Back to the talks Previous by track: corfu+yasnippet: Easier than I thought Next by track: Questions and answers to help you fly with Hyperbole Track: General

Zettelkasten for Regular Emacs Hackers

Christian Tietze (he) - @ctietze@mastodon.social https://christiantietze.de https://zettelkasten.de, hi@christiantietze.de

Format: 24-min talk ; Q&A: BigBlueButton conference room Etherpad: https://pad.emacsconf.org/2025-zettelkasten
Etherpad: https://pad.emacsconf.org/2025-zettelkasten
Status: Q&A to be extracted from the room recordings

Talk

00:01.400 Introduction 01:25.534 Advocating Freedoms 02:29.680 What Is This About? 04:36.534 Write - Essential Mechanic 05:09.601 Connect - Essential Mechanic 05:34.268 Correct - Essential Habit 06:49.434 Design for Use - Habit 07:43.920 Create Structure - Mechanic 08:47.968 Start in the Zettelkasten - Mechanic 09:32.401 Start with a Link - Mechanic 09:54.568 Recap 13:22.034 Facilitate Growth 14:46.140 Emacs demo 20:39.068 Learn, Share, Grow 22:45.297 Outro

Duration: 23:18 minutes

Q&A

00:49.560 Q: What do you use for the fancy animations? 03:30.080 Q: Are you not a fan of using , , headings in org-mode? 07:13.720 Q: Can you use org files and all its features inside Denote? 07:49.000 Q: Where or how do you like to capture fleeting notes? 11:29.040 Q: Zettelkasten feels like a very "cagey" approach to note-taking and knowledge management. Doesn't it restrict one to think in certain ways rather than what feels natural to someone? 15:07.328 Q: How does denote compare to org-roam? 18:25.040 Q: I noticed that the wikipedia link you wrote was typed wrongly - and it got me thinking about how to deal with broken links at scale? Do you have any thoughts on this? What about archival? 21:20.591 Comment: When I completely re-worked my config some two years ago, I also tried out some of these packages for making notes in Emacs... 24:34.800 Q: How Zettelkasten is useful for highly mathematical STEM academic fields like computer science or engineering fields? 39:40.960 Q: In your experience, would you say that you re-use most of your notes? 44:15.280 Q: How are notes structured and accessed when the notes grow from 10K to 100K notes? 48:55.160 Q: I would be very interested in your thought on this video by Westenberg - https://www.youtube.com/watch?v=CjSWwmg-JRM (Why I Deleted My Second Brain: A Journey Back to Real Thinking) 57:59.160 Q: Is there a danger that with the Zettelkasten process, that the process gets a bit in the way of the content? 01:03:20.640 Q: How do you navigate looking at all posts with certain tags

Listen to just the audio:
Duration: 1:17:07 minutes

Description

There's this one thing you can do with your Emacs that is not chiefly a technological problem to solve: thinking through writing.

Emacs offers a malleable environment where you can tweak every key stroke, and every pixel on screen to your needs.

Since we're all here at EmacsConf, the intention is clear: to use and enjoy Emacs, and spend our lives in this amazing environment.

While it's easy to ditch modern UI conveniences and pull technology like email, chat, database and server management, and editing book drafts into Emacs – well, these are tasks that have been implemented, for which there exist alternatives, and which you can teach Emacs to do in a similar way.

Oversimplifying: we can copy and tweak existing solutions and have a good life.

Now while everyone's email needs to use the same protocol, everyone's approach to thinking is different. There's no cookie cutter solution to merely rewrite in Emacs Lisp. We all need to figure out how to do this on our own, and then find an implementation that suits our needs. (Including using paper, but we're not talking about paper here.)

This is where I want to show you one simple foundational method to deep thinking, understanding, and problem solving: create yourself a Zettelkasten, an environment of linked notes that scales well over decades, so that you can take it with you into retirement and beyond for a lifelong journey of learning.

For this presentation, I merely assume that you agree that writing improves the quality and depth of thought. I also assume that you know how to type and move around in Emacs. The rest is just convention, and we'll walk through a couple of examples and exercises together so that after this talk, you're equipped with the simple tools that help you unlock new insights in your future.

About Christian

Christian is a macOS/iOS developer with a strong focus on user experience and clean architecture. Driven by a passion for accessibility and performance, Christian has shipped many apps and authored three technical books and 895+ blog posts, helping developers world-wide to realize their app ideas with a deep understand for the technologies they use. In 2013, Christian wrote about the Zettelkasten topic on his own personal blog and eventually moved everything over to https://zettelkasten.de where he and Sascha continue to teach and write.

Code and Notes from the Talk

Minimal Emacs + Denote Configuration

This is the relevant configuration Christian used in the Emacs demo. Save this to a folder as init.el, and launch via:

$ emacs -nw --init-directory . init.el

init.el contents:

(load-theme 'modus-vivendi-tinted)  ;; Dark theme to match dark slides
(menu-bar-mode -1)                  ;; Disable topmost menu bar

(package-refresh-contents)

(use-package denote
  :ensure t
  :hook (dired-mode . denote-dired-mode)
  :bind
  (("C-c n n" . denote)
   ("C-c n r" . denote-rename-file)
   ("C-c n l" . denote-link)
   ("C-c n c" . denote-link-after-creating)
   ("C-c n b" . denote-backlinks)
   ("C-c n d" . denote-dired)
   ("C-c n g" . denote-grep))
  :config
  (setq denote-directory (expand-file-name "./notes/"))

  ;; Automatically rename Denote buffers when opening them so that
  ;; instead of their long file name they have, for example, a literal
  ;; "[D]" followed by the file's title.  Read the doc string of
  ;; `denote-rename-buffer-format' for how to modify this.
  (denote-rename-buffer-mode 1))

;; For the demo, I forced notes to show in full screen by default.
(setopt display-buffer-alist nil)
(add-to-list 'display-buffer-alist
         '("^[D] " ;; Denote buffer name prefix
           (display-buffer-reuse-mode-window display-buffer-full-frame)
           ))

Notes created during the demo

These are the notes Christian created during the recording. The recording was sped up 3x--4x, so this could be nicer to read than having to squint at the video:

20251121T204827--mechanics-and-habits-to-use-a-zettelkasten__zettelkasten.org

#+title:      Mechanics and habits to use a Zettelkasten
#+date:       [2025-11-21 Fri 20:48]
#+filetags:   :zettelkasten:
#+identifier: 20251121T204827

Essential:
- Write: put in effort; [[denote:20251121T205045][GIGO]]
- Connect: Link notes to create trails
- Correct: Improve as you go

Additional:
- Design for use: give things a purpose, e.g. blog about it
- Create structure. [[denote:20251121T205312][Types of structures in a Zettelkasten]]: design new entry points and trails
- Start in the Zettelkasten: use the system to learn, make it grow, get better
- Start with a link: keeps your notes connected. [[denote:20251121T210416][Create notes as link first to avoid orphans]]

Tool doesn't matter as much. Small files are good. [[denote:20251121T210107][Use Denote in Emacs]]

(Christian Tietze: Zettelkasten for Regular Emacs Hackers, EmacsConf 2025)

20251121T205045--gigo__programming_quality.org

#+title:      Garbage in, garbage out
#+date:       [2025-11-21 Fri 20:50]
#+filetags:   :programming:quality:
#+identifier: 20251121T205045

The principle known as "Garbage In, Garbage Out" (GIGO) in computer
programming states that for any system, the quality of output is
directly depending on the quality of input.

See: https://en.wikipedia.org/wiki/Garbase_in,_garbage_out

20251121T205312--types-of-structures-in-a-zettelkasten__structure_zettelkasten.org

#+title:      Types of structures in a Zettelkasten
#+date:       [2025-11-21 Fri 20:53]
#+filetags:   :structure:zettelkasten:
#+identifier: 20251121T205312

Non-comprehensive list of structures that could be used in a Zettelkasten:

- Opposition pair: 1 note for the pair, 1 note per position/opposition each; pro/contra
- Table of contents: list of topics/headings
  - an outline for a writing project
  - recreation of a book's contents (so I can write in detail about the book);
- Argument: recreation of a distilled form to get to a conclusion;
  list premises and evidence, and how these support the conclusion
  - Counter-argument: address parts of the original argument to support another
    view and test the argument's strength
- Table of things
- Graphics, like a concept map, Mind-Map, diagram: visually bring
  elements into a relation, then write about the relation and the elements

Metaphors:
- Iceberg: visibile tip with hidden depth; someting that appears small
  but is large; good metaphor for obstacles
- Black box: focus on inputs and outputs, and ignore the 'how' in the middle
  - e.g. most Emacs Lisp functions (hoping the documentation is good)
- [[denote:20251121T205739][Atom, molecule, organism]]: how small parts compose to larger parts,
  which compose to even larger pieces


(Christian Tietze: Zettelkasten for Regular Emacs Hackers, EmacsConf 2025)

20251121T205739--atom-molecule-organism-metaphor__composition_recursion.org

#+title:      Atom, molecule, organism metaphor
#+date:       [2025-11-21 Fri 20:57]
#+filetags:   :composition:recursion:
#+identifier: 20251121T205739

Atom: smallest part, indivisible elements
Molecule: comprised of atoms; structure added to combine elements
Organism: comprised of molecules; different level of analysis, irreducible to molecules/atoms

This can be used e.g. for programming: functions compose into larger
functions which compose into packages.

Actual biological organisms also work in such a way. Organs are part
of bodily functions (high abstraction), but from a chemical or
physical perspective, you can decompose them into atoms. That doesn't
tell you anything about the organ's function.

20251121T210107--denoteel-for-zettelkasten-in-emacs__denote_emacs_zettelkasten.org

#+title:      Denote.el for Zettelkasten in Emacs
#+date:       [2025-11-21 Fri 21:01]
#+filetags:   :denote:emacs:zettelkasten:
#+identifier: 20251121T210107

While not shipping with Emacs out of the box, Denote is easy to install and a great start to manage notes:

- it comes with unified way to create new notes, and hides time-stamped ID's with "[D]" in buffer lists
- it can manage links, and show incoming links to notes ("backlinks")
- it supports quickly adding links, and creating links _first_ out of
  the box -- it just creates the note immediately, so you don't notice
  - [[denote:20251121T210416][Create notes as link first to avoid orphans]]

* Example configuration

#+BEGIN_SRC elisp
(use-package denote
  :ensure t
  :hook (dired-mode . denote-dired-mode)
  :bind
  (("C-c n n" . denote)
   ("C-c n r" . denote-rename-file)
   ("C-c n l" . denote-link)
   ("C-c n c" . denote-link-after-creating)
   ("C-c n b" . denote-backlinks)
   ("C-c n d" . denote-dired)
   ("C-c n g" . denote-grep))
  :config
  (setq denote-directory (expand-file-name "./notes/"))

  ;; Automatically rename Denote buffers when opening them so that
  ;; instead of their long file name they have, for example, a literal
  ;; "[D]" followed by the file's title.  Read the doc string of
  ;; `denote-rename-buffer-format' for how to modify this.
  (denote-rename-buffer-mode 1))
#+END_SRC

20251121T210416--create-notes-as-link-first-to-avoid-orphans__linking_zettelkasten.org

#+title:      Create notes as link first to avoid orphans
#+date:       [2025-11-21 Fri 21:04]
#+filetags:   :linking:zettelkasten:
#+identifier: 20251121T210416

Recommended practice by Christian Tietze: start a new note with a link
first from some place, any place, then create the file to match that
link (which would initially lead nowhere).

- Reduces orphans in the Zettelkasten by guaranteeing every note to
  have at least one link.
- Supposedly teaches you to think about possible connections early,
  which makes the network better.

A very loosey-goosey approach, when taken literally and used
liberally: it doesn't matter where you are, just leave a forward link
to something you wish existed. Even if it doesn't fit the current
note. After all, you came up with the idea just now -- so maybe
there's a connection that you just can't spell out, yet?

Discussion / notes

  • Q: I wonder what they use for the fancy animations (what is your presenation software stack?)
  • A: Apple Keynote -- sorry for the proprietary product, I'm just fastest with that. PowerPoint also works, and when PowerPoint can do, LibreOffice can also do. It's more about sticking to a few simple transitions and timings, like fading and directional movement/wiping, and you're good. Read "slide:ology", that's amazing and full of great examples.
  • Q: Are you not a fan of using , , headings in org-mode?  (It's interesting to see how people have different styles of writing org content.)
    • A: Didn't see a need for hierarchical structure for the presentation. Also I'm leaning towards sequential writing, i.e. 'notes as blog post' pattern, or in our technical terms: "A Zettel should be like a StackOverflow post: a self-contained answer, specific to the question, with references to learn more if needed." An outline doesn't fit that well.
    • Fair enough.  Thanks.
  • Q: can you use org files and all its features inside denote?
    • A: Yes, this is an org file and Denote uses Org headings.
  • Q: Where/how do you like to capture fleeting notes?
    • A: Uses small paper notes. (Also don't get hung-up on terminology like this please :))
  • Q: Zettelkasten feels like a very "cagey" approach to note-taking and knowledge management. Doesn't it restrict one to think in certain ways rather than what feels natural to someone?
    • A: Take that feeling of being caged in as feedback to adjust the process. Otherwise, the rules I subject myself to are voluntary. I give my best during the writing stage, which is 'cagey' and taking more effort, too, for the benefit of my research in the long run.
    • Makes sense. Thanks! I'll give another shot to zettelkasten and rewatch your talk!
  • Q: How does denote compare to org-roam?
    • A: Denote is smaller, allows finding notes, but gets out of your way otherwise.
  • Q:I noticed that the wikipedia link you wrote was typed wrongly - and it got me thinking about how to deal with broken links at scale? Do you have any thoughts on this? What about archival?
    • A: Use any script to automate fixing broken links
  • Q:When I completely re-worked my config some two years ago, I also tried out some of these packages for making notes in Emacs. But none of them actually fit my purpose because I do not like the idea of splitting up my ideas. I am writing prose, I collect material, reading lists and so on in just one big file everything fits in. In this notebook.org file I do archiving from time to time in a monthly basis, and I search these files in Finder on macOS to find older notes I would like to go back to. So, if I write a paper, it's basically an extract from my old notes that only have one structure, viz. when I took them. Time structures my thought, and my notes. But they would get lost if I split them up in a network-like structure like a wiki, or even a zettelkasten. My 2 ct. BTW, I work on legal and sociological topics. I'm a lawyer by training.
    • A:Cool! Happy for you :) See Twyla Tharpe.
  • Q:How Zettelkasten is useful for highly mathematical STEM academic fields like computer science or engineering fields? Like when studying a STEM field, what should we make notes of? The important bits are interconnected in a heirarchy, usually chapters of a book (1 -> 2 -> 3). Main points are usually dense, highly context-dependent sentences. One feels like he is copying information to another box. It's usually not possible to reword scietific results in ones own words.

    • A: (probably by oliver epper!! )Zettelkasten is great for Mathematics and incredible for CS. With org-babel you can have living source code that you can execute from the note. (Zettelkasten using denote, using org is the link here). You can even use Agda if you want to have living proofs in your note. Even inline LaTeX is an option for beautiful commutative diagrams or math terms. Quote the sentence as is and add ...?
    • A: foundational knowledge is important for tests mostly; and eventually becomes your inventory of skills (you can do X proofs from the top of your head).   Copy box: fridge me
      • see appendix A below: local link in supply chain of knowledge
      • engagement doing the work of not copying, but rephrasing, organizing, but sometimes also retyping alone
  • Q: In your experience, would you say that you re-use most of your notes? Watching your demo, I though that more notes your create, less you might re-use them... Thanks for sharing!

    • A: No, that'd be impossible. I have 11k notes. "Most" would be 50%, 5500 of them. I can't use that many things most of the time, it's impossible. Focuses shift, and I go from one department to another, metaphorically, as the hours, days, weeks pass.
  • 2:09 PM EST Q: How are notes structured and accessed when the notes grow from 10K to 100K notes.

    • A: Like a fractal; self-similar structures
  • 2:13 PM EST Q: Not a Question but I would be very interested in your thought on this video by Westenberg :- https://www.youtube.com/watch?v=CjSWwmg-JRM , Why I Deleted My Second Brain: A Journey Back to Real Thinking

    • A: dunno, probably a lot of collection, not a lot of thinking in the ZK to some purpose that worked...? Usually, posts/videos like this conflate note taking as a dumping ground for 'stuff' with something that is valuable in the long term. 
  • Q: Is there a danger that with a Zettlekasten-process, that the process gets a bit in the way of the content?

    • A: STrive for speed of thought tools, not e.g. typewriter. THEN: reduce ritual and ceremony. 
  • Q: How do you navigate looking at all posts with certain tags (this might be in emacs already; new to emacs ;))

    • A:I don't; tags are just a way to loosely group notes (and stuff in general) so that you can find potentially relevant pieces of information more quickly. To navigate, use links. Create nativigational aids. 
  • Q: I thought denote uses md files
    • I think it can use either markdown or org.
    • denote is agnostic to the format
    • and I already use org-roam but this makes me think I may want to switch.
    • denote is more minimal
    • Denote is simpler
    • Org-roam feels more hackable for me
    • org-roam tries to implement Roam Research
      • I'm just used to org-roam since years
      • after some research I went with org-roam just beacuse it handles org files. boy if I knew that back then
      • I have a lot of stuff that depends on org-roam, because it does more than just the zettelkasten. It is more of a database structure.
  • His slides are very aesthetic.
  • all these speakers need to provide there emacs ui setup in emacsconf repo
    • it was a compliment, all of them have eye-candy emacs
  • How did I not know the guy who runs zettelkasten.de was an Emacs user.  (It makes total sense, but I didn't know until now.)
    • oh wow, is that a video editing thing or an Emacs thing?
    • the video-editing is really great
  • So meta! Loved seeing how they  turned a zettelkasten into a talk.
  • Very, very interesting talk!
  • I'm going to have to rewatch this one a couple times. There's so much good stuff here. I don't know if the Zettelkasten idea has every fully "clicked" for me, but I think this is the closest I've been to "getting it" 
  • (add-hook 'org-mode-hook #'visual-line-mode)
  • oh yeah I have started to download the webpages I refer to
    • I use wget manually though.
  • I have tried zettelkasten through org-roam before. it felt too overwhelming, so I just do "bunch of org files in a directory" now
    • I think everyone has to find their own style.
    • org-roam can feel confusing in the beginning, I do get that
    • I feel like it is okay to just do what you feel suits you and slowly improve your flow
    • I am very addicted to it at this point, but I understand this
    • it doesn't have to use the hottest thing in the market
    • I don't do second brain thing now, I just write whenever I want to
    • I use org capture to maintain a journal and separate org files later.
    • I always start with writing and link things after I have a first draft of a file
    • I can confirm that 2k notes need to be very well organized to not get lost. Once you get to the point where your notes grow too fast, you really need a very good structure
  • it is an org file?
    • it looks like an org file. He just doesn't use headings that much.
  • I'm going to have to rewatch this one a couple times. There's so much good stuff here. I don't know if the Zettelkasten idea has every fully "clicked" for me, but I think this is the closest I've been to "getting it" 🙂
  • Excellent talk (slides and presentation itself). 👏
  • agreed, it's going to require several listens!
  • easily one of the best talks during emacsconf
  • Thanks for the talk!
  • lol, 'so is wearing underwear'
    • Amazing parallelisms
  • "so is brushing your teeth everyday"
  • Q: does christian make videos too
    • I have seen them somewhere
  • the point of zettelkasten is to see varying differences between various notes and make interesting connections. so the only 'cage' is to just write notes, the creativity will happen when you see the interesting connections.
    • as luhmann put it: "without noticing differences, one cannot think"
  • so far this talk has been very good
  • off topic, but I really dig there overall room aesthetics
  • https://github.com/yibie/org-supertag
    • it's by the emacs-china guy
    • though I don't like the ai integration in there

Appendix A:

202402121625 Zettelkasten can be the fridge-like buffer of knowledge

zettelkasten 

Any local knowledge database serves as a buffer between all available information sources on the internet and actual problems that have been solved.

[T]here's no "local link" in this supply chain that we've developed. Google is the closest link, meaning we use the massive, unsorted network that is the Internet and Google's interpretation of our problem as our "second brain".#20240212ds

A Zettelkasten, as a trusted system201304061037 for information retrieval, becomes the first line of defense when looking for a solution:

    Problem → Zettelkasten → Google

As Aseem Thakar points out, we don't go foraging in the woods whenever we get hungry, but instead go to the fridge in our apartment, then head to the supermarket if we don't find anything there, then maybe check out farms if the market is empty.#20240212ds Defaulting to web search for a solution and never building up a local storage buffer of answers to our problems would be similar.

software developer", 2021-09-26, <https://aseemthakar.com/how-to-build-a-second-brain-as-a-software-developer/>

Transcript

[00:00:01.400] Introduction
Hello there, fellow basement dwellers. I'm Christian and you are watching " Zettelkasten for Regular Emacs Hackers," my Emacs Conference 2025 talk submission. In this presentation, I'll be showing you a couple of things about Zettelkasten, very basic mechanics and habits that you can pick up and implement in Emacs, the most malleable of all environments, to make a thinking environment happen in your life that stays with you potentially forever. The subtitle "For Regular Emacs Hackers" implies at least the possibility of irregular Emacs hackers and regular Emacs non-hackers, so the target audience here is people who are comfortable tweaking their setup when they run into issues and use Emacs to write, no matter if it's prose or code. So that's regular Emacs hacking. You don't need to be an irregular Emacs hacker, for example, being a core maintainer or whatever. You just need to be a normal user who also modifies the setup. However, you should probably not be an Emacs non-hacker. Or maybe you shouldn't stay an Emacs non-hacker, someone who is not tweaking their setup ever. So, if you just open your application to write with a double click, and it happens to be Emacs, this may not be for you, but you ultimately be the judge there.
[00:01:25.534] Advocating Freedoms
It is my sacred duty to, of course, advocate all the essential freedoms during this presentation. These are the following. You shall not be shackled by a proprietary tool. (You shall be shackled by Emacs. (Which is free software.)) You shall also not be shackled by an esoteric method that turns out to be a grift where you need to visit annual workshops, walk on broken glass and stuff to be a true ""knower"". You shall be empowered to do great things for the rest of your life after this session alone. This is not a sales pitch. Finally, you should also not be shackled by whichever sources of information you rely on in the future. You shall be free to think and explore new ideas, ideally forever, in an environment you built to your liking, without the degrading web searches and the dead internet getting in your way. No libraries, no dead trees. It's you and your knowledge base that can truly produce magnificent things.
[00:02:29.680] What Is This About?
So what is this about? In the teaser text for this session, I brought up that when people talk about Emacs pulling in everything that people do on their computer, it's usually things they used other software for in the past. Like email, chat, playing music, browsing the web, managing tasks, you know, stuff like that. We don't have a good blueprint for thinking environments though. So it's not a trivial task, just port this or that to Emacs and then you'll be happy and productive. That's different from doing your emails or task management or writing in Emacs, where we have a lot of experience with existing software to adapt and deviate from, where we can essentially port the task to Emacs. We can practice to think, purposefully and productively, on complex things over long periods of time when we create bespoke environments that help with that. The first assumption is this: Writing is very important to form complex thought. Without writing, you won't be able to cross a certain complexity threshold. Thinking in your head alone without any externalization makes you prone to loops, repetitions, and worst of all: jumps. Jumps that get you to a point, but not backed by reason or argument. So you and your future and others cannot follow. Written words direct thought. The linearization or sequence-making of thoughts frames your next idea. That's the same for reading, which enables you to pick up existing ideas and continue to write about them later. So for "good thinking", writing, reading, and writing is mandatory. Emacs is good at showing text to read. It's good at processing keyboard inputs to write. So let's go. Let's set up an environment within Emacs to make this thinking thing happen. I'll walk you through some mechanics of the Zettelkasten machine and habits for you, dear thinker and regular Emacs hacker. And for your reference, highlight these things as I present them in the bottom left corner of the screen. So working with Zettelkasten as a thinking environment only requires very simple mechanics.
[00:04:36.534] Write - Essential Mechanic
First one is to write. I mentioned this. It could be one large text file, could be many small ones. We use the power of Emacs and small files because Emacs is cool, and individual files put boundaries around ideas that force you to decide what goes where. Most importantly though, write like you mean it. The principle of "garbage in and garbage out" holds. You don't need to use your novelist voice when taking notes, but it also shouldn't be shorthand only, so that your future you has an easy time reading and digesting what you wrote.
[00:05:09.601] Connect - Essential Mechanic
Next mechanic, which is also essential, is to connect. We think in associations. Connect notes to capture the associations that come to mind and that you want the reader, which is the future you, to make. Traveling a path of connections indirectly via tags or keywords and directly with links can feel like reading an essay you make up as you go. That's where connections show their power.
[00:05:34.268] Correct - Essential Habit
The next essential habit is to read and correct and reconnect as you go. You spend time and effort at the writing stage, you are the primary audience, so do your past self a favor and read what you wrote. And then as you read it, make it better. Make it better, continuously make things better and capture new ideas as they come up as you read. And things you can only now remember because you learned about things in the meantime. You'll also get better at this whole thing with practice. So improve old notes when you find them lacking in detail, their tone pretentious, their mere existence an insult to your intelligence. Pay [knocks on table] attention [knocks again] to pain points in using notes. Yes, I knocked on my desk to emphasize. And fix things on the fly. From this principle follows a lot of common practices and tips. This principle truly is essential. Too long and you didn't read it? Add a summary at the beginning. You can't understand what you wrote a year ago? Do your best to rewrite it in your own words. It only gets worse if you wait longer. Can't find anything in the mess? Collect links to what you could find in a new "meta" note so next time, you have a navigational help. This practice will form the basis for structure notes, maps, and overviews, which we will come to later.
[00:06:49.434] Design for Use - Habit
The next habit, non-essential though, is to create notes with intent to use them. It's one thing to write about facts, capture information, but all this is just collecting stuff. It's another thing altogether to write about a train of thought, about an argument you found compelling, about a model to understand the world, or yourself, in a better way. So collect to remember, but work in your Zettelkasten to think. What does that even mean, though? Recreate how intriguing books lay out their premises and arguments, for example. First this, then that, also that supports the premise, and so on. That's the structure of an argument. You can recreate it in list form, as a graph, you can draw and import the image, whatever. The structure of that argument is one thing, and the details, like the evidence for each claim, can be separate things. These can become their own sub-networks over time.
[00:07:43.920] Create Structure - Mechanic
To facilitate all that, you will be needing to create structures. You've connected notes, so links already leave trails to traverse between your notes. Structures can emerge from these with a sheer volume, but they also can be designed by you to be navigational hubs of similar shape and form over time. Patterns like that reduce confusion and improve feeling at home and finding your way around, so that's worth investing in. For example, use outlines for complex topics. Tables of contents of a book, for example, that you love and processed in great detail. Just recreate the table of contents, the nested structure of it, in your notes, and then you have something to hang your future thoughts onto. Another example is a pro/contra table or list to discuss opposing facets and perspective of a thing. Another example would be models or metaphors like the iceberg model where you point out something has a hidden depth to it or the metaphor of a tree to model a thing as a branching and growing idea.
[00:08:47.968] Start in the Zettelkasten - Mechanic
Another habit which is also not essential is to start in your Zettelkasten. Starting in your Zettelkasten removes the cost of deciding what goes in there and what doesn't. It gets you moving and up to speed with the method and the tool much easier. Importing stuff later into the Zettelkasten can feel like a chore, but starting the work you need to do anyway in it? That reduces the mental hurdle. As a regular Emacs hacker, you'll eventually develop your own tools to make initial exploration smoother over time, like do you start in a particular place or just create a new note from scratch somewhere. You won't know this until you experience this stuff for a while and try different things. So don't worry and be open for change.
[00:09:32.401] Start with a Link - Mechanic
The final habit, also non-essential, is to start with a link and not with the creation of a new file. Start with a link, create the file later. This avoids orphaned notes. Orphaned notes are those no others are linking to. To these you can only get with a full-text search or maybe by accident when you browse your notes, but there is no orderly way to get to them. To recap: Write. Don't be sloppy. Put in effort early to get faster at this. This is essential, because without putting effort into writing, you won't have anything to use. Connect and leave trails to navigate. That can tell a story when you traverse the trail later. This is essential because without connection, you will not get anywhere. Correct and improve things as you go. The last essential thing: well, don't worry about perfection, and then, be gentle to your past self. Adapt to what you learn along the way. It's essential, because without this attitude, you can easily get stuck in analysis paralysis, like where do I need to put this, or what would be the perfect way to phrase this. Design for use. This helps both finding your voice, and to have criteria for selecting what to spend time and effort on in the first place. It takes into account the opportunity cost of high quality work from writing and connecting. Create structures. You won't be able to scale and stay organized and find your way around without structures. You can practice this early and design structures deliberately, but it's also okay to ignore this for a while and wing it. So it's not marked essential, although it may hurt you sooner than later. The habit to start in the Zettelkasten? Well, do the work you need to do in a place that can pay back dividends on the effort you put in. That's powerful, but also not essential. You could just as well continue to write and think and scribble somewhere else, and then do the Zettelkasten importing stuff later. Start with a link. That's really useful practice, but more like a lifehack and not an essential habit. So you can also create new files from scratch for ideas that come up as they come up and then try to connect them later. Well, that's better than not writing at all, right? So if you look at this, you may ask yourself, why is this create structure thing a mechanic and not a habit? What's the difference? It seems kind of random. Well, do create structures as an imperative is a good habit, yes. Structures facilitate growth of the Zettelkasten and help you discover useful patterns and the things you care about. Patterns that work for you personally, which make navigation easier because they fit your personal expectations for what is and what is not. That's something for you to do. That's a process. But from the perspective of the Zettelkasten as a system, that's a mechanic or rather dynamic, the Zettelkasten grows organically. Thanks to your constant intervention and usage of course. That's how time passes in your Zettelkasten. That's how a process of transformation enters the system. The transformation affects the network. Every new or updated note, every new connection changes the network. The existing network then imposes demands for new stuff to fit in, slowly solidifying how things are organized to be perceived as orderly. This is calcifying. That's emergent creation of structure from use. Structure here is expectation for what could come next. On top of this interplay of emergent structure in your notes and processes that operate these constraints, you can design and influence and architect and have explicit structures and patterns, and therefore you can influence what is expected, what is unexpected and what fits and what needs to change to fit in. So the time you spend designing these things will influence how the Zettelkasten will behave in the future.
[00:13:22.034] Facilitate Growth
To prepare for growth you will probably encounter thresholds along your journey. Like 1-10 notes, well, you can easily remember all of them. 10-100, you will have forgotten some details, but will probably remember writing most of these notes in some way. 100-1000? Bad luck, you will have a hard time going through everything one by one to find what you have. You will have to rely on filtering results. For example, with a full text search, you will crave to use tags and keywords more to group notes into more manageable departments or collections. By this mark, search results produce way too many results. Popular tags become overcrowded, and you have the same problem you had in the last stage, but for each of these tags. So manual structures will take you through this. Anticipate growth pains by starting from structures. That's the recommendation here. Design your entry points into your current projects and research topics and interests as 'departments' of your Zettelkasten. Keep a list of, for example, 12 darlings, like Feynman did: a list of 12 things that you can check mechanically where you capture something new, and then you can see whether the newfound knowledge can also push one of your darling projects forward.
[00:14:46.140] Emacs demo
Now, finally, let's get to the demonstration in Emacs. Here is a very minimal init file. I will share it with you in the show notes. And this is the Denote default configuration. Here I'm using the shortcut to create a new note immediately for this talk. And there you see. That's an empty new note. Here, sped up like two or three times the normal typing speed of me, is how I would process this very Emacs conference talk. The essential mechanics and habits, additional habits, mechanics, and then from there after I capture everything. Make sure that I have a reference. This is not a thought-out implementation in Emacs, so this is just plain text. Christian Tietze, Zettelkasten for Regular Emacs Hackers at the bottom. You can use reference management systems that you like, but I don't want to get into these details. Here I'm creating a note with the denote shortcut. Based on the selected text, I'm starting a link. This link is creating the note for me. It's also default Denote functionality and garbage in garbage out. I needed to edit the title because the selected text became the note title. Didn't want that. That was the abbreviation. Notice that the default configuration does not in fact include auto-fill-mode, so the lines get infinitely long. Looks a bit weird. Just garbage in, garbage out. Processing this from Wikipedia. So we have a detail note from this overview. So that's an overview with one link already. Starting from here, now I want to write more about my talk. And next we create structure, types of structures, etc. It creates a weird link, but I can edit this easily thanks to Emacs being so nice to work with. A couple of examples. I mentioned some of these in previous minutes of this conference talk, like position pair, one note for the pair, one note per pro and contra, table of contents, like lists of things you like, to talk about recreating a book's content, table of contents so you can process the book in detail, argument structures, I believe I mentioned these. Look at this up if you're not into arguments, but arguments are very well structured, usually. A table of things like two-dimensional table or grid. Graphics. You can also include graphics, images, and then write about these. And then there are metaphors. And into one metaphor that I'm presenting here, iceberg, black box, and then atom, molecule, and organism, I want to get into this. Atom, molecule, organism. That's a composition and recursion because I have Big Ideas there. Atom, smallest part; molecule, comprised of atoms; and organism is comprised of molecules. Different level of analysis. Because this is irreducible. In fact, if you have no clue about reducibility, irreducibility -- that doesn't mean much to you? -- but look this up. You can go very deep with this kind of stuff. It's basically that if you decompose organs into atoms, you cannot get back to the organs. You just have a bunch of atoms. There's information loss, more or less. Here you see that I create a new thing at the end so that I can write about Denote. The tool doesn't matter, but when you use Emacs, use Denote because, well, why? Let's get into this. Fix the link. These are good reasons to use Denote. Denote is very simple. Denote has a couple of sane defaults. That makes life easier. Backlinks. We will see a backlink view at the end. I have to create a couple of things. I'm copying the source code there, the Elisp source, so that you can see, hey, this is just an Org Mode file. You can style it to your liking and then you can even execute the code if you want. Very powerful. Create notes as links first to avoid orphans. Forward link again. At least I wanted to create a forward link. I pressed the wrong shortcut. But anyway, I can fix this easily. You see, there's no link. Dammit. Now I need to create the link after the fact. Here's a list of shortcuts. The denote keymap. It's a recommended practice by me, starting your note with a link. You've heard this all just a couple of minutes ago. It reduces orphans and supposedly teaches you about thinking in connections early. It's a good practice to practice. So with that note, trying to switch back. Denote note switching, that wasn't as smooth, but inserting links is. And there you go. Here's a backlink view. And that's it. In a somewhat self-documenting way, here you see a structure note which is an overview that represents the gist of this Emacs conference talk, with a couple of links to details. From these details, as you've seen, you can go into even more detail. That's all there is to it. Repeat this for infinity, and you get really really complex networks and can do a lot of amazing things in parallel without interference.
[00:20:39.068] Learn, Share, Grow
I just want to stress that the Zettelkasten can help you to learn when you publish, when you share, and when you grow it and yourself in the process. Again, design the Zettelkasten to be used. Publish something, write a blog, share stuff with co-workers. That's powerful and that's so rewarding. This can in turn influence how you do it the next time in your Zettelkasten, because now you can anticipate these kinds of arguments, maybe I can do this early on, and then you're prepared even more for the future to share what you learn. You are also invited very warmly to our community of practice in the Zettelkasten forums. Just share your journey, write about your projects, ask questions. Everyone's welcome, newbie to pro. Just get in touch with people, talk about the processes, improve them, and eventually you'll figure out, well, reaching enlightenment in that regard may not be that hard after all, and then you're fine and good to go for the next projects that you tackle. Most importantly is to make this thing your own. The Zettelkasten, the method, the environment. Create a thinking environment for you. Create your own tools to think with. This goes back to the meme of Shuhari, which is basically imitate and then deviate and innovate. And this invitation here is to imitate what I just laid out. Imitate for a couple of years. One, two, three years. The time goes by faster [snaps fingers] than you think. And then figure out ways to deviate from the doctrine, to figure out ways to improve and change the processes to fit you better. But you need to try to manifest the best practices in your life first, for a while, to then figure out, well, they are not that best after all and I need to change some of them. But you wouldn't know if you didn't try. So do try. Yeah, and with that I want to thank you. Thank you so much for watching. That's it. That was the conference talk, my short introduction to the Zettelkasten mechanics and habits. I want to thank you so much for watching and spending time with me on this topic, on these two topics actually, that are very near and dear to my heart. Do share questions, ask questions in the etherpad. And if you watch this after the conference and all the live participation is long over, step into the forums and ask around there. Thanks also to Sacha and team for organizing EmacsConf 2025, for having me. Well, I'm looking forward to hearing from every one of you. So that's it. Peace out and see you in the next one.

Captioner: sachac

Q&A transcript (unedited)

Yes. All right. Take it away. Thank you, Christian. Thank you, too. If you could have the pad open at the same time, you can read the questions. Or I can start reading some to you while I'm here. Oh, no. I can read them. I was wondering whether I should maybe copy them into a new buffer. So they are also on screen. Increase the font size a bit. So I'm trying to do this on the fly. Maybe a bad idea. Let's see. Of course, you're going to capture it into your Zettelkasten then. Is that what's happening? Well, I can start you off. The first question is, I wonder what they use for the fancy animations. I was thinking about this because that's the first question and I was thinking about this while I copied this stuff over. What part is the fancy part? I can recommend books, like if anyone wants to have book recommendations for how to make presentations with PowerPoint-like software in a very simple way, we taught this at university. was it now, some 15 years ago, to students to make animation abuse where everything was flashy and typed in or something. Don't do this, but instead do a couple of very simple tasteful things like fading through colors like filmmakers do, right? Fading through black to make a scene cut or just fade between things, the fanciness. Presentation software stack, the fanciness. I didn't dial fanciness up, I just resorted to a very simple fade animations, like fading stuff in and wipe, I think, for text effects. That was it, more or less. And for the lines, maybe the lines are fancy. I was using Apple Keynote because I'm fastest with that, but I also usually, we taught this workshop with PowerPoint and I think the LibreOffice stack got much better with that as well in the recent years, but I haven't tried that in a long time to like fiddle around and find all the knobs to dial. Because the Apple presentation thingy has this nice feature where you draw an arrow with a tip, and then you have a special animation for line drawing, which is only available for line art. And then it draws the arrow that moves around like that. So yeah, presentation stack, Apple Keynote, probably not of interest for anyone here. I usually don't use plain text presentation stuff, right? So I tried this, I tried this with markdown presentations, slidey things, org presentation. It's always not enough control for the fiddly things that I'm interested in to make the experience great. So I wonder what are the fancy animations. And the stack, the stack is Apple Keynote, sorry. Okay, that's enough. Thank you for capturing. Are you not a fan? Okay. You saw this in the recording. Why on earth is autofill mode not enabled? I don't know. I thought it is the default, but apparently it isn't. I could also use visual line mode. It's built in as well, right? Visual line mode. It's so weird to be in this vanilla setup and not have all my key bindings and my normal stuff ready, so. I am not a fan of using asterisk headings in org mode. It's interesting to see how people have different styles of writing org content. Am I not? What exactly? What did I do? And nodes, just open one of these. Okay. I could see how you could, in examples like this, use org-mode to read-only. Why? How you could use org-mode to use headings for this, because it's an outliner first and foremost, so outlining is very natural. But when I do in my personal setup, I also have this what's called start indentation thingy. enabled, so that means that everything would be indented one level from the start. So I would probably fiddle around with that to not get like crazy. But also, I don't see, I don't see, like really see when I look at this, I didn't see a potential to create subheadings, I mean, or even headings. The only heading here is the title, like in my perception of this node, and it's one, It's two lists. You could group these in headings. I didn't. That's right. Where's another one? I opened the autosave file because I'm stupid. OK. So this one could also be probably subdivided. But I'm also not sure. I'm also not sure if I would gain anything structurally if I do it like this because then I'm in a structural level and the stuff that just flows naturally as a sequence of text paragraphs, this here, it's not disconnected. It wasn't meant to be under organism and I need to create a new same level thing to say, hey, these are, I don't know, details or whatever. So that's maybe the real reason. I'm fine with writing snippets that are self-contained and essentially one or two or three or five or whatever, how many paragraphs, but it's not like, it's always an outline form. So I don't resort to this. My onSettle custom is usually written in Markdown. So I use subheadings for actual headings to subdivide like I would subdivide an essay or an article, blog post. But that's the mode of writing that I'm in here. It's like blog post writing. I wouldn't start with four headings for this. So I'm not doing that. That's a thing. Okay, I guess I'm going to read the questions, but I'm going to stop copying them in because I lose time. I want to reply to you, folks. Didn't see a need for a rack constructor. Thank you for capturing this. Next question.
[00:07:13.720] Can you use org files and all its features inside Denote?
Can you use org files and all its features inside Denote? Yes. Like this is, this is an org file and Denote uses org headings instead of, with Markdown, you get YAML front meta. And with org, you get these attributes, value attribute thingies that are then the metadata for the node. So yeah, the answer is yes. You get everything and on top, a small layer of a link, link management. That's all there is.
[00:07:49.000] Where or how do you like to capture fleeting notes?
Next question is, where or how do you like to capture fleeting notes? Oh, fleeting notes. If you say fleeting notes, you also need to say non-fleeting notes and figure out what these are. And when you say non-fleeting notes, and fleeting notes exist, because it's a distinction, and there's something on both sides of this distinction. If you say there's a distinction, there are two sides, maybe you want to subdivide the non-fleeting notes further, because, well, it's a non-fleeting note isn't very descriptive. So what else do you say there? And people have said a couple of things to define non-fleeting nodes, for example, permanent nodes. I guess that's a Sonke Ahrens, permanent nodes is the most popular. Let's stay with that, stick with that. So if you have permanent nodes and fleeting nodes, now we have two types of nodes. The thing is, in books, the only, when we would talk about note-taking and you would ask me, hey Christian, how do you take fleeting notes? I imagine the discussion would be in German because that's how people usually talk to me. I would pull out a book, this is Object-Oriented Software Engineering, interesting book by Iva Jacobson or Iva, Iva, I'm not sure because, you know, it's an English or American person. And inside I have these fleeting notes like these are actual paper notes square what are these three by something inch American standard size I guess and A6 minus the tariff part from from another notepad these are notes i took engagement notes if you will like engagement notes because the margins of the book don't suffice to take and this is related to i need more space so this is more space these are fleeting and as you see I have them in my pile of books right next to me in the shelf and folded them in this piece of paper, labeled it with the author because I lost them. couple of times they just fell out when I reached for the book and this is an envelope I shove it into the book and then forget about processing the book again for five or ten years. So what do I do with fleeting notes? I do fleeting notes on paper or if I'm using an e-reader maybe with an e-reader software annotation tool but fleeting in the most like simple non contrived sense, fleeting notes are meant as engagement notes and then you do something with them or like I just showed you, you don't because life's short, right? Time runs out, then you need to forget, then you need to remember how to restructure everything in your head to make sense of the notes again because they were fleeting, they were just little scribbles and it's It's basically ballast or waste. I'm just carrying this with me for years and maybe never get around to using these. But also, I put stuff on there, maybe I do. It doesn't hurt to keep them in there. It's just an odd collector's habit I can't get rid of. fleeting notes, I just capture them wherever and then either I process them or try to throw them away or forget about them because they're in some inbox file on a smartphone and then, you know, they might as well not exist, they're just there to engage me during the reading and if I don't process them in time, meh like, chance is up, it sucks Next question, like fleeting notes, permanent notes. I'm not going to talk about these unless someone asks a question. Zettelkasten feels like a very cagey approach to note-taking and knowledge management. Doesn't it restrict one to think in certain ways rather than what feels natural to someone? Well, yes, but so does wearing underwear, right? A bit of personal revelation. When I'm working from home, I'm wearing pants. You don't see them, but I don't need to. But I'm also kind of self-restricting myself in a way because there's a window, it's dark outside, and I'm well lit. And if I just stand up and have no pants on, I don't care that much what the neighbors like 10, 20 meters across think. your mileage may vary then, right? So there are restrictions that make sense to get to some point in your day-to-day life. And in the case of Zettelkasten, I try to tell people the shortest story possible to convey a story there, but also not to cage them in with 20 definitions of different notes. Because all that matters is try to move your thinking into an environment where you can write, where you feel comfortable writing, and where you can keep this stuff. Because I do feel comfortable with a pencil and paper. But what I just showed you, this might as well never have happened. Something has changed in my brain. most likely, when I engaged with a book like this. As far as publications and writing is concerned, this might as well not exist in my life, because I didn't do something with it. I mean, I was reading the book probably on a commute back then, on the train. I didn't take notes on the computer right away. So if you feel caged in, that may be a sign that you went too far on the, I need to do this, I need to do that stage, and lose some of the playfulness. There needs to be playfulness inside of the whole procedure, otherwise it... If you don't enjoy doing it, you won't be doing it. And if you don't do it, then you don't get a benefit. And it's just another chore in your life. Then rather not do it. So try to avoid the feeling of caginess and do whatever you want with it as long as you try to give your best. That's the cage that you will benefit from like really trying maybe not trying hard but trying in earnest to produce something that you can read in a year in a week whatever and then make sense of it like that's that's that's the minimum bar and the rest If you find yourself writing for hours on end, during the nights, during the weekends, whatever, and can't stop, then you can tackle more of the prescriptions, let's say. But everything I showed to you was there are three essential things, and the rest is just add-ons. Add-ons to inspire you to think in certain ways, but you don't have to do them. You can scratch all of this. You just need to write to think, and then you need to connect to make the thinking happen in an organic way. Otherwise, you get a storage that's very hard to navigate and doesn't grow well. Thanks. Yeah, you're welcome. I'm not writing another show. Yeah. Oh, combine the forums. Like it's a Zettelkasten anonymous group there. Like everyone's struggling. Everyone's struggling and it's okay to like, I'm feeling like stuck at this and that. And then other people will be able to relate. I can promise. How does it denote, question mark maybe, compared to org-roam? Roam research was setting a new trend of connectiveness because you could create recursive structures with links and everything is linkable. And it's like you put every sentence in its own org heading and then link to all the headings possibly and then transpile them so you can expand in place where you would link to. I think this was the approach at least. So this was a very interesting transclusion, not transpile. Transclude the contents, like move them in right then and there and not navigate to another page. It was interesting. It never made sense for me when people told their stories of how they used it. I guess you can use this tool for a lot of purposes, but the prevailing story of the Rome style note-taking tools was like, just dump everything in it. And garbage in, garbage out still holds. I do keep my personal journal away from my Zettelkasten nowadays because I want to not restrict myself when I feel the urge to journal, when I want to capture something that, for example, my 15-month-old baby daughter does. Like, first time she pulled a chair out of another room to the next room just to create a ladder to then climb onto another thing. It just was, okay, this is now the time to take note of this. in her life, like you were this and that many months old when you did this. So we have some reference parts. And I could put this kind of factual information, et cetera, custom, but also, like, what if I have a toothache or whatever, and just want to gather data for whatever reason? I don't know. Like this kind of personal stuff. That's too much of the non-useful stuff in a tool that I want to use to think. I do see the appeal there. But anyway, the org-roam, like that's a roam part. You can watch a lot of videos on that to get a feel for this. Org-roam adopted this to make the connection easier and replicate features. As I said, porting tools that exist into Emacs, very nice pastime. A lot of people do this for a lot of things, but it's not a thinking environment. It's just another tool to take notes in that behave weirdly with links or interestingly with links. And Denote does do even less. It only offers you a couple of shortcuts to help you find files quickly. It does, like you see in the bottom here in the mode line, it does just show square brackets D instead of the whole ID, which would be rather long. So you have a buffer list that is bearable and works. You can read this and find the title quickly. Things like that, Denote does things like that. Oh, I closed the wrong. Denote does things like that very well, but it gets out of your way otherwise. Just a couple of conventions to get stuff into your file system. I'm not sure if Org-Roam now, what's the current state there? Do they do ID generation out of the box? Is there a standard popular convention or whatever? not sure, didn't follow, but I'm very happy that org-roam still exists because I think porting tools into emacs is very cool, I guess next question, I noticed that the wikipedia link you wrote was typed wrongly, oh no garbage in, garbage out, I think garbage in, garbage out, garbage well spotted how do you do that oh I typed it I typed it from yeah right i didn't i didn't paste this right yeah yeah this janitorial tasks that's the umbrella term how do you deal with this like if you put stuff like this into into your notes you hopefully get a 404 code from wikipedia because the page doesn't exist let's not check um you get a 404 and then you could write a tool that checks all the links periodically and tells you about broken links like a broken link checker that then looks up maybe automatically the last good version on archive.org and then the Wayback Machine and then corrects the link with the archived version because the live one is gone like this happens all the time on the internet and even if you don't mistype you could you know things go out of out of order so What do you do with that? You need to automate this. I'm not clicking every link manually. I have like, what's it now, 11,000 notes or something. This is not a thing that humans are good at. I'm not good at this. I will never finish anything else if I do this manually. So write a tool, write a script, get a script from the interwebs. There are tools like this plenty that can do it. And then you just need to do the wiring. and that should help. You could automate this in Emacs, of course, right? For every file in my node directory, look if there's a link inside and then check the URL or whatever. Also works for inter-node connections. This could also break if you rename files or remove files and don't use a denote function, which I believe takes care of backlinks and forward links and keeps them organized. If you do this manually on another device, out of denote, out of Emacs, in Vim, I don't know, then yeah, mistakes happen and you need to deal with mistakes and heal and create scar tissue there in some way so write janitorial tools to do that it's also, i'm mentioning the term janitorial tools a couple of times now because that's the term that we in the forums at least usually talk about things like find orphaned notes, notes that no one links to find broken links, clean up references reformat nodes, stuff like this. Find large nodes and then suggest them to the user to break them up. Like these are tasks that you can do mechanically, but also it's better if you use a tool to get started, otherwise you waste a lot of time. Next question, when I complete, next long question. I'm going to copy this over because maybe When I completely reworked my configs from two years ago, I also tried some of these packages for making nodes, but none of them actually fit my purpose, because I do not like the idea of splitting up my ideas. I'm writing prose. I collect material, reading lists, and so on in just one big file. Everything fits in. In this notebook file, I do archiving from time to time. I search these files in Finder to find old notes I would like to go back to. So if I write a paper, it's basically an extract from my old notes that only have one structure. when I took them time structures my thought and my notes but they would get lost if I split them up in a network like structure like a wiki etc custom my two cents by the way I work on legal sociological topics i'm a lawyer by training you know Luhmann was a german german style weird sociologist and sociology studies are like the first four semesters at university, where a lot of, I don't know any of the terms in English, a lot of these things were theories in the loosest and broadest sense. Some were just models of how things might behave, but nobody knew because they were not empirically backed. Some were empirically backed, and then also there was some sense making. And then also in sociology, you have some structure analysis of, like demographics and things like that. So very, very hard fact and statistics heavy things. So broad topic. And there's a lot of interconnections. I can see the approach here. I'm not going to defend Zettelkasten so that you in the end become a Zettelkasten aficionado or whatever I mean, if it works, fine, more power to you like reminds me of Twyla Tharp's like the shoebox approach or whatever it was called where she just dumped everything inside for her next project like an idea box, like real physical things like oh, I want to, I don't know, like this weird 8-bit do controller inside and then eventually she would spread out everything she collected, newspaper clippings and I don't know, maybe toenails or some weird stuff, I don't know what she actually put there and then make sense of this and think of something of putting these sorted pieces into context and if you just dump stuff into a notebook file and then rework the stuff, yeah, fine One answer is Zettelkasten is great for mathematics. Or maybe you can have... Oh, next question. Sorry, I jumped. So my answer is, this was not a question, and I'm happy for you. So that's still an answer. Next question. How Zettelkasten is useful for highly mathematical STEM academic fields like computer science or engineering fields, like when studying a STEM field? I need to unpack the acronym again maybe for the audience. M is mathematics. Engineering is E. Science and tech, tech, techno, I don't know, maybe technical. I could look this up, right? STEM, what was STEM again? I'm doing a bad job. Science, technology, engineering, mathematics. Okay, so probably technology is the computer stuff. Yeah, STEM. Dear audience, that's science, technology, engineering, mathematics, academic fields. What should we make notes of? The important bits are interconnected in a hierarchy, usually chapters of a book. Yes, someone did the job of making the order out of the chaotic findings that people did over hundreds of years for you. That's an order. But there are also many textbooks. So opinions may vary. The presentation may vary. Some textbooks may click with you, while others stay opaque, and you can't get into the meat. Sorry, I'm just typing the answer right now. Copying information to another box is not, yeah. Yeah, let's see what the first answer is. Someone posted that Zettelkasten is great for mathematics, incredible for computer science. With OrgBabel you can have living source code, yes, that you can execute from the node, also true for a formula, and LaTeX in interspersing of images and SVGs. You can even use Agda if you want to. Ah, I think I know who typed the answer. Hello, Oliver. If you want to have living proofs in your nodes, even inline LaTeX is in here. That's, yeah, yeah, yeah. quote the sentence as is and add, yeah, add something, maybe your own summary. I would suggest add your own summary because we're looking at the etherpad, right? Maybe I just copy this over so you can all the either pets indentation is missing with what Org is doing. And the answer was probably submitted by Oliver just in case someone wants to see who that could have been with the Acta reference because nobody uses Acta, nobody in the world maybe so Acta is just a yeah let's say it's good for proofing the code that you type as you type it in the compilation set very weird if you've never heard of this so check that out so question was how Zettelkasten is useful for these kinds of things yeah it sounds like in these kinds of fields that you only deal with facts that's also kind of sort of true for psychology where you have a lot of and also law where you at least in at least in German and our style of law where you have a lot of where you have thick books of law and very very little records, that's not the right phrasing, where you don't rely on the ruling as much as you do in, for example, the USA, where you need to be aware of every court ruling to then put stuff into context and all of this is like the current how you do law. I don't know. I'm not a lawyer and also not good at this in English. So, sorry. If I don't find the cool words and technical terms, you probably all know. What do you do with these kinds of things? You need to collect a lot of stuff. Like the previous questionnaire mentioned, hey, I'm just copying stuff over and then I have a good time and can write about this and create prose in the end. Like with mathematics, you can't. You can't do a lot when you're still working on the foundations. Like, what do you want to do there? You don't innovate there. You just need to learn the stuff. Typing things out and creating your own organization of how things are connected there, that's where the depth of processing increases, like the actual depth of processing so that it takes more effort and affects more parts of your brain, basically, which then helps with remembering stuff. I'm trying not to go into a 10 minute tangent here, but If you've ever tried to memorize stuff with flashcards and you put a, I don't know, coffee stain or a weird scribble or spilled something on one of these cards, actual physical cards there, this card now has another piece of information that is actually important because it makes the card stand out more and more worthy of being remembered because now one of these cards has this coffee circle thingy and all the others don't so this one stands out and this may actually improve how you learn this particular fact that you want to remember or memorize Similar, like what kind of principle do people postulate is at play here? Again, last time I went into this kind of stuff was at university, so I'm not sure if there's something new in the field of flashcard learning. I'm not aware of anything, let's say. And the idea back then that we spread and that was taught to us was like this. That more information and more context and more stimuli and making things more interesting, that's what makes it memorable. if you go for a walk and then associate things with places. Going for a walk is a different activity than imagining going for a walk. It's also different from not doing anything and just sitting in your chair and relaxing because you're not engaged. You're losing, like you're using a bit of your brain matter to do the work while the rest of your body is stagnant. You don't smell anything different. You don't hear, see, feel on your skin, everything's constant, like you're in a, what are these called? These chambers where you're locked into very salty water, I think. Floating chambers, so sensual deprivation chambers. You're depriving yourself of a lot of sensual input if you just try to really focus into everything with your head. If you engage more of your body, if you engage more of your senses, then you have an easier time there. What does it have to do with this answer? My answer would be the value is in rephrasing, in making your own connections eventually. I don't know, the foundational laws of geometry, there's no room for innovation, but there's maybe a room for association. Like, hey, I remember this, like, if you can't figure out what the name of a certain axiom, lemma, whatever it is, and it has a name, adding your own story to the note doesn't hurt. If you say, hey, this is what my grandfather taught me with wire and sticks when we tried to measure whatever in the sand on a vacation one day. And these kinds of things make it more memorable. And that's at least something that you could need to then, which is the ultimate purpose of STEM at university, then you can pass the tests. Like that's it. You either, and eventually you embody the knowledge, right? If you do this stuff over and over again, then you don't need to look up how to do a certain kind of proof. And you just do the proof when you need to do the proof. But maybe you forget all the kinds of weird proofs that you encountered in your life because you only remember the 10 most weird ones, but you had 15 in your life. And what's about the, what do you do with the five missing ones? They just fade away. But if you capture them in some way or another, as superficial as it may be, and you still have access to this later when you need to check your personal inventory of mathematical proofs so my answer would be let's type in my answer on either pad would be I'm not able to type. Sorry. My answer would be this. Foundational knowledge is important for tests mostly and eventually becomes your inventory of skills. Synth programming, like I don't need to look up for loops, I just do for loops and that's it, right? But still, I do have notes about for loops, I think, in different languages, like Python, the, what is it called, list comprehension. I'm not using Python, but it's so quirky. It's quirky and I captured this on a note because I needed to get myself used to the pieces and I needed to look at the syntax anyway, and then this is that part and this goes there. And I did this work, like this separation stuff on a note to then do the work with more engagement, like engage more so I can remember more later. I don't remember all the details, but I do remember there were different parts, and then I can look up the note. And it helps me refresh my memory rather quickly, because it's written by me for me in a way that I'm writing these notes for 15 years or something. 16, at least, 16 years or something. So it's much better than looking, especially now, it's much better than going to the web and doing a web search for the same topic, because I will probably not get the useful resource that I relied on some eight years ago anyway, but some garbage instead, maybe AI generated stuff, I don't know. So I have my personal copy of this and it's not just copying the thing into a box, like from one box into another, from the public internet box into my own buffer, that's a custom that's closer to me, like a fridge is closer to my kitchen and where I want to eat when I'm hungry than going to the supermarket or out into the wild and hunt animals, like this is all about proximity and also brought about ownership, like I own these notes and I still take responsibility for what I put on them to some degree like sometimes they're really bad and I want to distance myself, but still yeah, so that's it, I'm just I'm leaving, I'm not typing this out, I'm leaving copy box fridge, there was a term for this, wait a second, setup custom fridge can be fridge-like buffer of knowledge I'm just, let me just try to copy this note verbatim into the etherpad and then see whether you folks can do something with it. The formatting. See appendix A. I am doing this here. Appendix A. I'm not going to read this, but then you have a FridgeLab buffer of knowledge. Local link in the supply chain. That was the term. Didn't remember that one. It's at least a local link in supply chain. Also, engagement doing the work of not copying, but rephrasing, organizing, but sometimes also just typing the stuff in that's how people romanticize learning a basic like there were magazines with a lot of code listings and they just typed the code it was quite some effort you eventually got better at this but you did do mistakes and then you needed to correct and look through the source code carefully to spot the typo That's apparently virtue, at least people claim it's a virtue. I don't deny it's a virtue because I went through the same kind of crap and I believe it strengthened my character, for sure, but still. Maybe it's not that useful, but maybe it is. I don't know. There's studies I remember, but I don't remember any interesting results. It's not like it's a 10x improvement of recall if you type this, but it may be significant. Just check it out for yourself. I'm not sure. Next question, sorry. In your experience, would you say that you reuse most of your nodes? Watching your demo, I thought that more nodes you create, the less you might reuse them. Yes, that's true. I have 11,000 nodes or something, and I don't use 11,000 nodes every day. I can only use a fraction of them. But let's say you have the Feynman Darling nodes, then you would probably use these. Imagine you have the same 12 Darlings for a year. Then you use these 12 thingies for one year more often. than most of your other nodes, because these become your entry points into what's in my Zettelkasten today. These become your entry points of what do I want to work on today? These become your, where can I put the weird stuff that I just found? Does it stick to any of these surfaces? Entry point into the process. But let's say, five years later, maybe these 12 Feynman darlings were your darlings during university studies for STEM field, where you had a hard time figuring out how the technical system works for your benefit. And then five years later, you're in a job and you don't need all of them because your subject matters from university, they're gone. you're not at a day job and you have different darlings whatever your day job is maybe you're a mathematician at an insurance company so you will probably have insurance related domain darlings pouring in and some academic darlings falling off the bandwagon because no time for these maybe other hobbies come in like i didn't have anything related to child Rearing, bearing is when you produce them and put them into the wild, right? Rearing is when you let them grow up, I don't know Like when you have children and make them not die and get larger and do stuff, right? I did have a couple of things in advance, but That was not a daily concern. Now it is. Now I'm a father and that's a new concern in my life. Getting things done also tackled this, right? If you remember getting things done, everyone, you had areas of responsibility. It's a new area of responsibility and you could represent, sorry, you could represent these in your Zettelkasten, if you find that you can spell out your areas of responsibility well, then you can see, well, I've learned something new, can I become a better person in one of my core areas of responsibility? Yes or no? Why not? Then you will use these more than the other notes. Like the weird thing you found last month. Last month, you will probably not reference that weird thing every day. Natural process. But it's there. And eventually you get back to something, and then you serve access to this like that's the that and whatever this changes like going through these processes change you if I if I spend a month getting nerd sniped on I don't know chocolate manufacturing or something then I don't need this ever these things ever again most likely but going through the procedure of becoming the master chocolate manufacturer of my town, at least in theory, will do something with me when I encounter the next topic. I will think of chocolate manufacturing. I wouldn't have thought of chocolate manufacturing before. In fact, I didn't think about chocolate manufacturing yesterday. I just made this up on the spot. What does this say about me today? Maybe because we were out of chocolate, maybe because I don't know, maybe there's nothing that looks like chocolate here I don't know so it just came up and now it's there and things have changed for the better, for the worse, I don't know but they've changed and that's the real takeaway there where do you put the effort? where do you want to put the effort? and will it pay off? maybe if not, then maybe spend the time on something else like don't try to exactly customize something you hate just to I don't know, do your due diligence if you don't want to because then you'll hate the process that's worse than not using nodes a lot next questions, how are nodes structured and accessed when the nodes grow from 10k to 100k nodes? yes, that's a field of exploration, I mean that the lower realm there I don't know anybody who's using a hyperlinked 100k node archive. I do know people who have a lot of text files like the Evernote style they capture everything and put the verbatim copy into their stuff and then they have a lot more items of course but these don't really count, these are like web clippings of websites themselves I've read how many hundreds of websites today I could have captured everything automatically, but they wouldn't, like the files there, the clippings, they wouldn't have contributed anything meaningfully to my knowledge base, because I'm not thinking about the stuff, I'm thinking about the stuff that I'm spending time on. What about that? Yeah, how are nodes structured and accessed when the nodes go from one to 100k? I will imagine that it becomes a matter that's speculation. We'll see. My speculation is this. it's going to be self-similar, like Mandelbrot images, where you have recursive patterns. First, you have a couple of nodes. You can leave through them digitally or physically easily. And then eventually, you need to rely on structures or some kinds of entry points, which we usually call structures, thanks to Sascha. Perfect. Thank you. And the clue there is, Eventually, you have hundreds or maybe thousands, not hundreds or thousands, of structure nodes. And then you have the same problem that you had on the lower base layer. And now you will probably look for another tool. But there is nothing else that's not also a structure node but is structuring other structures mostly instead of structuring the low-level nodes. But this distinction, I want to point out, doesn't pay off. it seems like it doesn't pay off to say hey i have these foundational permanent nodes and then i have these meta nodes these these maps and structures and then i have these super mega structures which are not permanent which are not structures which are not I don't know this doesn't seem to pay off because eventually there is another level and another level and also sometimes you want to talk about a structure as a data point like hey I did this writing project once it was really cool link to writing project and the writing project is a table of contents huge hierarchy or whatever. Is it not a link to a thing? Back then you get into all these weird academic questions about the nature of links and there's no gold at the end of that rainbow, let's say. At least I've never seen anyone come back with any gold. I've seen people come back with very long faces and sadness in their eyes because all the effort, all the metadata was for naught. So, how do you do this? You figure out, when you get to 10k, you will probably have figured, and if you do 10k, not like web clippings, but really like you did put in effort to get there and type everything yourself in one way or another you would have figured out something that kind of sort of works for you around that milestone. And then I believe in you, you will be equipped to get to 20k, 30k, because you feel comfortable in your notes. And then you can expand further. And it will probably look different for someone in STEM fields than for an artist. Maybe, maybe not. I don't know. Maybe a fine arts artist would woodcrafter, maybe these kinds of folks will have different structures and rely more on visuals, schemas, whole part relationships, whereas mathematicians may not. but I don't know, I don't know, we need to compare and this will be the interesting like if the internet survives and we all survive and also Zettelkasten as an idea sticks on I hope we can have this discussion every year in five years, in ten years and then see how people transparently evolve their gardens, their processes, their note-taking systems next question not a question but i would be very interested in your thoughts on this video by westenberg okay it's it's this is not spam i i don't uh there's joe westenberg's uh youtube channel why i deleted my second brain a journey back to real thinking uh I think I didn't watch the video but read an article. Not sure where. Let me check my Zettelkasten. It's not on screen sharing. That was stupid. I'm probably not going to watch it right now with everyone here, so my thoughts on deleting and stuff like that. I do remember there was a very cleverly written, well-written post and the claim was, from my perspective, and that's not doing the work of It's not doing the work of the author justice. But the shortest summary is that there seems to be false contradiction, where the contradiction stems from how the opposite side, the note taking, the deliberate note taking, taking a lot of notes, all the notes every time, stuff, like how this is presented. I would say this is more on the collecting and not on the connecting side of things. So like I mentioned for the, some 30 minutes ago when I talked about Roam, Roam Research. If you start your day with a daily note, journal style, and then branch out from there, you do at least have time as a structure that helps you to organize what you put in. But also, it can get very messy and very weirdly interconnected, like this meme of the with a guy standing in front of a board with all these lines connecting things and just, you know, with the madness in his eyes you can do a lot of stuff in there. But at a certain point, it gets hard to also get out of your own head, let's say. If you do it in this kind of style, it can get hard to get off your own head. And one litmus test that you may find enjoyable if you struggle with this is to publish more. Because having access to a very deeply interconnected web of stuff thoughts, ideas, notes, wisdoms, whatever very deeply connected, like not a pile, yarn, wool, what's the ball? Ball of yarn? No, that's boring. Let's say like you have a lot of yarn and then you just create chaos and entropy and can't untangle it. It's a very tangled mess. So it's a very dense, dense thingy that's now its own kind of structure. Like if you have this And in your head, when you think about your notes, you may have some interesting pathways there. But you need to serialize and put into a linear order in order to communicate. Otherwise, people cannot follow your thought. it's a pity because we still rely on the book style of writing and also hypertext which is making navigation easy but not telling a story easier in our modern times but we still rely a lot on books which take take a lot of intertwined stuff, put it into one order so you can read from start to finish. But then, in order for you to make sense of this, you need to chop it up, associate with things you already know, and then attach things and create a new kind of structure through the narrative, because it's only one kind of presentation that's given to you. And there may also be different kinds of presentations, like I also mentioned for the STEM fields, where you have different textbooks from different eras of mathematics foundations, and then you see, hey, this one clicks, this one doesn't. Why? It's the same stuff, but they present it differently. And the presentation matters. The presentation matters. And if you only deal with the chaos, the weird entanglement, if you capture everything and anything, how many, I don't know, how many hairs you lost that day or whatever. If you capture everything in there, you don't create the kind of value that is linearizable into thought, which also makes reading your own stuff later, maybe unbearable, maybe at least hard and making sense. Now we come back to one of the earlier questions, like, why do I not use outlines? Because OrgNode allows me to do outlines. Well, I try to write my notes as blog posts, very short blog posts, microblogging, if you will, paragraphs, a couple of paragraphs. Maybe it's also more nested because it's a huge topic. It's a complex topic. That also happens. But then it's more like an article, a long blog post, or an essay, but never just vomiting thought out onto the page, because then you will get back to the note and you will find, hey, there's a lot of vomit. But you never follow this kind of phrase with nice. I enjoyed reading that. It's always me. Oppressing because of its mass, it's also getting in the way of having new ideas because whatever is there and you need to tangle it and then you lost your new genuine idea, this can also happen. So I'm not sure what the Westenberg video contains, but the sentiment there of deleting everything, it makes sense if you find yourself in a situation where you don't have a productive system. and the notes from 2009 in my own Zettelkasten aren't of the highest quality standards I rewrite most of them when I encounter them, but I can still use a couple as is and sometimes they're just garbage, but as Luhmann said garbage like trickles to the ground and forms a nice sediment layer there in the septic tank of your thought machine, your thought thinking environment. But clear water separates and then rises to the top and that's what you work with. it's never going to be 100% useful. But if it's 90% garbage and if the water is, if the septic tank is never clearing, then you probably need to start over and rethink your approach. Maybe there was just too much garbage, too much toxic waste in there, whatever that means, right? So I do understand this, but also I would say maybe maybe they tried to do too much with a different mindset and it just didn't click and then they, you know next question is there a danger? also, wait a second, answer do not probably a lot of collection, not a lot of stinking in the Zettelkasten to some purpose. Usually. Conflation was the word I was fishing for. Usually posts videos like this. Conflate note taking as a dumping ground for stuff with something that is valuable in the long term. Also if this was our dear friend Westenberg I think the note count also was higher, but I might confuse them with someone else. The note count was way higher than mine. Note count. But then also, you know, there's the sign of going for quantity in... just capture anything, like capturing, that's not the bottleneck. The bottleneck is finding, making connections happen, having thoughts and working on problems you really want to solve in the long term. So next question. Is there a danger that with the settled custom process, that the process gets a bit in the way of the content? yes, especially if you do it like old school style if you find that your handwriting sucks and you want to type things on a typewriter like that kind of process definitely gets in the way that's going to be your bottleneck you can't find black paper and put it into your typewriter and then like that's not the speed of thought that's the speed of, I don't know, my long deceased grandmother but yeah it's not like a strive for speed of thought tools not needing a typewriter and then next step reduce ritual reduce rituals and ceremony What I mean by that is if you find yourself doing the same moves or similar moves over and over again, but they don't contribute to having more cool ideas per week, then maybe ditch these, apart from the day timestamp, having a title, hashtags, and content, and also being very, very careful about quoting my sources. I don't do much with my et cetera. I cannot give up quoting my sources because then I'm plagiarizing, stealing, forgetting where I get ideas from and confuse. And I get out because I'm aware of the problem there. I'm also not trusting my own notes anymore. would not be trusting my own notes anymore i need to properly cite so then i can see hey this block has no quotations that's mine and i can trust this it's not like hey this is maybe mine maybe this is a verbatim quote from a book like this is not helping that's a ceremony that's paying off in the long term also so that when you write something you have all the sources at hand that's a useful ceremony title is just a way of summarizing the content do it like an excerpt, right? you have the whole content, that's your etc. do a short summary, that may be your introduction. So have a way to quickly get a glimpse of the relevance before having to read the whole settle when you encounter it years from now. And then have a one, that's one sentence summary, and then have an even more condensed form that's a title. The title is basically your API, the interface for you and your nodes. And it needs to uphold some kind of contract. And if it just is, like, this is my second note, that's not useful. This is just a technical limitation because you can't have the same file name twice. That's not going to be useful. But if you give yourself the permission to take time of titling your notes in one way and also changing titles, because you will know from writing introductions the further you get along, the more and more you need to rewrite your introductions because your understanding changes same for titles titles don't always stay consistent or constant that's why I don't link by title, because I would then need to go through my node archive and also need to go to my node archive and need to correct all the links that's a janitorial task I like to not do, if I can So answer here, less ceremony, fewer ceremonies, plural. Try to only do the minimum possible, minimum viable thing. And using Emacs and D-Node, you get a lot of defaults. And then you need to worry not about IDs and tags that much, or where to put them and how to put them. You can omit keywords. You can leave keywords out. They don't scale anyway. But they may be interesting as a way to group, but also Sascha from zettelkasten.de, from our website, Sascha also found that, I believe, he, structures win in the long term. Manual created hubs and overviews win over machine generated lumps. Because you need to reorient yourself in the search results all the time. That's not as useful as having a pathway through your nodes that you can follow. It may not be a pathway through all of your nodes, of course, but it may be a pathway from entry point into some topic. That's why some things are essential and others aren't. And the essential stuff, try to do it quickly, get better at it, practice the movements, then they get out of the way, and then you can focus on having ideas and typing them at speed of thought, more or less, and then you're good. I hope. Wishing you all the best. Next and last question. How do you navigate looking at all posts with a certain text? This might be an emacs, already new to emacs. I just may have answered this. How do you navigate looking at all posts with a certain text? Let me go to my actual node archive. and then php no language learning ah this is not my setup I'm very clumsy I'm sorry, I don't know swift language learning ah that's that's a boring one php By this time, I could have also, ah, there's the model. This would be an entry point into my PHP atom molecule organism learning journey where I tried to practice, what's visual line mode, sorry, where I tried to practice this atom molecule thing on something new. I haven't used PHP in a long time and found that I wasn't, yeah, you can see it on the screen also like language primitives are the atoms, I think molecules are more like idioms or patterns that are PHP special in PHP and different in other languages in Python would be list comprehension that's in a molecule and but also you can make a point for hey, this comprehension is more like a syntactical thing, no no get out of here, like the for loop is the syntactical atom and the molecule is using the for loop with weird other stuff to then make the list comprehension happen. That's my argument there, okay? So what do we do with this? I use this as an entry point if I want to see, hey, what kind of stuff did I take notes on to get a better understanding on multiple levels of the language. And project structure conventions and usually another language that I use more here would be, but I can't remember now. Funny, right? I can't remember how I would need to look, but I think I'm too clumsy in Emacs to do it live on camera, sorry. But the organisms part here would also be like my own inventions, let's say in code, like my own discoveries where I found that in Swift, for example, you can express things on the type level in a very sexy way sometimes. And then I would put these there as well. So these are also, There are organisms to discover in the language reference, but there are also organisms to discover in the world, and then they are your own inventions, discoveries, more or less. And that's a sign of mastery. If you work more on these things, then you don't need to worry about the basics of the language anymore. So I can use this as an entry point to get into things that are very PHP-esque. When do I need this? When I want to talk about PHP, like I do now. Why do I show this when the question is about tags? Because if I search for... Which one do I take? Let's try grep. php. Yep, grep exited the anomaly. Dang, why? Option requires an argument. Yes, eshell is there. So let's just use rg, ripgrep and then ripgrep for php. And you see I get, I can do filenames only, right? Let's not worry. I get a lot of, let's just scroll up. I get a lot more hits here for the PHP tech, which is only about PHP specific things, not about anything that I do in PHP. Like, I don't know, I don't take Laravel or Filament, which are frameworks, libraries in PHP, the Zend library or whatever, I don't know many. Like I wouldn't take everything and anything that is related to the php ecosystem with the php hashtag. Why? Because then the tag becomes useless. I only tag things that are about the language. Like how do I pass variables to php includes that's php related. And you see these are... I'm still not at the top. now i'm at the top these are a lot more findings than you could these don't fit on the screen these are three pages i believe and the note about my language learning journey was wasn't even one full page here's white space there wasn't okay right but still there's much more stuff the hashtag I don't use for navigating. I only use this, let's say, the shotgun approach. I think Sascha on our Zettelkasten blog post, don't remember which post, mentioned this as well. Sometimes you need a sniper rifle. Sometimes you need a shotgun. And the hashtag-based search would be I don't know what I'm looking for. It was PHP related. I hope I use the hashtag correctly and then kill all these nodes from my archive. The metaphor doesn't work that way, but still you get a lot of results here and you could use this as a way to filter from 11,000 nodes down to 50 or so, maybe less. And if you got this far, then you can start to do it in a mechanical way again. But wait, if you have 100 or 1000 nodes in the results there, then you can't mechanically go through all the nodes again. You can maybe skim and look for something that catches your eye, if you're lucky, and if that's the thing that works for you. but usually no, like usually the text become useless then maybe tag groups become more useful like PHP and LSP which is on screen here like then maybe the combination of two tags that reduces the search terms enough like this is basically glorified full text search where you highlight certain terms of importance that's it, I wouldn't use I wouldn't put more effort in there It's also maybe useful to group things by topic. For example, study notes like, I don't know, this is my undergrad, third semester, what do you do in the third semester, I don't know, history of philosophy of science of birds, very long course name, and then week 10. And then you assemble everything with this very technical tagging. Then you can use these folders to loosely group stuff together that happened during that week. And if you have 10 required readings and then do five of them, but they're also not closely interrelated, you will have read five to 10 different sources that you could process. You could maybe then also create an overview node as a preparation for the seminar whatever happens at university. Then the hashtag becomes just a way to group things. But once you create the structure node, you don't need the hashtags to navigate anymore. You could then maybe use the hashtags not to navigate but to to clump, to create collections, pre-assembled to some degree collections of things that exist already, get them out there and then review them. Like Twyla Tharp again, dumping all the notes on a table, all the stuff on the table, not just notes, all the clippings and so on and so on, and then making sense of them. That's what you can use a hashtag search for in the longterm. And I'm not convinced about navigational use in the longterm. which doesn't mean that it won't work for you, but I can't recommend it. I can't recommend trying to do it that way. So my answer is going to be, how do you navigate looking at all posts with certain tags? I don't. Tags are just a way to loosely group notes and stuff in general so that you can find potentially, PP, why doesn't P work? Potentially relevant pieces of information more quickly. Navigational aids. Create navigational aids things that help you navigate like this is for a week what did I say 10 of semester three uh the history of the sociology of the philosophy of science of birds or something lump things together there then you have a then you have a scratch pad to think on you do this did do this when you went to university with paper anyway, right? You would get your college book or whatever and then you would take notes and these like that location of that piece of paper in your college book in your whatever notes folder that's the important piece of information and the scratchpad, the thinking environment for this course and if you have all the other things that you ever did at your disposal as well maybe pull them in maybe pull them in and then bring them to the discussion with your professor or your teaching assistant and asking about, well, I read these five books that you recommended for this week. Thank you for recommending five whole books for one week of reading. I really enjoyed that. Long pause. And then you say, I also thought about other topics and brought them into a, I don't know, dialectic or position. What do you think about foxes in that regard and their own history and subjectivity when it comes to the bird's point of view or something, right? You can only do this if you either know this by heart because you do this a lot, think a lot about this stuff or you do this, you stumble upon this and then look for oppositions, look for certain patterns in your notes and then try to pull them in and represent them again so they actually do exist and then you can work with this, you can get on people's nerves you can write weird papers, essays and then get through your university studies quickly maybe also become an interesting columnist, I don't know, I've never been a columnist Right. I think these are all the questions. I'm going to delete the leftover ones. I hope you enjoyed some of the answers. Thank you so much. Thank you. That's a lot. We'll go back and get everything transcribed as well and then you can turn that into other articles in the future because that's how this all works, right? It all turns into writing. It all does. All the rest is garbage. I don't know why we invented anything else. It's only useful to transfer language before writing, but now you can just read. We have two people in the big blue button room who are still around, but I think they've got their, like, they don't have microphones set up. So I don't know if you want to just, like, CryptKNFL, if you want to just follow up through the etherpad, or email, I guess, in case you have further questions. But yeah, if you want to wrap up, I know it's getting a little bit late over there, you're in your time zone. Thank you so much. Yeah, all right. Thank you. You're welcome. Thanks for joining and listening. Okay, I'll wrap it up here then.

Questions or comments? Please e-mail hi@christiantietze.de

Back to the talks Previous by track: corfu+yasnippet: Easier than I thought Next by track: Questions and answers to help you fly with Hyperbole Track: General