Zettelkasten or topics: how to link notes across courses

Bhalendu Chaubey · November 13, 2026

What a Zettelkasten is, how it differs from organising notes by topic, and a practical way for learners to link ideas across courses without a big system.

A Zettelkasten is a collection of short notes, each holding one idea in your own words, linked to each other by hand, so that ideas from different sources connect. Organising by topic is simpler: you file each note under the subject it belongs to. For most learners taking courses, the best approach is a mix of the two: topics to keep each course in order, and a few deliberate links between notes when an idea in one place explains, contradicts or applies to an idea in another. This article explains both methods, when each one suits you, and a short weekly routine for linking notes across courses.

The answer in brief

What is a Zettelkasten?

Zettelkasten is German for "slip box" or "card index". The best-known example belonged to the sociologist Niklas Luhmann. According to Bielefeld University's Luhmann archive, it holds about 90,000 note slips written between 1951 and 1996, and it was central to how he developed and published his theory. The Niklas Luhmann Archive makes parts of it available online.

The principles most people take from it are:

Over time, the links matter as much as the notes. A cluster of linked notes on one idea becomes a line of thought you can follow.

What is organising by topic?

Organising by topic means filing each note under a subject: "Sampling", "Loops", "Cloud security". A topic can be a folder, a tag or a heading. When you want to study sampling, you open the topic and everything about it is there.

Most course material is already organised this way. A syllabus is a list of topics in order, and an exam outline is a list of topics with weights. Following that structure in your notes makes it straightforward to see what you have covered and to review one subject at a time.

Zettelkasten versus topics: the differences

ZettelkastenTopics
UnitOne idea per noteAny note, filed under a subject
StructureA network of links between notesGroups, sometimes nested
How you find thingsFollow links from an entry pointOpen the topic
Best forDeveloping your own ideas across many sources, often for writingLearning a course or a syllabus, and reviewing by subject
EffortHigh: every link is a decisionLow: one decision per note
Main riskSpending more time on the system than on learningIdeas that span topics stay apart

Neither is better in general. They answer different questions. Topics answer "what do I know about X?". Links answer "what does X have to do with Y?".

Which suits a learner?

If you are taking a course, studying for an exam or working through a textbook, you mostly need to learn a defined body of material and recall it later. Topics fit that work. They match the syllabus, they let you check coverage, and they give you a natural unit for review.

A full Zettelkasten pays off most for people developing their own ideas over years, often towards writing: researchers, writers, students on a long thesis. For a twelve-week course, building one can turn into a project of its own.

But one part of the Zettelkasten idea is valuable for any learner: deliberate links between ideas in different places. The same idea often turns up in two courses in different forms. A statistics course talks about sampling bias; a machine learning course talks about training data that doesn't represent the real world. They are the same problem. Seeing that link helps you understand both.

Why linking helps learning

Connecting a new idea to something you already know is a form of elaboration. The 2013 review of study techniques by Dunlosky and colleagues rated two techniques of this kind, elaborative interrogation (asking why a fact is true) and self-explanation (explaining how new information relates to what you know), as moderately useful. The review's highest ratings went to practice testing and spaced practice. So links are a useful addition to recall and spacing, not a replacement for them.

This routine takes about 15 minutes a week and keeps linking deliberate. It assumes you already file notes by course and topic as you go.

  1. Pick this week's key ideas. From each course, choose one or two notes you would most want to keep: the ideas you would put in a summary.

  2. For each key idea, ask where else you have seen it. In another course, another topic, a book, your work. Search your notes for a few words from it.

  3. When you find a match, write the link with a reason. One line: "Same problem as sampling bias in Statistics, week 3: the data doesn't represent the population."

  4. Check the link goes both ways. When you open either note later, you should see the other.

  5. Turn the strongest links into questions. "How is training-data bias in machine learning like sampling bias in statistics?" is an excellent recall question, because it asks you to bring back two ideas and the connection between them.

  6. Look at your topics once a month. Merge topics that mean the same thing, and notice topics that keep linking to each other; they may be one subject.

Weak linkBetter link
See also: LoopsThis recursion example does the same job as the while loop in lesson 6; compare which is clearer
RelatedContradicts the claim in chapter 2 that larger samples always fix bias: they don't fix a biased method
Week 3The shared responsibility model here is the same idea as the security domain in the exam outline
SimilarSame pattern as the database join in week 7: matching rows by a shared key

The reason is the valuable part. A bare link tells you two notes are connected. A link with a reason tells you how, which is the thing you are trying to learn.

Say you are taking two courses at once: an introductory statistics textbook and Microsoft's Generative AI for Beginners. In statistics, chapter 1 covers sampling and data. One of your notes, filed under the topic "Data", says: "A sample chosen for convenience can be biased: it may not represent the population, and a larger sample doesn't fix that."

A few weeks later, in the AI course, a lesson on using generative AI responsibly discusses models that give poor or unfair answers for groups that were under-represented in what they learned from. You write a note under "Responsible AI": "A model reflects the data it learned from; gaps in that data become gaps in its answers."

In your weekly routine, you notice the two notes are about the same thing. You link them, with a reason: "Same problem as sampling bias: the data doesn't represent everyone the result will be used for." Then you write a question from the link: "How is a model trained on unrepresentative data like a biased sample in a survey?"

When that question comes up for review a week later, you have to recall both ideas and the connection. It is a harder question than either note alone, and it makes each idea clearer. You would not have got there by filing the two notes under their own topics and leaving it at that.

Paper or app?

Luhmann's system was paper slips in wooden drawers, and a paper version still works: index cards, a number on each, and the numbers of related cards written on the back. An app makes three things much quicker: searching for a note to link to, showing the link on both notes, and jumping from one to the other. Whichever you use, the work that matters is the same: deciding what connects, and writing why.

How to avoid building a system instead of learning

Note systems are satisfying to build, which is also their risk. A few rules keep the system serving the learning:

For course study, a structure that works:

If you are studying for an exam, use the outline's objectives as your topics. See how to study for a certification exam from its outline. If you take notes in the Cornell layout, the cue column is a natural place to note a link: "compare week 3". See Cornell notes on paper and on screen.

Doing it in Soundings

Soundings uses topics as its main structure and links for the connections between them. While you write in a session, choose a topic above the capture box and every new entry is filed under it; a topic's page then gathers its entries from every session of the course (Organise with topics). When two courses have topics of the same name, Topics across courses shows them together. You can rename and merge topics without losing anything, and a course's Map shows its topics as circles, joined when they share entries.

To link two entries, select Link under one of them, type a few words from the other, and turn on Every course to find it in another course. Both entries then show the link, as Linked to: and Linked from:, and each takes you to the other (Link entries together). Key idea marks the notes to reread first, and a Pinboard lets you lay entries out by hand (Arrange entries on a pinboard).

Common questions

Do I need a Zettelkasten to study well?

No. For courses and exams, notes organised by topic, plus regular recall and spaced review, cover what you need. The part of the Zettelkasten idea worth borrowing is deliberate linking between ideas, which you can do in any note system.

A tag puts a note into a group: every note tagged "statistics" belongs together. A link connects two specific notes, ideally with a reason. Tags help you find notes on a subject; links help you understand how two ideas relate.

As many as there are real connections, which is usually zero to three. A note with many links often means the links are vague. If you can't write the reason for a link in one line, it probably isn't worth making.

Should I organise notes by course or by topic?

Both. Course is the natural top level while you are taking it, because that is how the material comes to you. Topics within each course follow the syllabus. Links and topics shared across courses handle ideas that turn up in more than one place.

How did Luhmann use his Zettelkasten?

Bielefeld University's archive describes about 90,000 slips, written between 1951 and 1996, kept in an ordering structure that made the collection a tool for developing his theory and writing his publications. The archive calls it a kind of intellectual autobiography. The principles people take from it, one idea per note in your own words, linked by hand, come from how that collection was built.

More from the blog

Something missing or unclear? Tell us on the contact page.