Shallow Zettelkasten, potentially missing interesting loose relationships
Hi. I just joined here, I think because of some insecurities with my Zettelkasten. I thought to dump it all here and hope that anyone is kind or interested enough to comment. I have put a few rabbit hole-hours in zettelkasten youtube, even bought and (sloppily) read the Antinet Zettelkasten book. It seems though, im still confused.
What I like about Zettelkasten so far:
- Even if my whole zettelkasten turns out to be a huge unsuable failure, I enjoy having a format that allows me to sit down and interact with my material after I have read it.
Contrary to the impression the rest of this post will give you, I do have a few (maybe one) thought that my zettelkasten allowed me to produce, that I likely would not have thought otherwise (for the interested: that Christians throughout history seems to have used Aristotle to justify a lot of shit).
My setup
Analog Zettelkasten - With inspiration from Scott P. Scheper, Kathleen Spracklen and Forrest Perry.
A few base categories so far (10-15)
- 10: History; 11: History of ideas; 12: history of religion; 13: My owh thoughts (this will hopefully not survive as a category); 14: ontemporary polotics; 15: society. 14 & 15 have btw become almost indistinguishable.
Alter between numbers and letters when going "deeper". E.g 15.1a1b - This, with inspiration from Kathleen Spracklen.
My fears
I'm putting a lot of work into this now, and 3 books in, i'm starting to get a sense that im doing something wrong, not sure what.
The potential problem
I have read 3 Books since I started with Zettelkasten (Dominion - Tom Holland; A brief hostory of equality - Thomas Piketty; The Bubble - Andreas Cervenka), these have produced around 180 cards. Although these are very different books, which might explain the scattered feeling I have. Aither way, I have noticed that my cards rarely go very deep. With every card, I increase the index with 1 – and thats it. I have a few branches with a depth of 2-3 (e.g 10.1a1), but in most cases the depth does not stem from interesting connections made throughout multiple books, or even a singe book. They are the result a single argument I put down from a single page in a single book.
From the reading/listening I have done on Zettelkasten, I get a sense that its often recommended not to create further category/folder cards, but instead just put down the ideas and let the stack of cards grow dynamically, connecting ideas as they are created rather than trying to set precise folders and categories.
This is intreguing, but I feel now that most of the things I put down is quite specific (because I dont ever try to create higher level category cards), which makes them hard to build upon. I do not doubt that I might be able to create nice links between cards in the future (linking them by writing another cards number on the backside), but I was hoping to see more depth of layers, or local clusters of single topics.
To give an example, here is one card I created (15.11):
Share of securities portfolio held in tax havens:
Around 10–20% of the securities portfolios of Europe and Latin America are held in tax havens. The corresponding > figure for Russia and the Gulf states is between 30% and 50%.
Most of my Zettelkasten is built on these kind of specific facts/ideas, which seems hard to build upon. If I ever read an interesting passage in the future about tax havens, the card above is likely too specific to have this new tax haven idea under it.
Let's say I produce a new card about tax havens which is not directly related to the above fact. In my mind it would not make sense to create a new 15.11a card. Its too loosly realted. To have it as 15.10 or 15.12 would make sense, but in my Zettelkasten, those numbers would likely already be taken by completely unrelated cards. I have no real way of distinguishing "loosly related" and "direcly related" by pure card numbering.
Might be more of a rant than a specific question. I'm happy though to get some loosly related answers if any related thought strucks you while reading this. Hope to get some good inspiration.
Thanks!
Howdy, Stranger!
Categories
- 3K All Categories
- 153 Research & Reading
- 696 The Zettelkasten Method
- 10 Knowledge Work
- 100 Writing
- 467 Software & Gadgets
- 154 Workflows
- 731 The Archive
- 15 Plug-In Showcase
- 88 Resolved Issues
- 225 Projects Logs and Journals
- 83 Project: Zettelkasten.de
- 53 Critique my Zettel
- 173 Random
- 374 Introduce Yourselves!
Comments
Youtube is full of Zettelkasten videos that contradict each other. My recommendation is to pick one teacher and learn their system well, before you include other perspectives.
Scheper's Antinet is quite faithful to Niklas Luhmann's style of working with paper index cards. You can learn a lot from him about Luhmann's system.
The question is, if Luhmann's system makes sense for you.
Your expectations
You don't need a Luhmannian Zettelkasten to do this. Others use other tools.
You don't need a Luhmannian Zettelkasten to do this either. Any kind of note-taking system will have this effect, if you use it with that intention. Some teachers tend to hype "Zettelkasten" as some magic creative machine. I wouldn't take them too seriously. Many of them just want to sell their books and classes.
The question is, what you want to get out of your zettelkasten.
Paper vs digital
Some Zettelkasten teachers tend to gloss over the differences between paper and digital. I think they are two very different beasts.
As a rule of thumb I'd say that paper works best, if you like to write few and concise notes. If you have a more verbose style or want to capture lots of information, digital might be easier for you. I'm curious why you chose paper!
Folgezettel
This numbering system is unique to Luhmann. It's a recurring topic in this forum. Search for "Folgezettel" or "Luhmann ID", if you're curious. :-)
As far as I know only few people stick to this system after trying it out:
Folgezettel don't have "depth". That's one of the many differences to other systems.
Simply counting up the index was Luhmann's default.
Yeah, that's a basic principle of the Folgezettel technique.
I find that much easier with digital notes. For example you can query your notes by keywords.
Another reason, why a digital approach might work better for you.
I use a digital tool for such notes. I'd simply add such factoids to a note called "tax havens", without any special IDs.
I think of paper index cards with Luhmann IDs as a fringe technique. It has its fans. The rest of us can easily do without. :-)
I suggest that you get Bob Doto's book "A system for reading". In chapter 5 he gets into using that kind of numbering system and has some good examples. And visit his web site, https://writing.bobdoto.computer/zettelkasten/. Some pages that speak to your concerns are
https://writing.bobdoto.computer/how-to-use-folgezettel-in-your-zettelkasten-everything-you-need-to-know-to-get-started/
https://writing.bobdoto.computer/linking-new-notes-and-fearing-lost-ones/
https://writing.bobdoto.computer/question-i-just-finished-reading-a-book-and-took-lots-of-notes-now-what/
https://writing.bobdoto.computer/how-to-handle-facts-in-your-zettelkasten/
< ...
The thing here is that the most value of having notes like this is in the work you put in to digest it. Why did you bother to save this one factoid? How does it fit into the body of your notes? That's what a note like this should be about. It will work best when it is written in you own words.
Yes, there is a place for keeping note of bits and pieces for use when you want to develop them into something bigger. And you could put them all into a holding section for this purpose. But chances are there is some reason why this factoid caught your interest enough for you to want to keep it. There is some context here and capturing that context would be valuable. If the title can hint at it, that will help. Otherwise when you revisit this note much later, you may wonder "What was I thinking? Why did I capture it?"
Luhmann had a lot of notes he didn't revisit, so don't feel bad.
In this case, you might have thought -
or
or
or
or
Make the card and its lead sentence or title be about that.
It's not too important to distinguish "loosely" vs "directly" related. In fact, being too precise can work against discovering new ideas. For times when you need to be specific about a link, put the link inside the card along with the reason for it. If you follow a Luhmann system closely - as illustrated by Doto - you might have a series of cards like this (I'm going to indent instead of making up numbers but just replace the indentation by your numbering system's numbers):
. facial tissues are for blowing the nose and wiping the eyes
.. UK usage for "facial tissue": "tissue" (or "paper tissue")
.. Kleenex is a famous brand of facial tissue
.... Kleenex has become the generic word for facial tissues, especially in the US
...... There is some dispute as to whether using a facial tissue is more or less sanitary than using a handkerchief.
........ Public sanitation improvements have made the largest contribution to human health of any health measures.
.......... John Snow stopped the 1854 Broad Street cholera outbreak by removing a pump handle.
............ There is some dispute as to whether this action was really the cause of stopping the outbreak.
.... "Hoover" is another brand name that has become generic.
As you riffle through the cards you will notice these wandering trails and you won't need any more specific link indication than that. If you have a more general number for something like "Public Sanitation" you could link the card "Public sanitation improvements" to it using a link inside the card.
It’s true that other systems of note taking (maybe all systems) will have this effect of forcing me to sit down and interact with the material. I have used other systems of note taking before which already fulfilled this purpose.
What primarily lead me to Zettelkasten was when I realized that I never revisited my previous notes (mainly roam research/obsidian/Logseq). Backlinks (and other link types) was a great promise but I realized, to my disappointment, I seldom used them.
Zettelkasten seemed to be a system that not just required me to continue to add notes to the system, but to continuously interact with the system as a whole. With that said, aim very open to Zettelkasten being the wrong system for me, not sure yet what might replace it.
This is an obvious issue of mine, maybe the core issue. Im not sure. Im not even entirely sure why I read. If nothing greater, I’m happy if it somewhat aids my current frustration. That no matter how deeply I feel that I have understood a book, or how much I have learned from it. In a few months, or at least a year or two, I can be quite confident that I will remember almost nothing from it. For this I have also started to use flashcards in Logseq.
Different reasons. Some of it has to do with the feeling of writing with a good pen, and the idea of having a folder of cards to physically interact with is just engaging. It feels good. Not the most flattering reason, but if I embrace it, I hope it increases my chances that I actually sit down and take the time.
Another reason is that I often get the time to write just before I go to sleep, and I imagine it’s good for my sleep to be somewhat analog. It’s nice to get a pause from my computer, where I will spend my whole day for the purpose of work. (Speaking of work, I have had the intention of creating a separate digital Zettelkasten for work. But certain practicalities have so far kept me from it).
A third reason is that I have found the physical aspect of figuring out where to insert a card, to have some good potential in aiding my described frustration.
I have hopes that an imperfect system, that requires me to take some time and actually read physical indexes, and pick up and search among physical cards, will mean that I am required to interact more with my previous material. This, along with the simple fact that I can see the start and end of my Zettelkasten, and putting some faith in the research of some of the benefits of writing by hand.
This said, I’m only trying it out so far, and I already with 180 cards sometimes wish that I could search among content when I have failed to index them on the right words. I will keep being open to other ways.
That is a good suggestion. I will get my hands on the book and the pages you pointed to were new to me, and useful.
This might be a key idea that I should reflect on more that I have done. I have played with different length of cards (e.g putting more context into some of them), but always felt that while doing so, I broke the rule of cards being atomic. As soon as I add context, the card is not atomic anonyme (by my understanding of atomic). But I will experiment with this more and find a style that is useful.
Yes, maybe I should relax this distinction and see my Zettelkasten just grow. Regarding relation between cards though, I find it hard to extrapolate both yours, Dotos, and other examples I’ve seen into my situation. I understand the examples you put forward is less real world examples as they are designed to educate (must say though, that you examples are more real world than Dotos apples). Nonetheless, especially with my upper most layer (if I may call them that), the cards beside each other are much less related than in the examples.
Maybe my expectation was wrong, but I feel that if you or Doto would have an analogue Zettelkasten, you could likely pick up a bunch of cards from a random place, and while you skim through them, you would get some kind of overview of some topic (tissues or apples). My upper most layer would be a broad topic, like for example history. Now, history if ofc a topic too, but a lot broader than apples or tissues.
I do continuously read (as I do in the links you provided) that a Zettelkasten is made to helt produce some writing. And as one reads, one already pictures where in the Zettelkasten some information might end up. Likely because one created the Zettelkasten with a specific output (or multiple outputs) in mind, which sounds like it would concentrate its content and make it less chaotic. I suspect this is part of my issue, I’m not aiming for something specific enough.
Here I differ. I think of a ZK as a cognitive instrument, but not necessarily one whose main purpose is to help the practitioner produce writings. It may be so that one needs to write to fully develop thoughts, but writing a series of z-cards may fulfill that purpose. Or, I should say, the work of distilling the thoughts or the source material to the point where it can be captured in a card or several cards brings much of the value.
Let's face it, no matter how "atomic" or well-focused a z-card may be, it can only be a pale, flattened representation of the idea it tries to express. So a card that can help re-awaken more of that depth will be a more effective card. Links to other cards can also help fulfill that purpose. In a physical system, the other cards close to a given card will help, because they will tend to bring back the trains of thought that led to the card's production in the first place.
Recently I read parts of Kant's "Critique Of Pure Reason", mostly part IV of the introduction. I had something specific in mind, but Kant is so notoriously difficult that to grasp even a paragraph is a challenge. There would have been no point in noting quotations except to help me to return to them as part of the study. Because I had a definite purpose I could try to untangle the writing as to how it fit into the ideas I was looking for. As I got clear on the subject, I created notes. Each one is tiny and almost banal but when I look at the group I see the subject I was interested in and how I got there. Here's my little series of card titles -
Kant, Critique of Pure Reason Kant's Analytic Vs Synthetic Judgements In Modern Language: Identity Vs Attributes Kant's "By Identity" Is More Than An ID Example: By Identity Example: By Attributes "By Attributes" == RDF Blank Node Kant Maps To "By Identity"/"By Attributes"Here is the content of one of those "almost banal" cards, one that reminds me of my thought progression:
In Modern Language: Identity Vs Attributes ------------------------------------------- :id: tom.20260721000950.1 :created: 2026-07-21 00:11 In modern terms, an object **can be known** 1. by its **identity** (which may include a schema); [intension] or 2. by its **attributes** [extension]. Both can be expressed by modeling languages, e.g. RDF and Topic Maps.For the time being I don't need any links in these cards. Their juxtaposition provides the links that matter for now.
[Please let's not get into any discussion of Kant or metaphysics or whether I may or may not be understanding Kant correctly! I got what I wanted from this study, which took parts of several days]
Why does it matter to you how often you revisit notes? What makes this metric useful to you? I'm asking because I want to help you clarify your personal priorities.
Personally I prioritize other things:
Others have other priorities. :-)
Zettelkasten is not just one system. Different people use the label "Zettelkasten" for very different systems. For example, some consider Luhmann's IDs essential for "Zettelkasten", others don't.
Why do you want to "continuously interact with the system as a whole"? What does this phrase mean to you? How is this different from interacting only with some parts? Again, I'm trying to help you clarify your expectations.
If your frustration is less about note-making and more about reading, then you could approach it from that perspective. Why do you read? What do you want to get out of reading?
One thing that helped me was to differentiate different kinds of reading. I read different texts for different purposes and with different expectations. I can spend hours with a short poem, savoring every single word, enjoying the images and feelings the poem evokes. But I also sometimes quickly scan a 1000 page book in few minutes, because I care only for a keyword or a reference. It depends. I might spend less than a few seconds on a blog post in order to determine if it's relevant. It can take months to work through a single textbook, because I want to remember many facts and learn a skill.
I think that sensual joy and intentional friction are very good reasons to write on paper. And I think they are rather common.
Another excellent reason!
However, it doesn't necessarily follow that you need a zettelkasten for your notes. You could write your reading notes on paper and simply file them away alphabetically by author and title.
This one doesn't convince me. :-) The inserting doesn't seem to work, if I understand your initial question correctly.
Hmm. Adding intentional friction works only to a point. If you turn the whole system into a tedious chore, you'll give it up sooner than later.
How closely have you read this research? The research I'm aware of isn't that clear in its assumptions and conclusions.
As with reading I recommend to differentiate different kinds of writing. For some purposes copy-and-paste is smarter than writing by hand (for example when you want to include a literal quote in your notes). For other purposes, like reading at night, pen and paper are ideal.
You could also try to combine multiple systems. For example you could use digital notes for a searchable collection of factoids and paper index cards for a more writing-centered work.
If you like more inspiration for paper zettelkastens, you could check out Victoria Chowder's fascinating youtube channel: https://www.youtube.com/@AntiNovelist
The term "atomic" confuses many people. Atomicity is a recurring topic in this forum. Beware the rabbit hole. :-)
Confusing terminology is one of the reasons why I recommend to start with one teacher only. If your starting point is Scheper, then stick with his system for a while.
When your first teacher's system doesn't work for you, pick another teacher. And then—that's the important part—compare their teachings. Note what they have in common and where they contradict each other.
When they contradict each other, you have to decide for yourself what you're going to do. The more teachers you get to know, the more contradictions you will discover, and the more you have to make up your own mind.
That doesn't make much sense statistically. If you "pick up a bunch of cards from a random place" in a ZK with 10.000 cards, you're most likely to pick up cards that have zero relevance for your topic.
The difficult part is finding the right cards, the cards that actually give you an overview.
The keyword is "where". In order to have a "where" in your ZK, you need a structure that provides a location.
Luhmann's IDs define such a "where", but not in the sense of "depth" or categorization. They define strings of notes, aka "note sequences". That's a very unique and fringe way of organizing notes. It works only for very few people.
Another kind of "where" are more common systems like:
If you go digital hypertext, "where" can also be defined in terms of nodes in a network. If you work with paper index cards, "where" has also the physical aspect of "where" to put a card in the box.
Yeah, that is part of it. On the other hand it helps to know the options. Sometimes you only know what you need, when you see it.
Alright, good to hear another perspective on that, and the thoughts on atomicity might also clear up a little bit what I might want to strive for in a card.
Looking at your example from Kant's "Critique Of Pure Reason", do you often use a card for the title of the book? If yes, if you would come across an idea in another book that expands your knowledge on this specific topic, would you add it among these cards, or typically link the idea using IDs?
I want to feel that I learn and remember things I read. If you ask me on any given day about my thoughts on some material I've consumed, I'd like to be able to recall some of my take-aways. I'd also like to have a process that encourages me to build on previous material that I read.
Revisiting my previous notes in some way might not be necessary to achieve this. However, using note taking as a method to interact with and understanding material only takes me so far. If I do not remember much of it in the future, I'm not sure what im doing. If I can achieve a process where I continously work with (or recall) my previous efforts, that seems to be an indication of the process being worth it.
Good question, and im not sure. With some inspiration from @tomp , maybe what I mean is that I want my ZK to be a cognitive instrument to help me continously interact with my own knowledge.
When I go to my notes, I think I would rather ask questions such as "Have I thought about something similar before?". Or "This seems to relate to my thoughts on X, can I consolidate these ideas?". Or even have an experience such as "Oh, is this related to my thoughts on Y?" -> "Seems it was not really, but it was a fun detour recalling Y and figuring out it was not related".
The author and book title in itself is nice to have at hand, but the priority is the ideas themselves. Where they came from is second priority. I dont think I will (as often) ask "What now did I think about what this author said in this book?".
Fair enough. I initially felt some potential in the analog inserting. But it's true that I started to suspect i'm doing something wrong in the inserting process. So far, I have not really thought of the "analog" part as the problem. I did enjoy going through my indexes, seeing titles of irrelevant cards just to be riminded also of what I filter out. Hoping this could aid in my "recalling" of knowledge in general problem. Its good that it does not convince you, and I'm very thankful for you taking the time to push me on these things
I have not even read the research. I have just heard it be repeated so many times. I have no problem backing off here.
Thank you for the youtube link, I will definitely have a look. And I'll try to reflect more on different types of reading and writing.
I see that I did not formulate myself correctly. What I meant was that if I pick up a bunch of cards randomly from one place in my ZK (not searching for anything specific at all), I would not even expect there to be much similarity of any topic among them (except for e.g having something to do with history).
Not that this is a practice im looking for, but this is a pattern I would have hoped to see in my ZK. That cards, by their positioning, related more to each other.
For fun, let me pick up 5 cards from a random place and write out their titles:
11.23 - Luther makes faith personal through Paulus
11.24 - Luther makes the bible accessable to the people
11.25 - The freedom to interpret the Bible on one's own after the Reformation creates chaos
11.26 - Yin & Yang
11.26a - The emperor needed to keep a close eye on the calendar
I must (happily) admit, that I expected less correlation between cards with this experiment. There is a reformatino theme, and a tin & yang theme here. I repeated the experiment a couple of times, and found more relation between cards than I expected from my ZK.
What I found intriguing with the Luhmann system (and also btw previously systems I have used, such as Roam Research), is the step away from viewing my note collection as categorised information (or ideas). I have seldom enjoyed the process of finding information through searhing among well-named folders based on topics. Im looking forward though to put some more time into researching some alternatives. May I ask, what system are you using?
The background of this card is unique to this case. I had a cutsie idea that explains in a concise way what I don't like about modeling information with RDF (Resource Description Format). RDF expresses everything as "triples" - subject-predicate-object. My idea was to title my critique "Can't - A Critique Of Pure Triples". Afterwards, I got to thinking about Kant's actual work. It's been a very long time since I read any Kant, and not much stuck, and I thought I would try to understand one small specific topic in the introduction.
The card's title reminds me of that whole context. Given another book in another context, I probably would title a card in a different way.
If you follow Luhmann's rules, that's exactly what you should expect to see. :-)
11.26 feels out of place. How is it connected to 11.25?
Luhmann's basic rule was to place a new zettel where the connection made most sense ("anschlussfähig").
a) If it works in an existing sequence, increase the number, eg 11.23 to 11.25 in your example.
b) If it doesn't connect well to that sequence, branch off a new branch. If none of the branches fits, choose one of your basic categories as a starting point.
In your example, I'm curious what note 11.1 is about. If it is about religion, christianity or reformation, I would have expected it in category 12, may like this:
12 History of religion
12.1 Christianity
12.1a The Catholic church is the biggest Christian church
12.1b Luther's attempt to reform Catholicism created a chism
12.2b1 Luther makes faith personal through Paulus
12.2b2 Luther makes the bible accessable to the people
12.2b3 The freedom to interpret the Bible on one's own after the Reformation creates chaos
Yin & Yang might start here:
11 History of ideas
11.1 Yin & Yang
11.1a The emperor needed to keep a close eye on the calendar
I use a digital hypertext system, implemented in Obsidian. One advantage of digital notes is that you can combine multiple organization techniques in the same system. The same zettel can be found via categorizations and links and fulltext search. The same system holds zettels and reading notes and encyclopedic entries.
Paper didn't work for me. :-(
What doesn't work well is to have a predefined set of categories. Luhmann said as much. There does need to be kind of relationship, otherwise you could never find related information. But it should grow organically out of your own understanding - your own semantics. It's perfectly sensible to have categories of various kinds, as long as they express your own notons of the connection between ideas.
Luhmann's system worked that way. For example, his case 1, section 6, was generally about equality in one way or another (my versions of his titles):
6.1 References On Equality
6.2 Different Senses Of Equality
6.3 [Equality] Conceptual Basics
6.4 Current-Day Relationship Between Equality And Justice
6.5 Equality And Ideology
6.6 Equality As A Grouping Principle
This didn't mean he never had topics about equality elsewhere. He did. For example, in another section, I think it was 14, which tended to involve "justice", he had some cards about the relation between justice and equality. Some of these cards he cross-linked to section 6 cards.
He also went off on tangents inside of section 6, though they tended to relate to the main topics there.
Luhmann couldn't run searches. All he could do was to hope to remember where various things were, and riffle through the cards in a likely section. So the cards had to relate to nearby cards in some way, even if only loosely.
I call related groupings "Semantic Neighborhoods". With digital systems, we can search. The search will be most helpful if it uncovers semantic neighborhoods relating to your search terms. If you can navigate to a semantic neighborhood, and you can see or scan the nearby cards, you have a chance to notice cards which your search didn't hit but are in the same neighborhood.
A good search will tend to get hits in other semantic neighborhoods besides the one you may have had in mind. Then you may get interested in why you got a hit on them and that in turn can spark off interesting new thoughts and directions.
All the above is separate from any links you may have in a card.
It is good to realize that just because you have a heading or category, there is no reason why all the cards "under" it have to have the same kind of relationship with it. It is enough that you think they have some kind of relation, even if you can hardly articulate it. Here's an example:
This how people in the wild write things down. Obviously Cooking is not a kind of Food, though Ingredients probably are. Baking may be a kind of Cooking but it isn't the same kind of thing as Tools. Still, to me this indented list makes sense. Maybe not to you, but some other arrangement will. That's how to look at organization. It's a help for finding and relating things but doesn't have to be strict in some formal sense. It's nothing to agonize over.
Oh, and the organization had better be malleable, because your ideas of the relationships will change over time. Luhmann couldn't change his basic organization but we can - if our system makes it easy.
It's not. I started out 11 as the top category "History of ideas". Meaning if I encounter any card that fits under the history of ideas, but its not related to any existing sequence within it, I will increase the number with 1 (thus, starting a new branch). In hindsight, 11.23 to 11.25 could have been put in the same sequence by numbering 11.24 instead to 11.23a (and 11.25 to 11.23b or 11.23a1). But at the time of writing, I seem to have felt that Luther making the bible accessible was not directly building on the idea of personal through Paulus.
This is also why my title mention that I might be missing out on "loose relationships". In my current system, I have no way of differentiating between no relationship (except within the top category, of which I expect only a hand full in my whole ZK), and related but not building on. At least I have no way of doing so with card juxtification solely.
Putting a card "under" another, means buildding on. Putting it "beside another" means unrelated (but within the same top category).
This is what I did in my example, all cards except 11.26a are new branches.
Since every number under 11 are new branches, maybe you are not interested in 11.1 after all, but if still you are its: Controvery surrounding Jesus resurrection
Many of my cards fit in both 11 and 12, since the history of ideas often include religious thinking. Your example for ordering my cards is interesting though. I realise I am mixing different systems now, but going with the recommendation to not create folder cards, I would never be able to create the sequence you propose.
Lets say I encounter the idea "Luther's attempt to reform Catholicism created a chism". I would not necessarily know if this is an idea that fits under "Christianity" or under "Chisms" or "Reformatons" or "Luther" etc. I would either have to decide in the moment and create folder cards (like your 12.1 Christianity), which I have seen recommendations against doing, or have "Luther's attempt to reform Catholicism created a chism" as a new branch under 12.
This however, leads to very specific ideas being at the top most layer. And - again - since I have no way of differentiating between "directly builds on" and "somewhat related to", my ZK becomes quite scattered. The only reason I have 3 Luther cards back to back in my example is because they come from the same pages in the same book. These might as well have been hundreds of cards apart, if I encountered the ideas far apart in time. In this case, I would likely have found "Luther" in my index, and created links by writing them out on the cards themselves. But I would miss out on the chance to have them linked by positioning.
It seems to me that you are using Folgezettel to structure your Zettelkasten. If this is true, I'd like to remember that Folgezettel is not for structuring, and the very weakness of Folgezettel is that it resists being used for structure.
This follows from the problem of multiple storage. A note should appear in multiple positions. Luhmann solved this by linking and consequently wrote that the position of a note is not important.
I am a Zettler
Would you mind sharing some photos of your zettelkasten, @emiwid? I'd like to better understand, what images you have in mind, when you talk about zettelkasten. This would be interesting:
When working with paper, the spatial arrangement of design elements and physical objects can make a big difference.
Luhman used folder cards like that, but not consistently. I personally don't see why not to use them. There's nothing to say that just because you have a folder card as a guide that the following cards all have to be perfectly consistent with it. I wouldn't link a z-card to a folder card, though. And after all, if you decide one isn't working for you, you can always ignore it or even throw it away (if you haven't created any links to it).
Your ideas of what those relationships are can change over time. It's better not too be too specific about them, especially in a paper-based system you can't change later.
Yes, and this is a real weakness of a paper-based system. There are a few ways to deal with this situation.
Write better card titles. Skimming through cards near 11.25, let's say, you can mainly take in the titles. Good titles will help you see when a card sequence is going off on a tangent. Then you can skip over those cards, or look in them for links to more material on this tangent.
Put links to a more relevant section right after the title, so you can see them while scanning. I would leave a space below the title where you could add links like that later on, after you first wrote the card. In the example of the Yin and Yang card sequence, you could decide to have a grouping somewhere else on Chinese religion and philosophy, and add a link to one of those cards. Next time you scan the cards near 11.25, you'll see the Yin and Yang card and also spot the link to, say, section 15.38, Scan through the cards around 15.38 and you will see why you made the connection.
One or more cards near 15.38 should have a link back to some card in 11.25. With this scheme, your index card could refer to 15.38, 11.25, or preferably both. If your more general links are placed where you can see them while scanning, I think you will learn to move around and find what you hope to find. A digital system could make this all work better, but of course you have definite and good reasons for wanting to stick with paper.
Do what Luhmann did and if you don't want those later cards to be hundred of cards away, insert new sequence subdivision numbers at the point you wish you could put the new cards. Don't be too picky about what these sub-sequences "mean", just let the card titles tell you. They represent some of the semantics that you bring to the party.
Crafting good titles is an important part of the craft of zettling and it takes some thought and practice.
Yes, if I understand Folgezettel correctly, this is what I use
Sure, here I will upload some. To also answer your questions:
1. All cards are the same size, and I use nothing other than A6 cards, except the organazing letters (which are blank).
2. I did upload one of my longest sequences to address this question, but in truth, I have never used my cards for any output yet. I'm not sure how I would lay them out. Sometimes when inserting a card, I put a few out like this (maybe not as nicely aligned), but I cannot say I have a system for it.
3. Usually only text, but I do have a few symbols (one is shown in the image). Most cards only have text on the front, but in some cases I have experimented with references to other cards on the back.
Note: there is no difference between "bib"-notes and "bib"-cards. I mistyped.
Could have been more instructive in my cards were in engligh, but at least you can see the structure here.
Alright, all these 3 points are really useful ideas. I especially like the second one, and I will try to experiment with this more, how I can make links to other cards more accessible while scanning.
Also, being less picky about when to continue a sequence might be the right idea here. I might have been too picky about that the idea in a sequence should build upon (and not just be somewhat related to) the previous card in the sequence.
And regarding this, I do feel that there is some special characteristics of the first layer in that there is no card "above" it (well, again, except for the top-level folder card). Every card placed in layer 2 or below will have some decided relation to at least three other z-cards, as opposed to only two cards for layer 1. This makes most cards that reach level 2 and below a little more contextual, and interesting.
Thanks for these excellent pictures! I see that you're following Scheper's advice closely.1 All card types look good. You only seem to struggle with Luhmann's numbering system.
Personally, I had to examine Luhmann's original cards2 in order to get a feeling of how the numbering system works. Here in the forum you can find posts, where I experimented with facsimiles.34 You might notice that Luhmann's zettels don't look like Scheper's. Luhmann's zettels typically have no titles. They are are not atomic. They have no fixed layout. They can contain anything. However, they all somehow connect to a previous zettel.
Several people have tried to understand and explain Luhmann's zettelkasten.567 Some authors have developed simplified versions.89
I believe Luhmann's system is so hard to explain, because it wasn't intended for the general public. Luhmann developed the system in the early 1950s for his own personal needs. He needed a system that worked for him, for his way of thinking and his own long-term projects. The system is tailored to one person. We can only guess, what happened in Luhmann's brain, when he worked with his system. He left some clues, but not a user-friendly tutorial.
So you have to figure it out for yourself, what works for you. :-) I suggest an experiment:
I did similar experiments with my digital zettelkasten, where I can move zettels around without having to update them manually. At some point it clicked. It's hard to describe. Adding notes to a note sequence feels like adding notes to a notebook, not like inserting cards into a card index. I wish I could explain this intuitive quality better. And I wish I could ask Luhmann himself and compare experiences. Maybe he had the same experience. Maybe he had a completely different experience. We don't know.
I'm curious how you experience this experiment. Did it change something?
---
See also: https://zettelkasten.de/posts/introduction-antinet-zettelkasten/ ↩︎
https://niklas-luhmann-archiv.de/bestand/zettelkasten/suche ↩︎
https://forum.zettelkasten.de/discussion/3425/visualizing-luhmanns-folgezettel-in-logseq/p1 ↩︎
https://forum.zettelkasten.de/discussion/3473/facsimiles-of-luhmann-9-8-zettels-laid-out-on-canvas ↩︎
Kommunikation mit Zettelkästen: https://web.archive.org/web/20240629071815/https://luhmann.surge.sh/communicating-with-slip-boxes https://zettelkasten.de/communications-with-zettelkastens/ ↩︎
https://www.researchgate.net/publication/328624186_Where_Does_Niklas_Luhmann's_Card_Index_Come_From ↩︎
Schmidt: Niklas Luhmann’s Card Index: Thinking Tool, Communication Partner, Publication Machine, 2016 https://pub.uni-bielefeld.de/record/2942475 ↩︎
https://writing.bobdoto.computer/how-to-use-folgezettel-in-your-zettelkasten-everything-you-need-to-know-to-get-started/ ↩︎
Sönke Ahrens: How to take smart notes. ↩︎
I have studied some of his sequences and cards, and I pretty much agree. I think he was willing to violate any "rule" that we may think we notice, leading to confusion among the rest of us. He could do that, I'm sure, because it made sense to him in context. For example, if he ran out of space in a card, he would leave it in mid sentence and pick up on the next card. He didn't need to put a link or even an arrow at the end of the first card because he knew that being in mid-sentence meant the text continued, and he knew how he numbered in that situation.
Sure, @harr's right, we don't know much for certain, but that's a case that seems pretty clear. There were some other link indications I never quite figured out.
There's a moral here. Your system needs to work for you, not anyone else. You can feel free to deviate from what anybody thinks Luhmann would have done, or what anyone else thinks you should do. That doesn't mean that any way is as good as any other, though. It means to learn from others and from what you want to achieve in the long run. For example, with a paper based system you absolutely must be able to find cards without searching for words in them. You probably want to be able to spot similarities when you riffle through a handful (that's why I emphasize the value of good titles).
Yes! Ironically, this might be a valuable lesson to learn from Luhmann. He did is own thing. :-)