About Thinking Notes • Zettelkasten Method
About Thinking Notes • Zettelkasten Method
You can call many things a note, but only some are worth keeping. Others are engagement notes, or thinking notes, which you need to understand something, but not more.
Howdy, Stranger!
Comments
For completeness: The Reddit Post:
Dear Zettlers,
Understand Thinking Notes to Clear Up Your Workflow
Math and the other hard sciences are particularly resistant to the Zettelkasten Method. A common issue I get from mathematicians and physicists is that they truly have a hard time figuring out what belongs in your Zettelkasten.
The reason is that in mathematics and physics, you create exceptionally many engagement and thinking notes. Most of these notes don't fit well in your Zettelkasten.
For non-mathematicians, I label these notes often with pre-processing to actually get started with what most typically associates with actually working with knowledge. The problem is that both activities are labelled with terms like "exploration", "thinking on paper", and similar terms.
If you want the Zettelkasten to serve you as a thinking environment, you'll have to figure out which thinking steps can be done in your ZK, which will have to happen before, and which thinking steps will happen outside. An additional question is whether and, if so, how to feed back the results of thinking that happens while using your notes to write.
I don't think that you can think about mathematical problems properly in a digital environment. At least, this is the overwhelming consensus of the mathematicians I spoke with.
Thinking on screen, however, is very possible in other domains that don't require you to submit to an alien, artificial language like math. Rewriting belongs in the realm of thinking and can be done on screen and, therefore, can be done in your Zettelkasten.
To make your Zettelkasten your integrated thinking environment, you have to move thinking into your Zettelkasten. Atomicity then becomes the output of the Zettelkasten work and not an input function. Or: You will have a lot of unfinished non-atomic notes in your Zettelkasten.
Happy reading: Understand Thinking Notes to Clear Up Your Workflow
Live long and prosper
Sascha
I am a Zettler
What is the benefit of moving thinking into a Zettelkasten?
And what exactly do you mean by Zettelkasten?
If I understand The Complete Guide to Atomic Note-Taking correctly, your flavor of the Zettelkasten method is based on this claim:
What exactly are these building blocks? You list an inventory with six items ranging from concepts to theories. Then you say:
But you also say:
I have difficulties to put the pieces together. How can a building block be as complex as a whole theory, but also be as simple as just one idea?
I'm struggling to understand, what exactly you consider an "idea".
The article mentions several aspects of an idea. They can be atomized, so that you get idea atoms, but it's also possible to break down an idea into elements. Ideas can have an essence, but they can be incomplete. Ideas can be processed, refined, developed. It's possible to "get into direct contact with ideas". There is mastery. Ideas can be the result of a process, but also exist at the very beginning of a process. They can have several concepts. Ideas can be built upon.
My brain has difficulties to integrate all these aspects into one concept. I have no idea, what an idea is.
I agree.
Good question.
I agree.
A definition would really help. But for some reasons I can't find one in the article. I must have overlooked it. Would you mind summarizing in a few short sentences, what an idea is?
Clear definitions would make it easier to appreciate what a Zettelkasten has to offer.
How would you write an atomic note that captures the idea of an idea?
@ZettelDistraction I was curious about how you use Zettelkasten in this regard.
@Sascha I'm currently experimenting with Zettelkasten for music production and mixing. So far, what I've been able to achieve with Zettelkasten is (1) creating models for melody, chord progressions, song structures, and selecting sound samples to describe the feeling; (2) learning new techniques; and (3) transcribing the "knowing how" knowledge into explicit sentences to discover or refine new techniques.
However, I'm wondering if the music production field can fully benefit from Zettelkasten. Have any of the people you've coached used Zettelkasten in their music?
@harr said:
This has been extensively discussed in the forum. Search for "one idea".
"One idea per note" is the vague, intuitive way of stating the principle of atomicity. That was @Sascha's starting point at the top of "The Complete Guide to Atomic Note-Taking". The vague principle is then replaced, in the section "What Is Atomicity?", with the more precise principle of "One type of knowledge element per note" (in my words). (This is more like a Carnapian explication of "one idea" than like a definition of "idea".) Sascha provides his own inventory of types in that section, but everyone who follows something like this principle seems to have their own set of such types.
EDIT: To answer the question "How can a building block be as complex as a whole theory, but also be as simple as just one idea?": Note that Sascha's definition of theory is not what you may be thinking. For him, theory means a hypothesis (one statement) "with an inventory of methods". So it's not very complex. But in principle a type of knowledge element could be internally complex, as long as it is well-defined enough to be useful as a building block: for example, a design pattern.
I didn't find it. Hence my question. Sascha repeatedly says, that the definition matters. But my searching skills are not sufficient to find an actual definition.
No, this is much simpler. I just want to know Sascha's personal definition of a term that is so central to his method. I think this is a fair question, because Sascha himself asked rhetorical questions about definitions.
I just noticed that an external link in the previous post got lost during the editing process, sorry! The question
is from The Principle of Atomicity – On the Difference Between a Principle and Its Implementation. It's one of the articles I unsuccessfully checked while looking for an answer. This question isn't answered, but followed up with another rhetorical question:
Again no direct answer. Instead:
So… are idea, building block and atom the same thing?
I'm working from the hypothesis, that Sascha talks about the thing modularity, when he says the word atomicity. Sascha's concept of a building block with internal complexity would make sense if he were talking about a modular system, that has components and sub-components. The way he talks about ideas would make more sense, if he were trying to define clear boundaries and relations between modules. If you think of notes as modules, then their inner complexity becomes a matter of choice.
Note-taking becomes quite simple, when you clearly distinguish between ideas and building blocks, and when you think of building blocks as modules instead of atoms. It also helps to distinguish between those concepts and the scope of a note. The principle 1 note = 1 idea/concept/source/… is good rule of thumb. But sometimes it's more practical to keep several ideas in one note as "potential" components, knowing that they could be disassembled if needed.
In my current understanding of Sascha's writing, everything blurs together.
@harr said:
I was merely pointing out that this issue has been extensively discussed, including the relation between the words atomicity, granularity, and modularity. I'll point to a definition of idea below.
That post was written before the "Complete Guide to Atomic Note-Taking". In the latter, he said: "I define thinking as the process; thoughts are frozen processes, and ideas are the result of thinking." That's a definition of idea. It's a vague definition; idea as used in everyday language is vague, as can be seen in typical dictionary definitions. I think that vagueness is why he replaced the principle "One idea per note" with something more precise. In the "Complete Guide" he also defined atomic as "contain[ing] exactly one knowledge building block" (I would say: "conforming to a type of knowledge element").
Yes, as I've said before, modularity is another way of talking about it. But the context of Sascha's "Complete Guide to Atomic Note-Taking", as he said in the first sentence, is that "atomic notes" had become a buzzword, so atom/atomic instead of module/modular is the term he's using.
As I understand him, Sascha does "distinguish between ideas and building blocks", but I think what he means by a "building block" is a type (a type of knowledge element, in my words). As I said, I think module and atom are interchangeable terms, but it's not right (in Sascha's scheme, as I understand it) to "think of building blocks as modules instead of atoms"; rather, building blocks are types of modules/atoms, which are instances of the types.
Remember that English is not his first language, and he does not have a professional editor. I like reading his perspective even when it is not perfectly clear. It's fun to think through it.
EDIT: @harr edited his comment between when I started writing this comment and when I posted it. So the quoted text is from a previous version of his previous comment.
@harr asked:
No. How's this for a summary: When a note contains an idea (i.e. the note is not empty), and the idea conforms to a building block (i.e. type), then the note is an atomic note or atom (or module if you prefer).
Thanks for your replies!
It sounds like a definition, but what does it actually say about ideas? Not much.
And it contradicts other parts of this and other articles, where you start thinking by writing down an idea and then continuously improve that idea.
The sentence describes an aspect of his understanding of "idea". But it's just one aspect of many.
Yeah, I'd also compare them to types. But I'm not even sure what types:
One article lists: Concepts, arguments, counter-arguments, models, hypotheses and theories, empirical observations
Another article lists: Definitions, arguments, counterarguments, models, theories, facts.
I appreciate that definitions appear in one of those lists. Definitions are an essential type in my own note-making. When I'm trying to understand someone's thinking, I start by researching, how that person defines relevant terms in their words.
I'm working on a note that contains the sentence: "Sascha defines ideas as … (Source: …)"
I agree with not perfectly clear and fun. :-)
@harr said:
You're welcome! Thanks for the good questions.
I'd say the contradiction here is only apparent: there are really (at least) two types of thinking: thinking-in-your-head and thinking-with-pen-and-paper (or thinking with a digital device). You start thinking-with-pen-and-paper by writing down an idea, and then you continue thinking-with-pen-and-paper by writing down more ideas. But you do thinking-in-your-head before writing down the first idea and throughout the process of thinking-with-pen-and-paper. This is all compatible with the definition of ideas as the (intermediate) results of thinking. In popular online language, writing down ideas could also be called "content creation": creating the content of notes.
The exact list of types isn't important to the general principle of atomicity.
Here's a reference for anyone who might be reading this thread. The Principle of Atomicity – On the Difference Between a Principle and Its Implementation:
EDIT: My own brain wasn't intelligent enough to navigate Sascha's writing. So I asked artificial intelligence to weed out distracting asides and imagery. I think I found a partial answer to my question in the same article:
So we have (at least) two kinds of atomicity on zettelkasten.de:
Dijkstra defines separation of concerns as a "technique for effective ordering of one's thoughts" that helps examine various viewpoints and aspects by "focussing one's attention upon some aspect" of a thing. For Dijkstra a characteristic of intelligent thinking is "to study in depth an aspect of one's subject matter in isolation for the sake of its own consistency, all the time knowing that one is occupying oneself only with one of the aspects."
Sascha seems to be more interest in identifying units of knowledge, that have an essence, can be isolated and come in one of six shapes.
@harr said:
@Sascha's conception isn't idiosyncratic, only some of the language is; conceptually, it is well thought out and, at what he calls the "Level 4" understanding of thinking and knowledge, is roughly congruent with the research on structured argumentation in personal knowledge bases done back in the 1980s. The "Complete Guide to Atomic Note-Taking" is a big improvement over Christian's explanation of atomicity in 2013. Dijkstra's separation of concerns, applied to Zettelkasten work, is like a "Level 2 to Level 3" understanding: pretty good, but not as elaborated as Level 4.
From the "Complete Guide":
It strikes me here, @harr, that you may not be grasping the four levels of Sascha's iceberg model. Atomicity has a different meaning at each level (or no meaning at all at the first level). The "Complete Guide to Atomic Note-Taking" is not a laundry list of incoherent conceptions of atomicity; it's a progression from more informal conceptions to a more systematized and analytical conception.
Again, Sascha's particular set of building blocks, and the number of them, is not what is important, so don't fixate on that. The point is that at Level 4 you're applying what you've learned about the nature of knowledge in general, and in particular the nature of knowledge in your discipline. And you may not even need a Level 4 conception of atomicity; if you find that Dijkstra's separation of concerns makes more sense, then perhaps all you need is a "Level 2 to Level 3" way of working.
OK, maybe we need a name for the thing we talk about, so that we can compare it more easily to other things.
The thing is a system of thinking, that was invented, designed and authored by Sascha Fast. Let's call it "Sascha Fast's System" or "SAFAS".
Sascha is the creator. SAFAS is his creation. The ultimate authority for this system is its author Sascha. If Sascha declares the system to have a certain number of building blocks or levels, then it is so. If Sascha declares the system to also have engagement notes and thinking notes, as in the original post of this thread, then it is so.
Let's compare SAFAS to other systems of thinking.
No, Christian didn't explain atomicity. He made up a name for a thing that he finds interesting. He was looking at a principle and gave it a name:
I agree with Sascha that we shouldn't talk about atomicity like Wittengstein's beetle in the box. We shouldn't assume that we talk about the same thing, just because we use the same word.
Unlike Sascha I'm content with comparing concepts as they are defined by their authors. So in this case we have two different definitions that describe two different things:
I think the iceberg is upside-down. Abstraction should be up, not down. Why use an iceberg at all? It's a bad visual metaphor for levels. The iceberg is a great metaphor, if you want to point to hidden stuff. But for levels I'd go with a simple pyramid. Simple basic stuff is at the bottom. Higher levels of thinking are up.
I don't know what atomicity means. But I'm willing to learn and understand how SAFAS defines different levels of atomicity. I'm quoting from the Guide:
The language in the Guide is a bit vague for my taste. My brain needs something more concrete. Not sure if I understand Sascha correctly. I was distracted by icebergs and divers. :-) Here's my attempt to describe the levels in a more practical manner:
Level 3 and 4 are unique to SAFAS.
The "building blocks" are unique to SAFAS. They seem to be the philosophical core of Sascha's idiosyncratic thinking. The Guide mentions blocks in Level 3.
If I understand SAFAS correctly, Levels 1 and 2 are what you do with any kind of organized note-making, be it a Luhmannian or Ahrensian Zettelkasten or any other system that has some rules on how you organize your notes. I think that for most people and purposes Level 1 and 2 are just fine. Personally, I'm very happy with my Level 2 system. It solves the problems I care about. :-)
I'm curious how many people share Sascha's interest in the abstract notion of "knowledge" and higher "levels" of "atomicity". I'm curious how many people implemented SAFAS as Sascha intended and what experiences they made.
I consistently reach level three with most notes. Since I started defining the atoms I feel that my reasoning is sharper.
The building blocks I use the most are concepts, models and arguments.
At what point in your note writing do you think about building blocks?
Does it help to think of those blocks as an advanced level?
I agree that being clear about the type and scope of a note improves thinking. In my note writing process, choosing a note type is the very first step. I have templates for several note types. When I create a new note, I pick one. That's it. What am I missing?
Christian's usage of these building blocks in teaching makes sense to me. He presents them "as a starting point to define & recognize elements of information in the student’s sources". Personally I wouldn't use that particular list. But I agree that it helps to know, what you're looking for in a text. But that would also be a starting point, not a more advanced level.
I'm still trying to understand Sascha's conceptualization of "levels"…
@harr, I could respond to a lot in what you've said, but I'm losing interest in this discussion, so I'll just respond to your suggestion that the use of knowledge building blocks is "a starting point, not a more advanced level".
If you already have an understanding of what knowledge building blocks are, then it may seem natural to use them, as if they are a starting point. But if you were teaching the Zettelkasten Method to a child, and you told her to aim (at least after some refactoring) for one knowledge building block per note, she would not know what you were talking about. And even the early readers of the early versions of this website's "Introduction to the Zettelkasten Method" (first archived version in the Internet Archive) would not know what you were talking about, because there is no mention of "knowledge building blocks" in it; the only "building blocks" mentioned are Zettels.
So first you have to learn what knowledge building blocks are. And the knowledge building blocks that Sascha is talking about are based on scientific or systematic thinking. Here I think of Paul Hoyningen-Huene's book Systematicity: The Nature of Science (Oxford UP, 2013), which says that science is knowledge that has a higher level of systematicity in dimensions such as (copying this list exactly from the book):
Sascha's Level 4, as I understand it, is a "higher" (or "deeper" in the iceberg theory) level exactly in the way that Hoyningen-Huene said that science has a higher level of systematicity than other kinds of knowledge. That is, when you are using knowledge building blocks in a scientific or systematic Wissenschaftliche way, then you are operating at Level 4. You then bring that way of thinking to your work from the beginning, so it seems like a starting point, but it is an advanced level insofar as you have to learn scientific or systematic thinking first. And if you're not doing scientific or Wissenschaftliche work with your Zettelkasten, then you don't need this advanced level.
@Andy I am always amazed of the level of your Bildung. Particularly, I am surprised that you pointed to Carnap which is spot on. The building blocks are indeed a try of an explication of the vague concept of "idea". I told both Christian and my wife that I am amazed that you sniped that.
EDIT 1: My claim is that the building blocks are a worthy first draft of universal knowledge building blocks. Empirically, they worked and are stable since a long time. But I lack the systematic justification for their universal nature. So, I push back a little bit that they don't matter if you use them. However, as long as I don't have a systematic justification, I will just use my outrage if somebody doubts its universal nature.
EDIT 2: @harr Andy explained my position even better than I would've explained myself. I have little to add.
I am a Zettler
Thanks for the argumentum ad passiones. It answers the question, how much those building blocks matter to you. :-)
Sorry, I missed your comment.
Not one.
(You got a PM)
I am a Zettler
For math, a top-down approach is helpful. I record whatever seems salient at the time. Zettelkasten work for outlines, comments on talks, etc.An example, "The speaker wrote a representation of the diffeomorphism groupoid of opens but should have written $(\operatorname{Open}(M)^\mathrm{op}\rightarrow\operatorname{Vect})$."
GitHub. Erdős #2. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein. Alter ego: Erel Dogg (not the first). CC BY-SA 4.0.
Does a top-down approach mean delving deeper into the knowledge you already have?
And the example you gave seems similar in nature to engagement notes based on Sascha's classification criteria. Does Zetelkasten generate new ideas in areas like mathematics?
It means working from overviews to details:
Not in my experience. It's a fragmentary record. For working top-down, a Zettelkasten can provide scaffolding. You start with an outline, statements of theorems, outlines of proofs, etc., down to details. What you don't know yet leave open as an unresolved or "unevaluated" link. I haven't been consistent with this. The Zettelkasten is just a file cabinet, for me.
I can't speak to idea generation. As for what experts do, research suggests organizing what you know (not to mention what you ought to know) into a predictive framework, as described in "How to become a successful physicist" by Carl Wieman. Wieman/Price argue (in science & engineering) that experts organize knowledge into predictive frameworks for decision-making; I suspect this is true for mathematics, but I am unaware of studies either way. As for what one should know in physics, there is Gerard 't Hooft's site, but I am not a physicist.
Disclaimer:
I haven’t turned Wieman/Price into a Zettelkasten workflow (and I’m extrapolating from science/engineering to math); for retention, spaced self-testing beats rereading. Links aid navigation but don't substitute for retrieval practice.
GitHub. Erdős #2. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein. Alter ego: Erel Dogg (not the first). CC BY-SA 4.0.
@ZettelDistraction Thank you for sharing your experience.