Entropy to Order: How I Rebuilt My Second Brain in Mordor

For about three years I told myself I was researching. What I was actually doing was bookmarking. The folders multiplied. The PDFs piled up in a directory I had stopped opening. Every few months I’d run a search across the heap, find a highlight from 2022 that was almost-but-not-quite what I needed, and tell myself the system was “working.” It wasn’t. I was running a personal library and confusing the cataloging work with the thinking.

The thing that finally broke me was sitting down to write an essay I had supposedly been preparing to write for six months and realising I had no idea what I thought. I had quotes. I had screenshots. I had a document called essay-outline-v3.md that was just a list of links. The shelves were full and the head was empty.

So I tore it down. I called the new system Mordor — partly because I was deep in a Tolkien re-read, partly because the name fit the mood. The point of Mordor is heat. You bring raw ore in, and what comes out is something you forged yourself, in your own words, that you can hold up in an argument. Anything that can’t survive that process doesn’t deserve to live in the system.

This is the long version of how I got there, and what changed in my head along the way.

The Collector’s Fallacy, named

There’s a phrase Christian Tietze coined that put a name on the disease: the Collector’s Fallacy. The mistake of treating “I saved it” as equivalent to “I learned it.” Once you have the name for the trap, you start seeing yourself fall into it in real time. You see your own hand reaching for the bookmark button as a substitute for thinking. You see the little dopamine hit you get from filing a PDF into the right folder, and you recognise it for what it is — the satisfaction of a clerk, not a thinker.

The clerk’s instinct isn’t useless. Reference material has to live somewhere. The problem starts when collection becomes the whole job, because collection scales infinitely and understanding does not. You can save a thousand articles in an afternoon. You cannot understand a thousand articles in a year.

Mordor was born the day I admitted that out loud.

The conversation partner, not the cabinet

The model I had been running, without realising it, was library. A library is a place where finished things live. You go in, you look something up, you leave. The library doesn’t think about you between visits. It doesn’t surface anything you didn’t ask for. It is, by design, inert.

Niklas Luhmann’s slip-box ran on a different model entirely. He kept ninety thousand index cards in a wooden cabinet and credited the cabinet, in interviews, as a partner in his work. The famous line — “Ich denke ja nicht alles allein. Das geschieht weitgehend in meinem Zettelkasten” — translates roughly to “I don’t think everything alone. Most of it happens in my slip-box.” That isn’t poetic. He meant it literally. The cabinet had enough internal density and enough cross-references that querying it produced surprises. He’d pull a card on one topic and find it pointing at three other cards he’d forgotten he’d written, and the path between them would be the essay.

The shift from library to conversation partner is the whole game. In a library mindset, friction lives in filing — how do I make sure this thing ends up in the right folder so I can find it later. In a conversation-partner mindset, friction lives in connecting — how does this new idea touch the things I already think. The first kind of friction is bureaucratic. The second kind is generative. Most note systems fail because they optimise the first one and ignore the second.

Mordor optimises the second one. Every note I write has to earn its place by linking to at least two existing notes. If I can’t find the links, the note isn’t ready to live yet, and that’s information.

The philosophy of the atom

The unit of thought in Mordor is the atom: one note, one idea, complete in itself. An atomic note is one you can hand to a stranger out of context and they’ll know what you mean. If you can’t do that, the note is doing two jobs and needs to be split.

This sounds prim until you try it. I started splitting old notes and discovered that most of what I’d been writing was actually three or four ideas welded together by the accident of having read them in the same article. Splitting them apart was painful. Several of them turned out to be hollow once isolated — the “idea” was really just a quote with my name underneath, no actual thought attached. Those got deleted. The system got smaller and sharper.

Atomicity is what makes connection possible. Two ideas can link cleanly only if each one has a clean boundary. A note that says “everything I think about distributed systems” cannot link to anything specific, because it isn’t specific. Break it into ten atomic notes — one on consensus tradeoffs, one on the CAP folk theorem, one on how Spanner pretends to violate it, one on why “eventually consistent” is mostly a marketing term — and suddenly each one has edges, and edges are what link to what.

The Germans have a word Luhmann used for this kind of linkage: Verknüpfung. Closer to “interweaving” than to “tagging.” Folders give you containment. Verknüpfung gives you a graph. I gave up topical folders entirely the day I caught myself trying to decide whether a note on cryptographic hash functions belonged under “Math,” “Cryptography,” or “Data Structures.” It belonged in all three, which meant it belonged in none, which meant the whole categorisation game was a tax I was paying for nothing.

Stages of development, not subjects

What replaced the folders is a lifecycle. Notes in Mordor don’t live in topics; they live in states, and they move between states as they mature. The vocabulary I settled on is borrowed loosely from Andy Matuschak and from the gardening metaphor people in the digital-gardening community keep using:

• Inbox. Raw capture. A line, a quote, a half-sentence I caught on a walk. Inbox notes have no obligations. They are allowed to be wrong, vague, or duplicated. The only rule is they don’t stay there. Inbox is processed weekly; if a note hasn’t earned a promotion in two weeks it gets deleted. The garbage collection is non-negotiable, because the inbox is the place where the Collector’s Fallacy tries to come back through the side door.
• Seedlings. Notes that have a thesis but haven’t yet had to defend it. Most seedlings die quietly. The ones that survive do so because they survive contact with the rest of the system — because I tried to link them to existing notes and the linking exposed something I hadn’t seen.
• Growing. Notes I’ve actually rewritten in my own words, with at least two outbound links. This is where most of the real cognitive work happens. Writing a Growing note forces me to take a position, because I can’t paraphrase someone else’s argument without exposing whichever bits I don’t actually understand.
• Evergreen. Notes that have stopped changing. Not because the topic is settled, but because my thinking on this particular atomic claim has stabilised. Evergreens are the load-bearing bricks. When I sit down to write something longer, I assemble it out of evergreens.

The thing this lifecycle gives you that folders never can is honest signal about the state of your own thinking. If I look at my system and see four hundred Inbox notes and twelve Evergreens, that’s a diagnostic. It means I’ve been collecting and not metabolising. The numbers can’t lie to me the way folder counts can.

The reflexive interface

None of the philosophy survives if the tool gets in the way. The single biggest practical thing I learned, the thing nobody told me when I started, is that the cost of capture has to be near zero or the system rots. If catching a thought in flight takes more than four seconds, half the thoughts get away. The graveyard fills up with the ones that escaped.

So Mordor lives in Obsidian, configured as aggressively as I could make it. Vim mode, because every reach for the mouse is a tiny break in the chain. Quick-capture bound to a global hotkey so an Inbox note is one keystroke away from any application. Splits and tabs wired into muscle memory, because half of writing a Growing note is jumping between three other notes and stitching the new one into the conversation between them. The full configuration lives in Obsidian Config for anyone who wants the receipts, but the principle is more important than any specific binding: interrogate every interaction your system requires of you, and delete the ones that aren’t doing work.

The keyboard-driven setup has a side effect I didn’t expect. Working at the speed of thought changes the kind of thoughts you have. When the latency between brain and screen drops below a certain threshold, you stop self-censoring. You write the half-formed thing because there’s no penalty for writing it. That half-formed thing turns out, often enough, to be the actual idea — the formed version was the cliché your brain was trying to retreat into.

What changed in my head

I want to be careful here, because it would be easy to describe Mordor as a productivity system and miss the point entirely. The system has not made me more productive in any measurable way. I do not output more essays. I do not finish books faster. The thing that changed is harder to see from the outside.

What changed is that I now have something to argue with. When I sit down to write, the page isn’t blank. It’s surrounded by months of notes I’ve already had to defend to myself, linked into a small dense web that knows things I’ve forgotten. The work of writing has shifted from generating thought from nothing to editing a conversation that’s been running in the background. Most of what looks like writer’s block, in retrospect, was just trying to do all the thinking and all the writing at the same time, with no scaffolding. Mordor is the scaffolding.

The other change, harder to admit, is that I now know how thin most of my old “knowledge” was. The first six months of running the system were humbling. Topics I would have sworn I understood collapsed into three sentences once I forced myself to write them atomically. Books I’d “read” turned out to have left almost nothing behind. The system is honest in a way the old folders never were, and the honesty is the point.

Three years of bookmarks, in the end, gave me almost nothing. Eighteen months of Mordor has given me a working second brain. The difference between the two is not the amount of information. It’s that one was a graveyard and the other is a forge.

For the technical breakdown of the Obsidian setup, check the Obsidian Config.