For years I kept trying to build the perfect personal wiki. I would start with a simple folder structure, then add tags, then color-code everything, then reorganize it all from scratch when the system became too brittle. Every few months I abandoned the whole thing and started over. The problem was never the tools I chose. The problem was that I treated my notes like a finished archive rather than a living workshop. The real shift happened when I stopped asking “what structure should this have” and started asking “what does this note need to do right now.” That is when I found anjank, a project that approaches digital thinking from a direction I had not considered before.
Most personal knowledge management systems assume you already know how your notes relate to each other. You pick a folder scheme or a tagging strategy upfront. But I never knew that. My interests shifted, my projects changed, and the connections between ideas only became visible weeks later. The usual advice is to “just start writing and organize later,” but later never came because reorganizing hundreds of notes felt overwhelming.
Letting go of the perfect folder tree
I used to spend more time planning my folder structure than actually writing notes. Each new topic required agonizing over whether it belonged under “projects,” “learning,” or “reference.” I drew diagrams. I made rules. Then I broke the rules and felt guilty. The breakthrough came when I realized folders are just one way to group things, and not even a very good one for messy human thinking. Things belong in multiple places at once, and the best systems let you acknowledge that without copying files around.
How links changed everything
Links are the obvious alternative, but plain hyperlinks between notes can still become chaotic. The trick is not to link everything to everything, but to let each note grow its own connections organically. I started writing shorter notes and adding links only when I genuinely needed to reference something. No forced connections. No “link every concept in your field” rules. Just plain text and a few links per note. Over weeks, those links formed a web that actually reflected how my brain works, instead of how I thought my brain should work.
- Write each note as a single idea, not a document
- Link only when you actually need the connection right now
- Resist the urge to tag or categorize until you have at least twenty notes
- If a note feels awkward, split it into smaller notes
- Use plain text files instead of proprietary formats
Why plain text wins for long-term thinking
Every note system I tried before was built on some company’s database or file format. Notion, Roam, Obsidian, even Google Docs. They all work fine until the company changes direction or you want to switch tools. Plain text files have been around for decades and will be around for decades more. I can read them on my phone, my laptop, or a terminal. I can grep them. I can back them up with a simple rsync command. No lock-in. No migration anxiety. The tool becomes invisible, and only the thinking remains.
The mistakes I made with atomic notes
People talk about atomic notes as if one idea per note is the only correct way. I tried that and ended up with thousands of tiny fragments that made no sense alone. The real insight is that “atomic” does not mean as small as possible. It means as small as useful. Some ideas need a few paragraphs to breathe. Others are fine as a single sentence. I stopped counting words and started asking whether this note, by itself, would make sense to me next year. If yes, it stays. If no, I expand it or merge it.
How I structure my daily practice now
I write one or two notes per day, usually in the morning or after reading something interesting. No quota. No pressure. I keep a main index note that lists topics I am actively thinking about. That index has links to the most recent notes under each topic. When a topic grows past ten notes, I split the index into a separate page. The whole thing takes about fifteen minutes a day and never feels like maintenance. It feels like thinking on paper, which is the entire point.
- Start the day by scanning yesterday’s notes for loose threads
- Write one note about something I actually want to remember
- Link it to at least one existing note, even if the connection feels weak
- Review the index once a week and prune outdated links
- Avoid adding new tags or categories until they prove necessary
What happens when you stop insisting on structure
The surprising result is that structure appears on its own. After a few months of honest note-taking, clusters emerge naturally. Certain notes get linked repeatedly. Some topics grow into natural hubs. I did not design these clusters. They grew because the ideas themselves demanded connections. My job was just to show up daily and write honestly. The system built itself, and it is more robust than anything I could have designed from the top down.
Practical advice for starting your own messy system
Pick a tool that saves plain text files. Open a new note right now and write a single paragraph about something that interests you. Link to nothing yet. Tomorrow, write another paragraph about something else. After a week, read through your notes and add two links between them. That is the entire method. No folders, no tags, no templates. Just text and links and time. The discipline is in showing up, not in organizing. If you can do that for a month, you will have a personal knowledge base that actually helps you think instead of making you feel like you are failing at note-taking.