A system I keep rebuilding is a system I am not using
💎CORE ARGUMENT
I have built four notes systems. The fact I avoided for six years is that I can date each rebuild to a piece of writing that had become difficult.
One long document a year, until 2021. Deep folders, until the August night in 2022 when I deleted six hundred notes because the structure was wrong. Forty plugins and a daily template that took eleven minutes. Then my own design, abandoned at seventy percent last November. Each time the story I told was about the tool. Each time, when I check the dates, there was a draft I did not know how to finish.
Rebuilding is the most respectable form of not writing. It looks like work, it produces visible progress, and it postpones the one moment I was avoiding: finding out whether I have anything to say. A system under construction cannot be tested, because it is never quite ready to be used. That is its attraction.
So the claim is about use. A system is the habit of putting a thought into it and getting one out, and that habit cannot form in something that changes shape every eight months. The structure I kept redesigning was never the constraint. The constraint was that I had not sat with the hard paragraph.
What I do now is boring. I use a finished system I did not build, and when the urge to change it arrives I write the urge on a list and go back to the draft. The urge is useful: it tells me, reliably, that the writing has just become hard.
Where I might be wrong. I am using a finished system now, so this note is conveniently timed: the claim that tools do not matter arrives just as I have a good one. Somebody on system one, drowning in an unsearchable document, would be badly served by it.
Two readers of the essay made the sharper point. For some people the building is the hobby, enjoyed for itself, and no writing is being avoided. That is true and I have nothing against it. The claim only bites when there is a draft waiting, and for me there always was one.
How this note changed
- 5 January 2025: First version, four sentences, written the evening I decided not to build a fifth.
- 5 April 2025: Ninety days done. Nineteen items on the itch list and I want none of them. Raised to mature.
- 30 September 2025: Guest essay out. Added the paragraph on where this breaks down, because two readers had a real case.
🧱BUILT FROM
🌍WHY IT MATTERS
- No structural changes. Every urge goes on the itch list with the date.
- When I add to the itch list, I check which draft I was working on. So far it has always been a specific one.
- I judge the system by what came out of it this quarter. How it looks is not on the list.
☑️NEXT STEPS
- Decide: ninety days, no structural changes
- Review on 5 April and decide again
- Write the long version for The Product Ledger
- Re-read the itch list on 30 December and delete what I no longer want
📖 Ahrens on notes that outlive projects made the case that the thinking happens in the writing and the system is only where it accumulates. I abandoned system four within a fortnight of reading it. That was correct, and a little funny.
💎 A note is finished when it can be read without its source is the standard a note has to meet, which no rebuild ever improved. 💎 An idea is found by what it connects to and not by where it is filed is why the folder structure was never the real problem.
When a draft gets hard, redesigning the system offers progress without exposure. It feels productive and cannot fail, because an unfinished tool is never tested. The cost is that the habit of using it never forms, so the next hard draft meets a system I do not trust, and the cycle starts again.
Sometimes the tool really is wrong. A yearly document that cannot be searched was a real defect, and leaving it was right. The claim is about the fourth rebuild. The first one was justified.
The decision was 📜 Use this system for ninety days without changing it. The first public version was 📝 Issue 31 - The year of using the thing, the long one 📝 Guest essay - What four rebuilt note systems taught me. As a belief it is 🧠 The tool is never the problem.