Create Zettel from Reading Notes According to the Principle of Atomicity

System

Create Zettel from Reading Notes According to the Principle of Atomicity

The Zettelkasten note-taking method has made book writing and writing scientific papers easy for hundreds of years already.

Read the full story here

Annabella

Are these overviews structure notes?

ctietze

Meh, the post from 2013 is very light on detail in that regard. That could use some updates

Back in 2013, Sascha hadn't formalized the notion of a 'structure note', yet, so back then I didn't think of these. But they are good candidates for overviews.

The example in the post I give is to create a Mind Map -- it's an overview so you can find your way around e.g. core concepts in the book. Stuff that doesn't fit in the margins of a book because it spans the book's entirety. I believe I had informal overviews on sheets of paper in mind because I'm such a scatterbrain.

A structure note for the book is a good alternative because you get the overview plus it's part of the Zettelkasten.

The screenshot in the post doesn't show a structure note for this book. Given the fact that the same screenshot shows 20 notes referencing this book, that might've been a good idea in hindsight. -- Back in 2013 or so I might've asked myself: but do I really need a long-lasting overview of this book? Is it that important to compete with my 3 or 4 book structure notes of that time? Today, I wouldn't worry, because having the overview doesn't hurt. In my video series on "Range" by Epstein I started with a book overview note to collect assorted material as I went, so you see there's been progress in my approach.

Annabella

@ctietze

I'll look forward to the update then.

A suggestion: Consider including Sascha Fast's "method" to create atomic Zettels. Thinking about idea types and their structures can help in following the principle too.

p_nguyen

@ctietze said:

A structure note for the book is a good alternative because you get the overview plus it's part of the Zettelkasten.

In fact with a digital Zettelkasten you can combine exzerpt-like literature-notes with an outline of the resulting atomic notes (Literature-note can be analog, but you have to create a digital Zettel for the source). Atomic notes having one big disadvantage from my point of view: their information are completely removed from the original context. To overcome this I have a subsection called "Verzettelung" in every literature note where I copy my literature notes inside (with analog literature-notes I skip this) and process them afterwards. With this workflow the exzerpt keeps untouched and allows me to come back to the complete argumentation. I'm doing this for a while now and it helps me to get complicated argumentations much quicker than with atomic notes when I come back to them after a year or so.

ctietze

@p_nguyen So (pretending it's all digital for simplicity) you end up 1 long-ish excerpt and put links to your different 'literature notes' inside there?

kavinkumar

Hi Christian,

Your thoughts were resonated a lot with mine. I enjoyed reading it.

I couldn't understand one line from it. Could you explain a bit further in other words if possible?

Under section "Phase 3: Write notes from clusters"

and this is the line:
1) "I can re-use the arguments without buying into the assumptions because the arguments are of sufficiently general form. "

and one more sentence:

2) "The underlying principle I’d call the principle of atomicity: put things which belong together in a Zettel, but try to separate concerns from one another"

Could you elaborate the separate concerns please?

Thanks,
Kavin

ctietze

Let's see:

1) "I can re-use the arguments without buying into the assumptions because the arguments are of sufficiently general form. "

First: I don't remember why I picked that example

This is a very convoluted way of saying: I don't need to agree with the point to make use of it. Agreeing to the premises of an argument on a content level is not required to talk about an argument, to reference it, to use it as an example, or to refute it.

@Sascha in other places here uses syllogisms as an example of "forms" or argument structures. See here for a couple examples to get a feeling for various if-this-then-that forms: https://en.wikipedia.org/wiki/Syllogism

2) "The underlying principle I’d call the principle of atomicity: put things which belong together in a Zettel, but try to separate concerns from one another"

Could you elaborate the separate concerns please?

That's a reference to a design principle in computer programming called "separation of concerns": https://en.wikipedia.org/wiki/Separation_of_concerns

Here's what I think are better examples than the Wikipedia page.

Take an e-mail program. Gmail web app, Mail.app for Mac, Outlook/Windows Live Email, doesn't matter. Its source code contains pieces for different concerns or responsibilities; these could be: composing a rich-text email, managing email attachments, uploading attachments, sending the email text, or quoting original email text in a reply. -- As a user, you often don't think of these as separate pieces of the action, because you want to "send an email with 2 photos to grandma". But the internal organization of an app's code consists of things that focus on distinct concerns.

"An email program" is not a very helpful concern to reason about, but "show a list of all successfuly uploaded attachments" is.

So figuring out what a good "concern" is means to sub-divide the problem space (email app) until you get understandable things-that-can-be-done.

For Zettel-making, this translates to e.g. the following: a note on "Everything I know" likely won't remain useful. The note 202207181859 Life, The Universe, and Everything is too broad to be sensible, or even of passing value, if meant in all honesty. It would have to answer, well, everything. (Just like "make an email app" is not a single action you can 'do'.) -- If you understand the Hitchiker's Guide to the Galaxy reference hidden in this example, you can talk about that part from the book, though; The Question on Life, The Universe, and Everything in Douglas Adams books can be a discussion of ideas from the books.

---

To tackle an unknown space, I explore it to get to singular "concerns". Take for example 'Aquaponics'. You can look up the term and see it's combining aquaculture (growing fish) with hydroponics (growing edible plants without soil).

That's maybe a good first definition of the term. That would be the "concern": term definition.

But I want to know how to set-up an aquaponics system. So my structure note aka overview gathers everything I know about aquaponics. That's its "concern": to gather everything I know about aquaponics in a structured way so I get to the details easily. It's a hub for navigation.

Another note's concern is e.g. 202205021417 Environment conditions for nitrifying bacteria in aquaponics. It lists these conditions, like temperature, pH levels, dissolved oxygen per liter, ... to make the bacteria you need for the system to grow.

This is referenced by 202205021411 Nitrogen flow chart in aquaponic system; it's concern is to visualize the flow as a memory aid, and to help me reason about the flow of components in the system from a high-level view.

Another note captures findings on how carp feeding influences the carp's Omega 3 ratio, and how to optimize for high levels, and what to avoid. It teaches how to get the most Omega 3 from carps in the system.

Another note explains how the total system's capacity is limited by the water oxygen levels.

These are all small pieces, small ideas, that help me understand the whole aka the system. Each piece is self-contained and has a "concern", a purpose. I can use each piece in different systems. I can use the carp feed information for aquaponics system design and for a how-to on maintaining garden ponds, because the carp feeding info stands on its own.

In a pragmatic sense, the "concern" is the answer to the question: why do you want to link to a particular note? If you cannot answer that, the note might benefit from being overhauled.

In my personal experience, over time, I developed a sense for the kinds of reason-to-link-to-this I would like to satisfy before writing notes. (This may also depend on domain knowledge: I know a lot about many things related to computer programming, so if I approach a new programming topic, I have a rough idea of what I want to figure out to get a good first grasp of things.)

GeoEng51

@ctietze Excellent explanation - thanks.

In a pragmatic sense, the "concern" is the answer to the question: why do you want to link to a particular note? If you cannot answer that, the note might benefit from being overhauled.

I know you didn't mean to imply that there is only one reason to link notes to one another. One of the most interesting links to me, when I find them in my ZK, are links between disparate ideas. That is, the ideas are so different that you wouldn't at first expect there to be a link, but then you realize there is one.

emps

@ctietze said:
@p_nguyen So (pretending it's all digital for simplicity) you end up 1 long-ish excerpt and put links to your different 'literature notes' inside there?

I don't want to hijack, but I have a similar workflow. As I read, I copy everything I find interesting into one long note. I think it can be viewed as an analog of skimming through the text. When I've finished reading, I break the copied text into notes.
The positive side is that you can stop this process at any moment and then get back to it without losing coherence of original argument (and without shuffling through the book again): the notes about already processed parts of the book remain in order they were present in the text, the arguments that weren't processed yet remain in author's order. It's like one big pipeline.
Also, this way I can satisfy my collector's fallacy without much hassle.
Later, when the book is wholly processed, I tend to make "structure notes" in which I reshuffle notes thematically and connect them to other parts of the archive (if it weren't done already). I leave the original note with "book contents" intact.
As a downside, it's a time-consuming process. Consequently, I have some notes on books half-processed, with large batches of text that are kinda interesting, but I can't really connect to anything easily... and I find it boring to process them properly just for the sake of it.