After trying so many productivtiy tools in the past I settled on having two ordinary notebooks where I keep all my ideas and notes: One for the business side, where I write down meeting minutes and insights, in chronological order. Another one for technical and creative thoughts, also organized in a linear, chronological fashion. In addition I sometimes keep lab notebooks for specific projects.
If I look for some specific information (e.g. whom I talked to in a given meeting and what the take-aways were) I typically know the rough date range when I wrote that information down, so I just go through the notebook entries until I find it. Same with the lab and creative notebooks.
Another advantage is that writing by hand allows you to draw diagrams and write equations with ease. In my opinion nothing beats a paper notebook in this regard. I tried the new iPads and MS Surface tablets but the note-taking experience is just not the same and the results look disappointing IMHO. Also, if I really need to record an image or computer-generated diagram I simply print it, cut it out and glue it into the notebook.
My former colleagues in science used the same method of keeping lab books for decades with great success, I have never heard of anyone using a system like Zettelkasten (but maybe that's just my bubble).
Totally agree. Organizing notes into hierarchies is not necessary, most of the time you just care about whether it’s a business note or personal. Chronology is an organization method already available regardless of the note data, and we as humans experience most things rooted in time so it’s easy to recall a rough range as well.
I have also done the dance of going through various productivity tools and have often ended up back in the paper notebook camp. If you are interested in trying another tool out, I’d really love to know what you think of an app I am building. It is meant to be very lightweight with just one screen, and start up really fast, but when logged in, also lets you sync across all devices you have.
It’s available at https://notebrook.app with code ALPHA2020. Everyone is welcome to try. hello@notebrook.com for feedback. I think it’s striking a good balance so far of being extremely lightweight but also let’s you recall when you need.
> Organizing notes into hierarchies is not necessary, most of the time you just care about whether it’s a business note or personal.
...for some (most?) people. For others, like me, it simply is necessary - we have too many notes, ideas, reminders, thoughts, whatever, of too wide a variety, to use the system described by GP comment, or yours.
2. I also require that the notepads have perforations on every page so that individual pages can be torn out. This is important for keeping the notepad uncluttered -- and lowers the inhibition to freeflow write, because if things don't work out, the page can be torn out.
The two biggest takeaways from Bullet Journal for me were the task markup you mention and numbered pages + index. When I have to refer to some previous work I did, I note the past page numbers I had to look up. It's a simple way of cross-linking notes, but I find that it works quite well for me.
I try to avoid notebooks with tearable pages, because I found them to be too fragile. I carry my most recent notebook everywhere and some of them end up quite beat up. Tearable pages tend to start falling out on their own.
> I try to avoid notebooks with tearable pages, because I found them to be too fragile.
So in my experience Rhodia (and other "higher-end" perforated) notepads are pretty sturdy and tend not to fall apart even with rougher than usual handling (mine haven't). There is a slight price premium however.
I believe it works well for you. For me one of the important parts that got me to consistently write a journal years ago was also switching to the cheapest notebooks I could get.
With something (relatively) expensive I always had that small, subconscious fear that what I have to write in the notebook wasn't worth ruining the page.
One way to avoid this is to buy something like Rhodia that you want to use but is a bit more expensive. Then buy one even more expensive notebook. Now the Rhodia is your cheaper notebook and there is no longer an inhibition to writing in it!
I adopted the exact same technique this year after making the jump from engineering to product management.
Previously, everything was scattered in notes on sublime or notion. This is because in the dev world, everything was happening primarily over email or slack, if i ever did have a call with a client, i just made dot points in a text editor and that was it.
My first week as a PM I realized how much more meetings and talking there are. So, I switched to a notebook, but I quickly found the issue of general meeting and “standup” notes being scattered amongst drawings and diagrams etc.
Hence, I now also have two small notebooks. One is for sketches, rough wireframes or ideas, the other is a daily tasks which I use as my standup notes.
Of course, processes can always be further refined, and I’m now also incorporating notion back into my work process.
Paper and digital note systems are not in conflict. When I'm working through the math in a paper, I'm not going to type 50 lines of equations into a text editor. I take a picture of my notes, upload it to OneNote, and copy the link into my text notes on the paper. It's surprising, but you can have a 100% digital note system while using paper for 90% of your notetaking.
For years I took notes both with pen/paper and digitally, and either cross-referenced things or scanned pages into a master digital document like you describe. For me, the game-changer has been getting a RocketBook. It’s handwriting recognition is about 98% accurate for me, so it only takes a few minutes for me to scan some pages, drag the emailed text file into VS Code, join lines into paragraphs, and check for typos. Gives me a chance to review what I wrote too. Then it goes from there into a plain text file (or more recently, WorkFlowy) which is my preferred system. I started with the free PDFs that RocketBook makes available, but finally bought one mostly because I wanted to pay for their server resources. :-)
The RocketBook is actually what got me started on this. My pen died, so I started using the OneNote app (service provided by my employer) and it's worked well enough. I definitely recommend the RocketBook.
This is why, for digital use (since I cannot write without pain these days), Microsoft Word is imo the best digital notes software. Full text search, tag integration (#tag), images, hyperlinks, tables, diagrams, equations, and macros (Macros are incredibly awesome.)
There is OneNote but I dislike the org structure and the walled garden approach.
Zettelkasten can be done on paper too. Its point is making connections between notes and being able to do discover new connections giving you new insights.
Simple note taking just records what you know and think. ZH
can lead to new understandings of the same set of notes.
You can use a date as a "link" in a notebook. A normal person does not create that many different entries per day. You can add a keyword to the notebook margin as well if you feel particularly systematic (then a link/reference would be 18jul23 for example).
I dont see myself following permanent links by timestamp. It is so time-consuming and error-prone, I would drop the whole thing immediately. I know myself this far anyway^^
The idea is that it is a _good thing_ that it is time consuming, as thinking about the connections between the notes _is_ the actual productive work that you have the system for.
Another major benefit of pen on paper is it's far less tiring, at least for me. I find staring at screens for extended periods utterly draining and often use paper for notes and code reading.
Major disadvantages - lack of relatively easy search, security, duplication, archiving, integration, accessibility to others, etc.
So it's a complementary tool for me, but the digital-only brigade are missing a useful, and - vital to some - dimension.
Searching for information in analog notebooks can be a pain. Also they might get lost, spilled over with water, burned, eaten by the dog, etc... easily. In that respect digital notes with a backup in the cloud are superior.
However I agree that no digital solution beats paper so far when it comes to drawing diagrams or writing mathematical formulas. However the conversion of handwriting to LateX by https://webdemo.myscript.com/ is quite impressive.
I'm wondering: is there _any_ empirical data that shows that good old note-taking with a pen and paper is less effective when compared to Zettelkasten note-taking? (Disregarding advantages of digital note-taking such as easy to search, etc.)
I personally use successive relearning, sometimes also referred to as retrieval-based learning, it's not a note-taking technique per se, but more of a framework on how to learn/study in a way that maximizes retention and meaningful learning.
If anyone's interested in it, here are some resources:
It's only the latest hype around digital Zettelkastens that tout the supposed benefits of mass-linking notes or generating fancy graphs (which I find to be a gimmick with little practical value).
Like any note taking system: Your potential output is what and how much you put into it. No note taking solution - digital or analog - magically produces results without countless hours of input.
Key for non Zettelkästen is to refactor notes and delete/store those older notes, this way you also throw away part of that nested hierarchy. You simply combine it with your own trajectory and ongoing inevitable specialization, software notes would house the notes about software that you actually use.
I'm not a big fan of Zettelkästen for this reason, there is no need for those if you go through note refactors and make it a yearly routine (I combine it with OS reinstalls and trying new programs). I don't care about note program longevity for this reason, I'm not bound to a particular program because refactoring also means rewriting many notes from scratch, based on older notes and current knowledge.
This refactoring/rewriting is also a good refresher and helpful for really internalizing broader concepts (which you then can apply to other seemingly similar areas).
You can have tags instead of hierarchy and then you also have a network.
So instead of "java notes" and "design notes" you add the "java" and "design" tag, etc., so you can approach your notes in any direction: you can find "java" notes and then narrow the result sets to design notes (so java+design), and vice versa.
And you can add links too between the nodes, so you can also walk them in any order you define.
Existing tools can do this already, it's just a matter of using them.
Is there any more to zettelkasten? Its author invented the same thing on paper notes, but digital tools provide this naturally.
Zettelkasten is very simple and that is the point: You create small and connected notes. This constraint turns out to be useful for research and writing for some people.
hm, now that i am reading this, would that basically be the equivalent of paragraph tagging/paragraph connections in a longer document?
i.e. you use one note document for 1 specific book and it might be 1500 words. but instead of linking the documents, you link specific words/paragraphs to other notes/paragraphs?
Yes, you could easily create a Zettelkasten in a single file. Or with pen and paper.
The gist is focused and self-contained notes, linked together to form structure and connections. A personal wiki with attention of recontextualization.
That wasn't necessarily my point. I cannot visiualise how you use self-contained notes for something like book notes or meeting notes, browsing random links just seems completely inefficient and not transporting the actual meaning/context of the notes
The primary change of thinking that a Zettelkasten system requires is giving up the idea of keeping notes about a book or meeting. Instead, keep notes about topics you care about, using the book or meeting as a source.
You are not browsing the notes at “random”. You read your notes because you are working on something and what to see how your collection of notes might help you.
It takes effort to support this workflow. Notes have to be self-contained and of a somewhat permanent relevance. You don’t put raw reading and meeting notes there.
Zettelkasten fights this complexity by much more up-front work when adding and linking note, instead of a tree structure, you have a network.
I wonder if I am doing it wrong. I certainly organize my notes into tree format when I am doing intensive reading on a book. This is to keep my head straight on what topics had been touched on by that book.
On occasions, I do opportunistic linking, however.
> This is to keep my head straight on what topics had been touched on by that book.
Most of the time, it’s more be beneficial to file notes according to the situation in which they’ll be useful rather than where they came from: If you’re going to have a tree structure, the original sources should be out at the leaves as external references rather than the root. This manifests in many forms from lots of different people giving advice:
In Getting Things Done, Allen spends a lot of time on the importance of organizing your todo lists by where you’ll be able to do the actions.
Luhmann used his original Zettelkasten to store passages that he could pull to make drafts of papers, and cross-referenced them to other passages that could be included together.
In How to Write a Thesis, Eco recommends writing a preliminary outline of your thesis and then tagging notes with the section number they’re relevant to.
In his MasterClass series, Chris Hadfield emphasizes the benefit of collecting summary notes organized by the interface you’ll see when actually performing an activity.
I am wondering, are you selective on what you take notes too or just take lots of notes? I.e. do you take lots of notes in meetings and every website/paper/book you are reading?
As far as I can tell, the crosslinking required by the Zettelkasten approach (1) provides two main benefits:
- It forces some retrieval practice of the notes you’ve taken before, which reinforces all the ideas involved.
- Most of the benefit comes from the act of writing, but there’s an inevitable sense of futility that comes about if those writings are inevitably lost to time. By giving each note an ongoing purpose (to be linked to), the Zettelkasten system dodges this particular trigger to stop writing notes.
(1) as used by Luhmann; not necessarily any of the electronic tools
Research is difficult and results don't generalize that well. People are so different. Some can remember anything they read, while others don't remember even if they do take notes.
A better approach is also difficult, ie. asking yourself, do I need better note-taking and memorization of topics? Due to Dunner-Kruger effects, we rarely ponder this. Nowadays, a websearch is a very quick and accessible cop-out and, in many cases (not all), more effective.
Research typically show that pencil and paper works best for memorization. In my experience, I never refer to these notes later however.
Wiki was a good idea in 1995, but taking notes shouldn't require arcane knowledge and have arbitrary implementation limits.
Markdown is slightly better. As a tool, I prefer WYSIWYG into an accessible format.
For TODO-notes, a simple text-file serve me well.
However, for note-taking that I want to refer to later, mindmaps works best digitally in my personal experience, both for memorization and reference.
A free version, though a bit limited, is FreeMind. Works well enough to navigate to find what you need, provided proper node names and structure.
I don't think that the note taking part itself with a physical notebook can be beaten, yet a digital Zettelkasten has huge advantages that come after the note taking part: add links, images, formulas, search, etc.
The main reason it's so hip because Luhmann used it.
So, if your work is similar to Luhmanns and your cognitive profile is similar to Luhmans it makes perfect sense to try it out.
But I have no idea how you could test it's effectiveness in any sense that would provide results that are scientifically or statistically sane in any way.
Personally I've tried a "zettelkasten light" and I've found that I can keep my personal random research findings better organized that way - so instead of having just a page in a notebook, building an actual index of the notes as well, and linking thematically identical topics together seems to keep my thoughts better together. Been doing it only for a few months, though, need to see what I think about it in few years.
IMHO he is also the best example for why this method doesn't work that well. His books lack focus. They are collection of thoughts with no mention of the sources. I say this as a trained sociologist who once was a proponent of his ideas.
Can you expand on that? Everyone talks about Zettelkasten but not the products of Zettelkasten. It'd be nice to know more about your opinion on Luhmann's work.
This is anecdotal, but I recall reading somewhere (maybe in a foreword) that e.g. Luhmann's "Social Systems" (1984; English translation 1985) was composed this way in purpose. The reader was supposed to be able to start the book from any chapter without feeling that he/she is missing something.
But, once more, this is anecdotal recall.
A similarly structured contemporary book is David Fleming's "Lean Logic: Dictionary for the Future and How to Survive It" [1]. A true masterpiece of composition IMO, and a work that took the author 30 years.
Interestingly, this work ethic also links to Luhmann's views on the time span of research relevant to him. Quoting a paper I found on the matter [2]:
" In 1962 he received a scholarship to Harvard and spent a year with Talcott Parsons. In 1968, he was appointed professor of sociology at the newly established University of Bielefeld, where he worked until his retirement. Shortly before his appointment he was asked on what subject he wished to work at university. His reply was: “The theory of modern society. Duration 30 years; no costs.” He consequently realised exactly this theoretical program. At the time of his death in December 1998, at the age of 70, he had published an oeuvre of over 14,000 printed pages."
The problem is that for many years his work was way to overrated in German speaking countries. His work also promoted the decline of sociology into irrelevance. One could argue though that his popularity was rather a symptom than the cause for this.
This. ZK works very well, but not in all areas. People in the humanities are more likely to benefit from it than, say, software engineers. I tried making software-related notes in a ZK and it just didn't _feel_ right. Maybe I should give it another go.
Org-mode is my go-to tool for any kind of notes or document, sometimes even for PDFs I want to make, via latex export and then latexmk. I find it more comfortable to use for literate programming than Jupyter notebooks too.
Thank you for mentioning org-mode. I was actually finding a good set-up. You're comment leads me to this blog[0] that I end-up following this set-up which is quite very simple.
On the other hand, I recently saw a lot of Zettelkasten articles. I'm not sure if its been hyped but I still would like to give it a try. When I found emacs, I gave it a try (it was hard) and eventually I begin to like it because of its tons of features. When I found org-mode, I gave it a try and then I like it and use it everyday. When I found org-babel, I gave it a try and then I like it. When ... ... ...
I hope more people would like to try emacs and org-mode. I know: learning at first is pretty hard. My advise it just to use it everyday. I remember I uninstalled notepad++ from my PC in order to force my self to use emacs. It took a month for my brain to be comfortable using emacs.
Org-roam is great for Zettelkasten style personal knowledge base wiki. It builds on org-mode and shows the backlinks that make a Roam database so useful.
For those of us struggling with just what Roam or Zettelkasten are, or are wondering why not just use a wiki: Roam is to a wiki what Markdown is to a text editor. That is to say it's a more focused usage of the editor, with helper functions designed to streamline the workflow that is part of the Roam/Zettelkasten "system"
----
My own setup in Spacemacs (Emacs with Vim-like leader-key composed functions) is arranged around a Zettelkasten/Roam wiki with daily journal notes and TODO items. (Org-mode is a fantastically complex piece of code, listing its features for organization would take too long, but this has some: https://orgmode.org/features.html )
Daily note-taking is done in org-journal, with TODO items scattered in the notes. Creating a new day's note brings forward all the unfinished org-mode TODO items into the new document. More complex stuff gets linked to as an external document. Example: 2020-06-07.org may have an entry "Write new disaster recovery requirements" and within that entry have a link to the "ProjectName Disaster Recovery Requirements" document. The markup in org-mode for that task now that I'm working on it, looks like this:
* DOING [#A] Write new disaster recovery requirements
:LOGBOOK:
- State "NEXT" from "TODO" [2020-06-05 Fri 16:14]
- State "DOING" from "TODO" [2020-06-06 Sat 12:07]
:END:
[[file:~/Org/projectname_disaster_recovery_requirements.org][ProjectName Disaster Recovery Requirements]]
Now when I have the disaster recovery requirements file open, org-roam will show me which docs link to it. Which provides a big boost to my recall of important or associated things. For example the backlinks window shows me I linked to the disaster recovery doc from meeting notes a couple of months old. To refresh my memory I click through and read the meeting notes, to discover someone mentioned a new feature of our software that makes disaster recovery more difficult, or some new corporate policy that affects how we store backups, etc etc.
----
Why should people care about this? In my opinion, it's because we as "computer people" have a vast amount of knowledge we are expected to be able to use on whatever problems we encounter in our work. I think we need all the help we can get to remember that knowledge and its relevance to a given situation or subject. I personally am pretty loose about how I write my personal knowledge base, I don't write theses or long-form articles with perfect bibtex references, but I do try to get what I'm doing into text form so I can search for it later on.
I'm excited to see how this "personal knowledge management" ecosystem grows and develops. We're only human but we're expected to manage godlike amounts of information.
Is Zettelkasten better than 'the big .txt file'? [1] Sure, the example here is about productivity, but I know it can be used for general notes. Anything from your to do list, to notes from books, to ideas you want to remember. You want to remember what you wrote about a book on gardening? Seach gardening.
This seems to be a low effort, high reward way of doing things. What extra benefit does Zettelkasten bring and is it worth the additional effort?
EDIT: Another consideration is that you don't want the time you dedicate to sorting a productivity system to be especially big. You want to make stuff. The big txt file doesn't take up a lot of time to organise and allows you to spend time doing good things.
>> Is Zettelkasten better than 'the big .txt file'?
Might be but the ROI certainly isn't. That's why I use one large text file also. One relevant detail is not to consider that file read only, you must tree shake it regularly. Noise will be less and relevant stuff will be easier to find.
I use sections (materialised as markdown titles so I can easily navigate through them in a text editor) for recurrent subjects (so in some way this is my tagging) and never date anything unless is relevant to do so.
I tried using a text file for quite a while but being restricted to text was too limiting. Often a photo, voicememo, or sketch is the most efficient way to record something.
I use it to 1) store ideas, concepts, etc. in a markdown format. So, if I'm studying something - say backpropagation for neural networks - I create the note, search for material online, study it, and add links, images, formulas, etc. to the note so I can merge several good sources of information that would otherwise be scattered in PDFs, stack overflow, wikipedia, ... Then afterwards I can go back and read my note and maybe visit the links I used to obtain the initial understanding. It's really great so far.
2) I can link the notes to create a sort of personal wikipedia. So far I'm pretty happy with Zettlr for this purpose.
I find that even if I don't have to go back to many notes or actively walk through the graph it's really useful for distilling knowledge.
Zettlekasten was designed as a research tool, not a productivity system(for that I use task lists and progress bar).
Anyway, I used grep too, but I also navigate by following the links in my notes, which can sometime take me to different location. That can be a single file too or multiple files. That doesn't matter. What matters more is organization(or rather lack of).
With Zettlekasten, you build a concept map of topics that you wouldn't otherwise get if you write your notes linearly or chronologically.
So yeah, zettlekasten doesn't make sense as a productivity system, but that isn't its intended use case anyway.
Sure, to be clear, I meant using a big txt file as a place where you store notes from books, or your own ideas. Sure, you can also use it as a calendar or other organisational tasks as well, but is Zettlekasten better than a big txt file for storing (and referring back to) notes?
This is a false dichotomy: you can create a Zettelkasten in a single TXT file. The implementation (e.g. using multiple files, one per note) is just a convenient way to see your stuff in a file listing and note taking apps, for example. But as long as you can link between notes, you're good, and a single-file Zettelkasten works just fine.
> Is Zettelkasten better than 'the big .txt file'?
I tried it, and for me it is not.
The ideas stored are too granular, and the linking becomes unwieldy after a while.
I'd rather store small ideas in a linear big txt file (I use Google Docs), revisit parts from time-to-time and collect/coalesce similar ideas and rewrite them into bigger ideas.
This process of reviewing and rewriting has allowed me to cluster ideas and gain clarity over time. Inspiration is lumpy so everyone needs a cache of thoughts. But these ideas also need to be regularly iterated upon and synthesized into coherent narratives if they are to be useful.
Shameless plug: I literally just pushed a change[0] to my online note-taking app[1] last night that makes it a "full" Zettelkasten (e.g. there are backlinks now!) There are also some other cool features like multi-parent nesting that (in my opinion) make it much cooler than just a Zettelkasten. The main downside for many in the HN community is that it's online-only at the moment.
"Online-only" is not a deal breaker for me, but closed source and the missing possibility to self host it kinda is. But I'm gonna give it a try nonetheless!
Thanks for the feedback! One of our goals is to have a robust API that anyone can build apps for so that you can integrate this with any sort of platform you want (and then obviously the client could be open source), but yes, otherwise it is closed-source.
The source code is not available, and each mouseover triggers a lot of Ajax calls (one for the note hovered, one for each link in the hovered note), but the experience is quite pleasant.
My problem with Zettelkasten is the granularity of notes which are supposed to be a single short concept only. I can't remember hundreds of notes I wrote in the past so that I can connect them when I write a new one.
I actually prefer Martin Fowler's concept of bliki (blog+wiki) where the granularity is a blog post, instead of a singular note, which captures your understanding of a topic at that time.
The idea behind the Zettelkasten is that you traverse it a lot. So it's not like "take a note and forget about it", it's more like write a bit and read it a lot of times. Luhmann wrote his books that way. So it's more about reading than taking the notes in the first place.
This thing that apparently worked very well for that sociologist that mostly likely operated before the iper-connected and mostly digital world that we currently live in (he died in 1998).
It's now being marketed as another holy grail solution, it reminds me of the GTD method back in the day (in terms of marketing).
Index cards and custom sized drawers? lol nope.
To add some constructive feedback: I'm playing with MediaWiki and i noticed this nice links on the right side of all pages: "What links here". It reminded me of this Zettelkasten thing. But without much fuss.
Think of Zettlekasten as a wiki-with-a-narrow-focus. This site https://roamresearch.com/ can be considered a wiki, but its features are all designed around journals, linking, tags, etc.
Something like how a text editor can be configured to be a very useful IDE for a given programming language.
To your last point: there's now ZK-focused software which has exactly that: a 'what links here' button on the side, which sort of removes the fuss. (Obsidian, to name one)
I put my markdown files in a Dropbox folder, so I can use apps like 1writer on iOS. Syncs well and works like a charm!
Whenever I have an interesting thought, I get my phone, open 1writer, it creates a new [timestamp].md file, and I just jot it down. It instantly saves to dropbox so if I open Sublime on my laptop, the file's already there.
On my PC, I mostly write in Sublime Text but now experimenting with Zettlr and Obsidian.
I heard heard other combinations with other cloud file services and other markdown writing apps work just as good.
Having a folder with plaintext notes going back years just feels ... future-proof.
Here is my personal experience after applying this method for more than a year. The crucial idea of the zettelkasten method is to contextualize knowledge instead of categorizig it. So its processing information instead of just storing it. Hence retrieving knowledge is much easier because its much more connected in your brain. I experienced this as extremely powerful, my thoughts became much more internalized. Cross-linking becomes the standard, which is a main driver of creativity and idea shaping.
Reading the linked article I'm very surprised that the author says they can't talk about what they read after reading it. I'm quite the opposite as I often talk about what I've read to my friends, even before finishing the book. Sure I might not remember things as well a year later, but then I can go back to the ToC or index and find the bit I've forgotten. Saying that books don't work in general doesn't ring true to me.
This is the fourth explanation that I've seen on HN this year. Including my own [1]. I like the more practical approach of this one, but the long numbers (timestamps) used for referencing look confusing. It's easier to remember the title of a note or a number you made up yourself.
In case you're looking for a visual editor, you can take a look at Emvi [2]. We didn't build it exactly as a Zettelkasten, but you can reach that goal in pretty much any note taking app. Ours is more focused on collaboration.
Every time I see a new method of note taking I file it under notes for notetaking and never get around to it.
These days I do everything in pen and paper. I file it in a file cabinet. I think this is what Fermi used to do, could be wrong. It works for me. No dependencies, no installations, no updates, no latency, pretty decent storage and great portability.
I make an effort not to dwell on productivity tools, which feel like spending time to save time in spurious and inconsistent amounts. Yes, I wish I could memorize and remember everything worth knowing and doing. No, sometimes it's nice to forget things and prioritize the information that's in front of me.
I keep seeing articles about Zettelkasten pop up every couple of weeks on HN lately. Notetaking and productivity articles tend to get me thinking about how my own system works (I still have a bunch of Zettelkasten articles I would like to read...)
I've been using Anki as my SRS system for language recall for years. A couple of months ago I started putting in random facts and bits of information that I would normally put into a note. This sort of thing tends to get inputted, filed, and forgotten about until I do a purge of notes every few months. But since I've started using Anki for this, I now 100% remember these types of random facts. A few examples:
- FQ_CODEL stands for Fair Queueing with COntrolled DELay
- ADSB is to air traffic what AIS is to marine traffic
- In CAP theorem, must choose between availability and consistency
(For Anki nerds, I have these as cloze deletions)
The biggest benefit for me (aside from actually remembering this stuff) is that the cognitive overhead of inputting and studying is super low. I can stay focused on doing important work while not having cognitive overhead of learning new things affect my output.
Then you’re already close to the Zettelkasten method; you can use individual paper notes and never touch digital.
The difference is in how your individual notes are structured into a coherent system. Instead of filing a note in your cabinet by topic / theme, you give each note a unique ID number and over time add links to others within your note.
It seems as though Zettelkasten, despite being a relatively old phenomenon (e.g. see Niklas Luhmann), is experiencing a renaissance these days. Ultimately, it is just (yet) another way of keeping track of your notes and thoughts. You can think of it as cataloguing everything you write down. I reckon whether you stand to gain anything from converting your existing note-taking to a Zettelkasten operation depends entirely on how you currently take notes. What really matters is that whatever approach you employ works for you. In general, I wager the mere act of actually writing something down and approaching material in a critical, thoughtful manner as opposed to merely consuming is what drives both retention and understanding. Zettelkasten might help you attain that. It might also not.
The difference lies in how it's organized. An encyclopedia such as Wikipedia is organized using an encyclopedic hierarchy of knowledge. That makes no sense for a personal linked notes system: the key is to avoid hierarchy, and tag and link the notes in a way that helps them surface on the contexts you need.
I'd describe it as a certain style of private wiki. It comes with certain conventions how a wiki page should be. The article describes it as atomic, self-contained, permanent, and concise.
I am not disciplined enough to do the initial linking by using these timestamps, will also not break out of the flow to find the previous note in another folder and copy the timestamp around.
Instead I’ve been using a stream of small auto-timestamped logs with hashtags to take and retrieve my notes and ideas. It is not a tree-based system, it’s just a text log stream and has worked extremely well for me so far.
I used to do this in a private slack group but eventually created an app called ZenJournal. You may also find it useful.
Niklas Luhmann collected 90,000 note cards and popularised "Zettelkasten" note-taking as a knowledge management method with paper-based hyperlinks:
https://niklas-luhmann-archiv.de/projekt/beschreibung
(archive, supports download as TIFF scans and JSON transcriptions).
During my undergraduate studies I summarized each paper or book I read in form of a single plain text file. This was very helpful when studying for oral exams that required internalizing in the order of >20 volumes. While these weren't cross-referenced, and no special tool was used, this still helped to ace the exams, mostly because the practice of summarizing/compressing the material to its essence itself is essentially the same as learning it.
Main main advice is any tool or none is fine as long as the notes are in plain text so they can last - it's the most durable file format, and you can always build a full-text index to make it searchable e.g. using Lucene.
I almost never take notes. My to-do lists are on post-it notes scattered on my desk, and I rarely miss a task or deadline. I tried the notebook route for a while, but I found that it was more like a diary; I never looked back until years had passed and, by then, the data was no longer relevant.
I don't get it. Even if I write a bunch of stuff down, I still don't have the time to read it, unless it's for a specific project and I'm summarizing data specifically for that. If it's not tied to a project, I might as well not write it down at all - it won't be read.
> The linking is done by placing links like these [[20200606154308]] into notes. Most of the slip-box software allows you to click on this link and opens a note with ID 20200606154308 for you.
What's the purpose of using a number to link? Why not use words, like hyperlinks have been doing for decades?
Usually you use an ID in place of the title of the note, because you may want to change the title at a later date. If you were to change the title, then you would have to also change any of the links. Therefore, you keep the ID the same and can change the title of the note without breaking any links.
They're unique. Zettelkasten titles can be quite long: "Practices for nurturing healthy brain development during pregnancy" takes up a lot more space for a link than a 14 digit ID. You may say to just use a keyword of the title instead of the whole title, but then it is no longer unique and you must come up with another way of identifying which note it links to.
Also, using IDs means it is very easy to visually scan a note for links. The [[##############]] format stands out vividly.
I like it, it doesn't bother me. Do whatever works for you.
Is there a way to see a title or can you only show the ID in the preview on the left? I am very confused by this software because renaming a file brakes all my links... And if every file as a date as a name I would not know where to start.
That’s kind of similar to the format I usually use for writing for my blog. I’d use the blog itself as the permanent folder, and the act of publishing it the ritual which makes it permanent. As I am writing new essays they live in emacs or notes or notion or whatever note taking app I enjoy using currently. Such notes are almost never long lived for me.
If I were to write something large like a book, I would probably maintain distinctions between early drafts vs later drafts of chapters. I am imagining that most writers do have some system which captures the notion of the stage the work is in.
Publishing externally is the ultimate way to make it permanent though- otherwise nothing stops the author from amending.
I've been using https://github.com/vimoutliner/vimoutliner for a decade now. It works well. It's so easy to backup, since it's just text files. And you can just use grep and other standard tools to look for content. I believe it's the vim-counterpart to emacs org-mode.
I've been forced to use MS OneNote for one contract, since the laptop is locked-down. Works ok, but search-ability and keyboard-only usage sucks in comparison.
I don't see how this method of note taking is better than just writing note linearly. Sure, it is good if you want to recall information for writing a book, but it doesn't seem like it would actually increase your retention and improve your ability to apply things that you have learned. If anything it would make you less efficient at learning because you have to maintain this note structure.
It seems like it's only good for summarizing or collecting information for repackaging and publishing. Maybe that is all it's purporting to be, though.
There's been a few of these articles on HN recently. This method sounds to me like a personal wiki. And for that I use Zim-wiki. It has all the features the article talks about, as well as displaying pictures. It's a folder and text file structure as described, handles links, and back links, between notes and has a tagging system. Not only that it has a journal and task management. I use it at work and home. It's cross platform, but not mobile. Brilliant piece of software.
My blog/wiki has been created somewhat along the same lines for many a year (including YYYY/mm/dd/HHMM page names), but somehow I can't get past Zettlr's default ID scheme, or the lack of hierarchy (having all my notes in a single folder is not something I'm keen on at all).
(As an aside, both Zettlr and Obsidian completely freeze up on my 7000+ blog/wiki folder, but Obsidian eventually manages to parse and cross-link most of it given time)
I'll try to explain how I git where I am with this.
I used the same Google keep as you for multiple years, but Keep isn't available offline on PC, it's interface doesn't help in interlinking, and it is not open source. Last one hurts especially considering its a Google project, one of the decent ones at that, and thus it can always be cancelled next quarter.
I moved in to Emacs+org-mode on PC, and Orgzly/Orgro on Android. It syncs notes via syncthing, on my lan, no internet required, but can be used. This gives it a ruledundancy. Org-mode is also much Kore organic for note taking, since it allows collapsing trees, embedding code, equations. Adding links can be done inline, unlike Keep (I haven't used it in last year so that may have changed).
As for search, Keep is good enough if your notes only have shopping lists and one off stuff that you require every now and then. Keep is, however, terrible for organizing any amount of knowledge. Org-rifle, for eg. gives context aware search results across my entire notes folder. The power is unmatched by Keep, where notes are mostly siloes, and interface too verbose to let you a birds eye view.
And did I mention Keep takes multiple seconds to load on a good connection? And that its closed source? For some, that alone is reason enough.
Finally somebody makes some sense. I have an even lazier system, I type a search into bing like "bizidea - hydroponic urban farming" and save it to bookmarks and when search "bizidea" there are all the billion dollar ideas.
There's a note-management system that can be built that's much better than all existing systems, I am certain of it, but all my attempts to design it so far have failed.
It somehow involves a combination of organically determining a hierarchy of the notes, and then helping to algorithmically prune notes that are no longer useful.
Yes, totally! I feel like this Zettelkasten technique is a step in the right direction with the idea of linking notes, but relying on these long, random looking strings to do the linking seems wrong. Human-readable tags that serve as links seem better, but I suppose that's more of just a software issue, not a technique issue. My ideal note-taking software for what you're describing would be:
1. platform independent (or at least very easy to use on multiple platforms, phone + desktop)
2. your "organically determined hierarchy of notes" idea ... a clustering map of some sort.
3. links via human-readable tags
4. FOSS
I'm guessing this exists already but that I just need to look harder.
The most important idea I got from zettelkasten was not tagging ontologically your notes, but thinking in future actions like #havingatea, #planningahike, etc.
Any idea why would Zettlr (recommended application for Zettelkasten in the article) called one of the themes "Karl-Marx-Stadt" (which is a name used for Chemnitz during division of Germany).
I know it's a bit off topic - so I guess my comment fits right in HN discussions :D
edit: I figured this out myself; the author of the application is described:
Hendrik Erz is a research assistant at the University of Bonn. His main research focusses on the nexus between Marxist economic theory, digitalisation and violence in capitalist societies (riots and terrorism).
If I look for some specific information (e.g. whom I talked to in a given meeting and what the take-aways were) I typically know the rough date range when I wrote that information down, so I just go through the notebook entries until I find it. Same with the lab and creative notebooks.
Another advantage is that writing by hand allows you to draw diagrams and write equations with ease. In my opinion nothing beats a paper notebook in this regard. I tried the new iPads and MS Surface tablets but the note-taking experience is just not the same and the results look disappointing IMHO. Also, if I really need to record an image or computer-generated diagram I simply print it, cut it out and glue it into the notebook.
My former colleagues in science used the same method of keeping lab books for decades with great success, I have never heard of anyone using a system like Zettelkasten (but maybe that's just my bubble).