Org mode is very dear in this space (to us Emacs users anyway). I'm an iPhone user and was missing quick access to my content (view/add/edit) while on the go. Org is a rich markup, but I've been slowly chipping at my own iOS app with https://plainorg.com.
Compared to markdown, our non-Emacs org options are few but here are others I know about:
Worth mentioning here that LogSeq can work with markdown or org-mode files. In my personal setup one of my LogSeq nodes points to my BrainTool.org file which holds all my bookmarks, web-resource pointers and associated notes. This way I can bounce between working in Chrome, emacs and LogSeq all referencing the same underlying data model. (With Orgzly for a read-only view on my phone.)
See also org-roam (https://www.orgroam.com/) - an emacs-native bidirectional linking PKM solution.
Capturing notes in a decent UI seems to be a solved problem and there are a lot of tools out there for the job. Searching through a lot of notes also seems to be a solved problem.
However, the structured organization of a large collection of notes (think thousands to tens of thousands of notecards in boxes..) as well as the finding connections between seemingly unrelated notes doesn't seem to be a solved problem. Does anyone know of any solutions in this space? Zettelkasten (per my understanding) seems to make this manual on part of the note-taker, but (1) this is super time consuming and (2) very hard to keep track of all the note cards as the collection grows over time.
Cool! I really think this could be huge. Finding connections between ideas is basically the genesis of innovation. Would be very interested to see/learn about your approach.
I think part of the point of Zettelkasten is that by manually making the links yourself, you are forming the associations in your brain, which is part of the learning, thinking, and remembering process.
I am just now learning about it from the book How to Take Smart Notes, so I could be off, but that's part of the idea. With that said, I think what you suggest could possibly be a cool supplement or replacement.
I maintain an offensively big knowledge graph, and I feel like you're both right -- building those connections manually is both mind-expanding and somewhat labor intensive.
If I have a conversation with X and discover they're an expert about Y, learning fact Z, I like to record Z as a file, and link it to both X and Y. That forces me to keep X and Y as entities in the graph. Which, if I care about them, is worth doing.
The real beauty of making those connections is the act of classifying something can cause you to explicitly recognize what concepts it is related to, and formulate precisely what the connection is.
The real pain point is knowing what your future self will wish your past self kept a file on. Some ideas (generally more social science-related ones, but also something like a bug involving the interaction of a lot of disparate technologies) are so cross-cutting that I'd need to make five parent concepts to file them "completely", and often then I'd like to file those parent concepts further. I generally don't.
I used to use Semantic Synchrony, for which the creation and linking processes are much faster. Every little keypress counts. My SmSn graph is as a result much more connected than my org-roam graph. I still refer to it, but try not to build it out much, because the Git review process is so time consuming. For me org-roam's killer feature is that version control is so easy.
You could fall back to a publicly available knowledge graph like Wikidata for your general-purpose "meta" concepts. Then Wikidata itself takes care of linkages at the "meta" level, while your own private information can of course still be managed locally.
It's an interesting idea. Maybe I should try. I suspect, though, that I want control over every node, no matter how general. The thing about my notes is they contain only information I think I might need. I have, for instance, an "emotions" node. I'm sure the Wikidata node for emotions is interesting, but I want my own node too.
I tried to square this circle once by writing Hode[1], which permits encoding certain data as relevant to me, and then filter my view accordingly when I wanted. But the encoding process (i.e. the user experience when adding data to one's knowledge base) was so hard that I gave up.
Zettlekasten is more about making connections between ideas and keeping your ideas organized. The primary use is for generating ideas and writing, not for information retrieval.
The flow is:
1. Read something and take notes
2. Review and distill your notes into key ideas
3. Create cards in your zettelkasten linking these ideas with other ideas already in your zettelkasten
When you are writing and get stuck, refer to the topic in your zettelkasten and follow the flow of ideas.
I use org-mode, just FYI. I was wondering however: everyone shares how they store things, but how often or even how do you query your knowledge base?
There are often tips for creating habit to write down things „at the end of the day“ or such. But can anyone describe some habits when to consume this stuff? I‘d be interested. Thx.
I'll bite. My answer to the "Million dollar question": How to make useful notes?
- Notes are not "write-only": progressively summarize and tree-shake your notes each time you iterate them. You'll leverage your excitement instead of forcing discipline.
- Ideally, notes are organized by project, not by category. It can be a catalyst for action and reviews.
- Only store things that surprise you, not stuff you already know.
Some things you "know" today may surprise you "tomorrow". I have a massive collection of notes. I am constantly surprised by what I used to know and how much I have forgotten. Such is life I guess.
You need to pick a directory/file structure that works for you. Some people (including myself in the past) use a massive single append only markdown file, so querying is as easy as CMD + F.
I realized I was almost never editing/updating old notes within the massive append-only file, so now I have a at-most-one-folder-deep directory with somewhat descriptive file names. I still have some append-only type files, mainly in the form of logs. If I can't think of a name for the file, I just name it with today's date and try to put it into the "best" folder. Sometimes it's ambiguous where the best place is, so I usually put these files in the root directory until they make sense to put somewhere else.
I've also been a plain emacs then org-mode note taker for a number of years. I keep a log.org file with dated items containing notes from meetings and major project actions of the day. I generally have my laptop open during meetings making notes in realtime. After a meeting, or whenever, I'll tag things and note any follow ups, TODOs or things to remember.
I have a terrible memory for proper nouns but I can generally remember enough context so that a quick backward search in my notes buffer can get me back to the name of the person, team, customer, project, action item etc that I'm looking for. (EG it probably has this tag, or it came up during my meeting with Riley, or last Monday, or during the sprint planning meeting etc). I'm now fast enough at this that I can often start talking and do a search in time to mention the 'thing' by the time I need to refer to it.
I see this model as somewhat different than the more reflective, bi-directional link driven, PKM tools being discussed here. Its a great feeder into such a system (IMO) where more active curation can take place. Plus even if there's never any curation I still have all my todo's, calendaring etc accessible to org's agenda features!
Thanks for the answers. I think I realize that there is a difference between a todo list (have to do it) and a knowledge base (let’s organize what I’ve learned).
The former I also do with org-mode and then use the archiving feature for things I put to DONE.
I oftentimes do the latter, but don’t actually use it that often. In college, I used to summarize text-books, which helped me memorize my learnings. After finishing the book, by chapter summary served as a shorter version of the book. Was quite effective.
Haven’t been successful in replicating this for work, though.
I'm using PouchDB to query the small stuff that comes at me during the day like tasks, links, reminders, etc. One simple html file plus a little Javascript (even I can write that) is all it takes. I've tried more elaborate approaches, but they never lasted.
Likewise. I've tried and failed to pick up Zettelkasten, mostly because I can't really grok the idea of atomic notes and what they're supposed to look like. I'd love to see someone else's.
I've read "Taking Smart Notes", and Zettelkasten seems like an amazing solution for a problem I don't have. It seems to be geared around writing longer works out of notes you collect, and if I were doing that, I think I'd be all-in on Zettelkasten. But I'm not. A wiki-like setup is far more valuable to me.
I have a daily note with "Work" and "Personal" sections. As I go through the day, I add entries in those sections:
- Worked on the [[Foo project]] with [[Joe]]
- Called [[Mom]] and asked her about the [[Meatloaf recipe]]
- Helped [[Jane]] with the [[Qux problem]]
This morning it's very easy to see what I was doing yesterday, and have the context for all the mischief I was up to. Also, when I open "Foo project" and look at its backlinks, I get a timeline of when I'd worked on it, and can see what other things I was working on at the same time. Being able to instantly recreate that context is incredibly valuable, and it's very lightweight for me. I'm not going out of my way to associate "Foo project" with "Joe", but still end up with a graph view that has a lot of notes linking to both of those notes.
This reminds me of my personal Wiki that I use to keep track of information at work (I also use Obsidian for this). I start a daily note in the morning and write everything in there until the end of the day, where I have some time set aside to chopping up the information into the appropriate pages. It took a bit for this system to pay off, but when I was writing employee reviews and co-worker feedback, it was almost effortless. I have started to accumulate a critical mass of information that is giving me little hints of insights I hope to be able to leverage this year.
I am still working on actually implementing a personal ZK system in Obsidian or on paper (I am looking forward to Scott Scheper's book about Antinets). I manage people and write software for a living, but I don't really write articles or papers (which ZK seems a good fit for). I just finished "How to Take Smart Notes" by Sönke Ahrens, and it has inspired me to take another pass at starting a ZK for the benefits of greater understanding and insight about the variety of topics I read about.
> until the end of the day, where I have some time set aside to chopping up the information into the appropriate pages
Your way of working sounds wonderful, but I find that I would probably struggle with this part. How are you able to define the end of your day and ensure you have such time marked off? I have tried this but always end up getting sucked into endless other issues/problems/etc and basically passing out at some point, only to wake up the next day to dive right in -- no time to be able to do that sort of retrospective note-munging...
(this is not exclusive to trying to find retrospective time; it applies to things like finding time for food/dinner, blocking out vacation time, or other things, too. work never ends!!)
Not OP, but I feel compelled to answer. I have lots of my own thoughts and feelings on the topic. I'll leave it at two framing thoughts:
If you literally can't find time to eat, how will you even be able to function to execute work, let alone anything higher level than that? I know you probably didn't mean literally no time to eat, but my broad point is how can you work hard indefinitely without caring about your physical self reliably?
If your workload will never end, what exactly are you racing against by working as hard as you can every day? Even if you 100% everything for one day, you're guaranteed to have just as much or more the next day, right? What are you gaining by battling every day until you give out?
Power to the OP, but does anyone else find this knowledge mapping exercise mildly depressing, especially as it pertains to notes about friendships? I can’t quite find the words to describe how I feel about it, but I don’t particularly like the idea of reducing all my thoughts and emotions into text and links between texts. Maybe forcing text makes me aware of details more consciously, but I feel like it might reduce the rawness of life, reducing the resolution to fit into text space. When in reality, maybe the raw emotions are more human? Again, not really sure how to describe my feelings here. Am I alone with this view?
People have been diarizing for as long as there's been paper, so I assume that putting events, interactions, feelings, etc on paper is a pretty common human impulse, but perhaps not for everyone.
One thing I've personally found about maintaining relationships is how much the fact that someone bothered to write something down about me is meaningful. There's a lot to be said for the clerical role in friendships. Noting someone's birthday, the things they like, their family life, etc. so you can write a card or just drop them a note on their birthday is quite the opposite of "reducing" someone to text in my opinion.
This is just an additional tool. Use it if it works for you. For years I used to dump my interesting product/research ideas in individual files - now I moved them in obsidian. Even before then I used to put them in a computation notebook with light green graph paper pages. Writing, sketching, drawing pictures on paper is certainly more creative and freeing (for me) but searching through old notes is harder. Now with Apple pencil (or equivalent) you can get some features of both worlds in one place, with OCR and such. There is a recent free app called collanote that seems quite promising. Still, I’d prefer a cross platform app.
Are you though? You can just view the text as memories.
Same as taking pictures on vacations: "reducing vacations to pictures" sounds like a pretty bad vacation/more like photography work. "Taking pictures of cherished moments and places" sounds like something you might look at again from time to time to fondly remember.
A little of topic but I tried Obsidian out even though it isn't "free" software. I could still use it without the paid features. Then one day I went looking for the source and discovered it wasn't open source. That's when I stopped using it.
There was discussion on open sourcing it but I'm not holding my breath. Does anyone have any additional info on this?
So Logseq [1] is open (AGPLv3), also markdown with no lock in, more powerful in some key ways than Obsidian (it has block links, which I’ll let others expound on the benefit of or edit this when I’m at my desk).
The main thing it’s lacking for now is good mobile support, for which folks still use Obsidian mobile…
It's lacking good support for mobile, but the mobile app is usable, and in active development. I use it daily with no issues. It mostly just lacks a no-frills setup for syncing notes on multiple devices, but they are developing a subscription for backup/sync hosted by them (I use syncthing, but others have used iCloud and google drive).
To build on the features mentioned there's also:
- A community marketplace for plugins and themes
- Simple customization and extensibility
- A builtin PDF viewer with highlighting and linking
FWIW - I'm a huge FOSS guy, but I went ahead and paid for Obsidian because it looked so good and it at least feels as if "saving in pure Markdown" is darn near equivalent, or at least a reasonable non-free alternative experiment; i.e. I'm not sure if this is as good as FOSS, but it's interesting enough for me to throw a few dollars at to see what happens.
In theory, seems like a fine idea. But for, me, in practice, I went back to https://zim-wiki.org (like I pretty much always do.) Not exactly sure what that says about it all, or if it much matters.
One thought though is that Obsidian feels "overloaded," almost as if it wasn't allowed to grow naturally like Zim did?
I have used zim in the past, and I didn't like zettelkasten because of the autogenerated magic.
I also just learned about obsidian for the first time in this thread and it looks pretty impressive, but I do agree that the onboarding is a bit hard and all the plugins while impressive are also a bit overwhelming.
But it's interesting that the core plugins including things like slides which is definitely something that is great to have
The author also has a site called "Grok TiddlyWiki" that helped me tremendously when starting out with TW: https://groktiddlywiki.com/read/
I use TiddlyWiki and love it. Because TW's UI and functionality is based on individual "tiddlers" within the wiki (just like your notes), adding functionality becomes part of the wiki itself -- a fusion I like compared to other plain-text note tools that keep the note-taking and note-storing separate.
Can definitely recommend Tiddlywiki. Ive just spent several weeks implementing a mini ZettelKasten using Tiddlywiki. It is created in 'exploded' mode in NodeJS and then a script runs to re-assemble it into a single html file (with the images in a separate folder) and then I push it to GitHub where it is published on to a personal domain. It is visible here: https://chloetiddlykasten.chloegilbert.me
I used the template (TZK) created by Soren Bjornstad to build it.
That bothers me, too. On the other hand, I haven’t found a good Free replacement for it, and there’s zero lock-in: its database is just a folder of Markdown files. If it stopped working tomorrow, I’d still have all of my content in a convenient format, ready to start using with another app.
There's also obsidian-export[1] that converts the few things that are not plain markdown (`[[links]` and `![[links]]`).
To be honest, obsidian not being open source doesn't bother me too much, what is important to me is that I own the data and that it is in a fairly common format so i can move it to another software in 10 years. An open source software with a weird binary format to save the data will probably do more harm in the long term
Same for me. I wrote a little Python script to copy my Obsidian files into a set of Hugo files, then publish them. Proprietary or not, it's so easy to work with the files that things like this are possible. It's my data in my favorite format, ready for use with all the existing text manipulation goodies I already know. There's not much to improve on there.
Plug: I wrote a couple of other scripts and put them on GitHub[1]. If you use Drafts and its "Quick Journaling" action group[2], then `process-notes` will look awfully familiar.
That's the first thing I check for any system that would, if adopted, become a life-long thing. Zettelkasten is supposed to be exactly that: an ever-growing collection of linked notes, growing more useful the longer you stick with it.
I never understood why anyone would accept linking this sort of system to any third party, no matter how sound their business model, no matter how much trust you may put into the individuals running the company.
Slightly different take: OSS is the second thing I check.
The first one is data export. If it's a closed component that still generates value and I can capture that value via export, OSS becomes a tie-breaker between otherwise equivalent tools.
Obsidian is an editor on top of a bunch of markdown files. If they go away, meh. The world is not running out of Markdown text editors. Sync is an add-on feature, so I can sidestep potential privacy concerns there.
A working system always beats waiting for a perfect system.
I didn’t like moving to Obsidian from Joplin for this reason, but Obsidian is so much more useable because it stores everything in a regular file hierarchy. It also has a better UI. And above all, I absolutely hate Joplin’s “every note is a folder” approach.
The result is the same. Every sub-notebook is a notebook that can contain more sub-notebooks, but there is no such thing as a mere folder that can be placed inside another folder without also creating a confusing space where notes can hide inside the notebook but not alongside its sub-notebooks.
The other way to interpret it is just a very poorly planned UI that uses separate panes to list folders and notes that are located in the same hierarchy position. Either way, it’s madness.
This is a really good point, I am using Obsidian but I feel OK about the proprietary-ness of it because at the end of the day, it's a nice tool to read my well-organised folder of markup files with. It's possible I live to regret this ...
Been using this for a few months now, as I got fed up with emacs-specific configuration when using org-mode. There's also a plugin to integrate it with Anki, all thanks to solid package base of VS.
Before I used Obsidian, I used VS code and a folder of markdown notes. In general, if Obsidian went evil, I could take my existing note setup and go back to that, much as I started with it when I used Obsidian. I'd lose some of the nice todo list setup I have but I know I could write a custom program to replicate that and if it bothered me enough I could even figure out how to write VS code plugins.
So while I'd prefer an open source solution, I'm not in a huge rush, though I am keeping an eye on logseq's evolution.
Logseq is somewhat different from Obsidian, since it's an outliner. IMO Dendron would be a better replacement for Obsidian. Particularly so if you're already using VS Code.
There are other downsides with Obsidian as well. The thought development Luhmann experienced is watered down in the instantiation of Zettelkasten that obsidian provides. If you’re interested in an alternative approach, I’ve been using the Zettelkasten system the way Luhmann did for the past year. Here’s a behind the scenes look at writing a book using a physical Zettelkasten: https://youtu.be/fRgIX4azYOs
It’s not analog (writing by hand and doing things the hard way pays off. It did for Luhmann and continues to do so for many others who use notecards). Second, numeric-alpha addresses per card with strict character count restrictions. Third, tree-like structure for branching thoughts infinitely. Fourth, an index that forces one to neuro-imprint keyterms in their mind to act as cues for thought. All of these create a unique structure Luhmann communicated with. Obsidian = bubbles that connect markdown files. Completely different concept. It’s not a zettelkasten even though people have hijacked the term and ensconced it with digital notetaking apps. This is my opinion.
Obsidian is just a tool, like notecards are; you set your own process and organization. With the exception of handwriting in place of typing, it's entirely possible to set up an Obsidian workflow with all of the properties you list here.
My Obsidian vault, for example, is organized into 2 parts: Ad-hoc notes and an organized zettelkasten. This latter part is a single flat directory with Luhmann-style alphanumeric hierarchical addresses as file names. My first task every day is to split up the previous day's ad-hoc notes and file them in the zettelkasten, which naturally leads to a review of the existing material as I look for the proper place in the sequence.
Do you think it is possible to recreate a digital version of this by basically handcoding your own tool for thought organisation, but coding it yourself needs to be part of it? Like writing a cheatsheet being the main point, and you might not even look at it in the exam?
[0] Foam is a personal knowledge management and sharing system inspired by Roam Research, built on Visual Studio Code and GitHub.
https://github.com/foambubble/foam
It might not be a direct alternative, since it's not using markdown files directly, but I really like Trilium notes (https://github.com/zadam/trilium).
It's open source, you can selfhost it and you can write your own scripts in JS to really make it your own.
I will always run into these threads to state that: these tools will not help you until you yourself have organized your thoughts! You can't have messy idea of a subject topic or some category of ideas and "just" use Zettelkasten or some other method or tool to unfuck it.
And now when that's done, I just use a set of markdown files and some kind of Zettelkasten indexing system.
e. also this graph understanding of ZK is misunderstood way to do ZK.
They're just a better form of writing, and writing can help you organize your thoughts. Sometimes the line between writing and meditation can even blur.
Thanks for the article! It's interesting to see how your notes cluster and what you can glean from them. I have a single file, people.md, that I keep important notes about 1-on-1's I've had with others in.
Also: I'm a big fan of customized personal note-taking using open standards. I didn't intend to do this, but somehow over time I ended up rolling my own note-taking approach and "software" (if you can call a shell script that) which is too lightweight to have any kind of visualization piece. If you're interested in reading about that journey, it's found at https://dev.to/scottshipp/an-amazing-note-taking-system-with...
This is cool. I'm looking to level up my Obsidian usage and I really like the author's thoughtfulness and optimism in their daily notes.
My favorite thing so far has been using Daily Notes to track a morning score and an evening score in the markdown's frontmatter. The score is just a "how I'm feeling" and can encompass really any aspect of my wellbeing. I'm working on a project to parse the data out of my daily notes so I can try to visualize trends. Eventually I want to mash in more data (listening history, Strava activity level, steps from Garmin, etc.) to see what kinds of things have, or seem to have, effects on my well-being.
How useful to people find these "cluster graphs" to be, which are so popular in zettelkasten/pkms systems? It seems to me it would, instead, be more helpful to just have a table of the most connected nodes, sorted by # of connections.
There’s no such thing as “pretty diagrams”, lornajane. I believe you simply has boarder interests in life which inherently make inter-links less realistic.
Compared to markdown, our non-Emacs org options are few but here are others I know about:
https://beorg.app
https://braintool.org
https://flathabits.com (I author this one too)
https://organice.200ok.ch
https://orgro.org
http://orgzly.com
Karl Voit also has a great effort going to advocate org outside of Emacs and documents some of the tools at https://gitlab.com/publicvoit/orgdown/-/blob/master/doc/Tool...
edit: formatting