into the system? Via GTD’s separation of capture & process, Zettelkasten’s fleeting vs. permanent notes.

A new piece of information or thought can be captured quickly - now your brain can go back to the job at hand. Later (hopefully not too much later) one goes through all that captured information and decides what to do with it. Some of it may, in retrospect be not important - you discard. Some may be important - you store it. Not only that but during storage one synthesizes it with existing knowledge which may immediately create new information.