Zettelkasten Forum


What is going on in a Zettelkasten visually

A Zettelkasten may look something like the image above, where the arrows represent the direction of the links and the boxes a particular kind of Zettel..

Links aren't always bidirectional. E.g.: Let's say I have an overview Zettel on Maria Popova's note-taking method. From this Zettel, I may link to a Zettel explaining how she uses sticky tab notes. But, the latter Zettel might not link to the former.

Another thing to note is that the item in the bottom of the image may link to further details. This might happen when you break down a Zettel that violates the Principle of Atomicity. E.g.:

  • Topic-based overview: "Maria Popova's note-taking system."
  • Idea relevant to the overview: "Idea index definition."
  • A detail of the idea: "Reading is about identifying patterns."
  • A detail of the detail: "Pattern definition."

I based my diagram on what's been discussed in the Zettelkasten blog, so others' Zettelkasten might look different.

Questions for the reader:

  • What would you change of my diagram?
  • How does your Zettelkasten look like?

Comments

  • @Dilan_Zelsky said:

    • What would you change of my diagram?

    Nothing, if this works for you.

    • How does your Zettelkasten look like?

    You're gonna think I'm messing with you ...

    I like to think of my Zettelkasten as "just" containing notes. Each note is atomic, with links to other notes that are also atomic. I've tried the classification thing (per Sönke Ahrens), but it got in the way too much. I was spending more time figuring out how to classify something, than simply writing and linking notes.

    The end result is a messy Zettelkasten, but one that meets my needs (helps me learn new stuff and think). I guess my brain is messy as well :-).

  • edited August 29

    My ZK

    that's all ;)

  • edited August 30

    @r1tger

    I feel you. Maybe I should stop classifying things too. Thank you for sharing!

    @Ubaldo_Passamonti

    Thank you for sharing your Zettelkasten too. You have two Zettelkasten instances? Mind sharing why?

  • I went through the same process as @r1tger. I don't regret what i learned from this but it is ironic how much more beginner friendly my Zettelkasten is now compared to then.

    my first Zettel uid: 202008120915

  • Hi @Dilan_Zelsky
    sorry for the picture being a little confusing.
    The problem is that, given my bad English, my writing could be worse :smile: .
    So please forgive me for this.

    My ZK goes this way:

    1. Software I proudly use: The Archive
    2. All notes go in one folder (ZK in my use case)
    3. All important files relevant to my interests (PDF, JPG,PNG) go in one subfolder (ZK/DOC in my case) and are linked to the rilevant notes -> (![](doc/ID-n.jpg))
    4. All notes are created equal
    5. I put in the main folder (ZK) every kind of note: atomic notes (Zettels) but also every note I think could be useful later on (quotes, encyclopaedia/dictionary entries, etc)
    6. Zettel notes get a #zk tag - links to other ZK notes (sometime to non ZK notes too) - no classification of ZK notes - #tags when I need
    7. Non Zettel notes get a tag depending of the kind of the note (#dictionary, #quote, etc...)
    8. All notes (Zettels and not) get a unique, time-based ID + "-" + a title (for example: 202108301443 - gli Italiani sono molto simpatici :smile: - Title and ID are repeated in the body of the note
    9. All linked files get a ID (the same as the linked note + -number) - -> es: Note ID 202108301443 linked to 202108301443-1.jpg, 202108301443-2.jpg, 202108301443-3.pdf and so on

    It seems to me a little messy ("un gran casino" as we say in Italy) but i like it ;)

  • @zk_1000

    I'll ditch out categories soon, then. I'd really like to make my ZK more friendly.

    @Ubaldo_Passamonti

    Thanks for elaborating! Our Zettelkästen share several similarities. E.g.: I also dump .png files and the like in my ZK folder.

    It seems to me a little messy ("un gran casino" as we say in Italy) but i like it ;)

    If it works, I'd say that's good enough!

  • Went ahead and mapped out my workflow

  • @Nick

    Interesting; thanks for posting.

  • @Nick

    Thank you for sharing your workflow. It looks quite interesting. Also, I think that you made a really clear and detailed diagram. Nice job!

    I'm particularly interested in that idea index. Could you say more about it?

  • edited September 2

    @Dilan_Zelsky it is nothing fancy. It is just the practice of creating a personalized index (like at the back of the book) of the ideas you find most salient/interesting while you read.

    So I have a piece of paper with Title of the book at top and bullet list underneath

    • Idea 1 - p. 3, 10, 35, 98, 120
    • Idea 2 - p. 6, 16, 76
    • Idea 3 - etc
    • etc

    Comes from Maria Popova and her book synthesis website Brain Pickings

  • @Nick

    I see.

    Do you find the practice helpful? It sounds to me like you could fall for the Collector's Fallacy there: You make that index but forget about doing the actual reading.

    To me, it feels like it could help in the processing stage, when you form clusters from reading notes. Then, the ideas from the index would be what those clusters would be about. Have you tried doing that instead of while reading?

    Would love to hear your answers.

  • Very cool that y'all share these diagrams. It's super interesting to see how each of you think of the Zettelkasten or your workflows in different terms and include different things in the pictures.


    I used idea indexes to add something like an overview of keywords, esp. for fiction books, as a preparation step to help with clustering ideas: https://forum.zettelkasten.de/discussion/1930/use-idea-indexes-for-journals-and-fiction

    If you process a book from start to finish, this jumping back and forth can be skipped a lot of times, though: say you have processed the first 200 pages of a book already, and find an idea that relates to something from page 50, you can look up a Zettel about this. Of course I page back through the book to see if I missed a notion originally, but ideally, my Zettelkasten got me covered and I can expand the existing note.

    So a linear process in the end covers similar ground to clustering, provided you do process the whole book, of course.

    An idea index still adds a personal, bibliophile touch to a book, but I personally don't find much use in them. They could add flexibility if you just want to extract e.g. 1 of 20 concepts that's scattered across multiple pages, though.

    Would love to hear more about what @Nick thinks about this, but this thread is not the place for this :)

    Author at Zettelkasten.de • https://christiantietze.de/

  • @ctietze Tbh, I don't have anything else to add. I just wanted to thank you for contributing to the thread.

  • Do you find the practice helpful? It sounds to me like you could fall for the Collector's Fallacy there: You make that index but forget about doing the actual reading.

    @Dilan_Zelsky well you have to do the reading to create the index in the first place. I don't use existing indexes, I'm creating one as I read. I'm also not creating an index of the whole book, instead I'm creating an index of the ideas that I particularly find interesting or useful. The key to avoiding the collectors fallacy is to always make sure you are adding to the information and connecting it to what you already know when creating a new note.

  • @Nick

    I see. Thank you for sharing your workflow. You even found a way to avoid the fallacy.

    I may give the idea index a try. Sounds interesting,

  • Here is a 2014 (sick!) graphic by @ctietze, from https://zettelkasten.de/posts/zettelkasten-building-blocks/, with several contributions to the comments section.

  • @thomasteepe This acurate to this day. The issue with trying fix too much into workflows is that you treat yourself like an beaurocrat treats people: With great inflexibility and an attitude you have towards machines (compared to people).

    I am a Zettler

  • @sfast
    I'm skeptical whether the picture is accurate, either as a description of what people actually do or as a prescription what they should do.
    In my view, in 2014 the sketch was meant as a well-stated starting point for a debate - in the passage directly above it @ctietze wrote:
    "I’m iteratively getting a better picture of the Zettelkasten using the systems metaphor. The picture I show you today is still work in progress, so of course I’m interested in your opinion and I’d love to debate this model."
    The tiny fraction of forum discussions I was involved in since 2014 produced the following aspects, that could immediately be represented as changes to the sketch:

    • the Archive can trigger ideas out of itself without external input,
    • the Archive can influence what new external information is consulted,
    • ideas can certainly emerge inside the system's boundaries,
    • in the framework of the zettelkasten method, a bicameral system of a heuristic and an epistemic structure can be used, with tool notes as a practical implementation,
    • an ecosystem of several cognitive support tools can be used, in particular a "feeder system" to the Archive can be used, where ideas can develop some maturity before they enter the Archive.

    And I'm sure there are countless other and superior ideas and new options around the zettelkasten framework I'm not even aware of.

    I understand that you are involved in teaching zettelkasten practices to clients, and I imagine that it is very helpful to start with a basic model that is flexible but does not include every single niche option. But I would still question the value of calling such a simplified model "accurate". I see here the danger of curtailing the future development of a method that is currently in a very interesting and dynamic stage.
    I'm confident that in this forum, together we can create ideas for the next generation of zettelkasten systems.

    Remark:
    I am a fan of the picture - it has triggered my first comment on the zettelkasten forum, if I am not mistaken. The "(sick!)" part was meant as an approving stereophonic joke on the slightly academic comment (sic!) and, as a nod to the 18 year old original poster, the youth slang usage of sick.

  • @thomasteepe said:
    @sfast
    I'm skeptical whether the picture is accurate, either as a description of what people actually do or as a prescription what they should do.
    In my view, in 2014 the sketch was meant as a well-stated starting point for a debate - in the passage directly above it @ctietze wrote:
    "I’m iteratively getting a better picture of the Zettelkasten using the systems metaphor. The picture I show you today is still work in progress, so of course I’m interested in your opinion and I’d love to debate this model."
    The tiny fraction of forum discussions I was involved in since 2014 produced the following aspects, that could immediately be represented as changes to the sketch:

    Christian might doubted the accuracy of this picture. I don't. I don't agree with everything with this picture. But it is accurate in its spirit. I'll show you what I mean by commenting the points.

    • the Archive can trigger ideas out of itself without external input,

    You could make another arrow to demonstrate this. However, then you should also make another self-referencial arrow from the text to itself. If you edit a text you basically let the conversation you have with your draft generate the changes itself.

    • the Archive can influence what new external information is consulted,

    Yes. However, this would mean that you no longer could label the arrows "process". You'd need two arrow-types to distinguish actual flow of action and various types of influences. To incorporate this aspect would mean to change the part of reality that the model maps.

    • ideas can certainly emerge inside the system's boundaries,

    This would be an ontological error since ideas are not something you directly process. They are always outside of the system's boundaries: In your head which is not part of the system. They are mentioned outside of the system's boundaries in an ontological correct manner since they are part of a provisional group of family resemblance (Familienähnlichkeit). This blur is needed to be receptive to change (similar to the function of the right hemisphere btw.)

    • in the framework of the zettelkasten method, a bicameral system of a heuristic and an epistemic structure can be used, with tool notes as a practical implementation,

    I don't understand. :)

    • an ecosystem of several cognitive support tools can be used, in particular a "feeder system" to the Archive can be used, where ideas can develop some maturity before they enter the Archive.

    Yes, but still outside of the system. I do have quite some tools inside the Archive and outside. It is a technical question: Since we, humans, are donuts what you eats never gets inside until it is actually absorbed by the intestines. (This includes the grace that our feces are never part of ourselves even before we do our bathroom deeds)

    This image is not valuable in a way that it shows you the complete picture of everything relating the ZKM. It's value lies in showing us the simplicity and the principal architecture of the overall set up.

    And I'm sure there are countless other and superior ideas and new options around the zettelkasten framework I'm not even aware of.

    I am sure that some ideas will challenge this picture. :)

    I understand that you are involved in teaching zettelkasten practices to clients, and I imagine that it is very helpful to start with a basic model that is flexible but does not include every single niche option. But I would still question the value of calling such a simplified model "accurate". I see here the danger of curtailing the future development of a method that is currently in a very interesting and dynamic stage.

    Perhaps, we have a different understanding of what we mean by "accurate". In the book, I wrote a short section on models: Models map reality by deviating from it as much as possible for a given purpose. If a model deviates too little from reality it blurs vision and decreases the utility of itself.

    I'm confident that in this forum, together we can create ideas for the next generation of zettelkasten systems.

    I am, too. I'd rather like to develop modular components, rather quite some atoms for the molecules that are the personal overall systems.

    Or: An iterative model that is more like a model family with simple models at the bottom and more and more complex models at the top.

    Remark:
    I am a fan of the picture - it has triggered my first comment on the zettelkasten forum, if I am not mistaken. The "(sick!)" part was meant as an approving stereophonic joke on the slightly academic comment (sic!) and, as a nod to the 18 year old original poster, the youth slang usage of sick.

    I am a Zettler

  • I use Notion and my basic breakdown goes like this:

    1. Jot down notes
    2. Finalize notes
    3. Add links and comments
    4. File in ZK
    5. Whatever "cluster" is increasing, I usually turn that into some writing

    Pretty simple really. Honestly, I've been writing since I was a kid, and really don't have trouble with that. I've written plenty (years and years of blogs and articles and a couple books) before I had the ZK. So, for me the ZK is just a level up on what I've already been doing. (And I love it!)

    Here's the visual of how I collect notes, etc...!

    First, I write a Quick (fleeting) Note in the top box pictured. When I hit send, it flies away into the "Needs Filing" section, which is a subsection of my Zettelkasten page.

    Below shows the ZK page itself. The top screenshot is the ZK page with boxes closed (including the "Needs Filing" box). Below that are screen shots of the Permanent Notes box opened. Notes are sorted by title code, coded by concept relationship. The "Topics" are there just in case I need to find a cluster quickly (which I often do). Notes are internally linked and backlinked.

    The bottom two screen shots (above) show what happens when you click on the note or the quote. A box pops up showing the recorded text. This mimics the effect of pulling up slips from the box to see notes.

    Tags are mostly there for reference, though I use them less and less once I realized you could just filter the notes by searching keywords in the note text.

    The "Making Connections" box (in first screenshot above) is for filtering and searching cards, laying them out in "Gallery" which is essentially a "card view." Here I can move cards around, connect ideas visually, without disturbing the organization above.

Sign In or Register to comment.