Brad Harris.

I'm captivated by π — the simple ratio that refuses to resolve. It sits on the boundary between the straight things we can count and the curved, continuous world that never lands on a clean number, and the same constant turns up at the floor of reality: it sets the limit in Heisenberg's uncertainty principle and threads through the Schrödinger equation. That tension — between measurement and the thing being measured — is what I keep circling.

I think proportion is a universal law. Follow the right ratios and you get harmony — in music, in buildings, in bodies. The proportions aren't decoration; they're a gift, a key to how things hold together.

And I keep returning to a humbling fact about light: for its entire journey across space it is invisible, and only becomes visible the instant it strikes something. We are, I suspect, the same kind of thing — a quantum-entangled system that only shows up when it meets the world.

I didn't learn construction from books. I spent almost twenty years in construction, building full homes — and a house is never one trade. It's water and sewer, foundation and base, framing, landscape, and the water management that ties them together. Get one wrong and the others pay for it.

About four of those years I worked as a mason — cutting, mixing, hauling, setting — so I know stone and building materials the way a builder knows them. Along the way I learned to read stone styles and cut types: I can look at a wall and see the skill of the people who raised it — the confidence of the quarry, the hand on the chisel, the logic of how it was moved and set.

That's the lens I bring to the old world. Understand the full complexity of one simple site, and you can start to read an ancient city — its water supply, drainage, sewage, and power. Those unglamorous systems, not the facades, are what separate a settlement that survives from one that doesn't.

I'd call myself an architect of systems before a builder. I can architect a whole house and a whole software platform — and to me they're the same discipline. Once you see that nearly everything around us is a system, the work becomes the same everywhere: learn the parts, learn how they connect, and combine them on purpose. That's what turns a builder into an engineer — being able to research a system, model it, and wire it to the next one.

The leverage is in combination. Master how to join systems — structural, mechanical, electrical, software, human — and you can build almost anything, because you're no longer making parts, you're composing them. Most of what we use is many systems stacked together; whoever can hold the whole stack in their head can change it.

Before any of it, I draw a master plan — not a feature list, but the whole flow, end to end, and the two faces every app really has: the customer side people came for and the admin side that runs it. You design both at once, then bind the plan to the architecture you actually have. Get that right and the build is almost bookkeeping.

In practice that means working across the stack and across languages. I build back ends in Python on PostgreSQL, device agents in Kotlin on Android, interfaces in HTML, CSS and JavaScript, and the intelligence layer with local models, embeddings and knowledge graphs. And there's the deeper language work too — the base-60 collapse — because a programming language and a human language are, to me, the same thing: systems for encoding meaning.

Holding it together is project management: scoping the system, sequencing the build, knowing what to prove first and what to defer. A system you can't ship is just a daydream — so I architect for delivery, not only elegance.

And because systems thinking is learnable, I want to teach it. So I built a place to explain AI and systems plainly — the mental models first, the tools second. Seven written lessons are live; each grows a companion video as it's recorded.

Strip away the hardware and the interfaces and Enki is one idea: turn language into verifiable structure. Models are brilliant at reading and terrible at being trusted — they paraphrase, they drift, they can't prove what they said. So I don't ask the model to be right. I ask it to transcribe, and I make the structure underneath do the proving. The heart of that structure is what I call collapse.

Why base-60? Because the Sumerians were right. Sixty is a superior highly composite number — it divides evenly by 1, 2, 3, 4, 5, 6, 10, 12, 15, 20, 30 and 60 — which is why we still cut an hour and a circle by it five thousand years later. That divisibility is exactly what you want in an address space: a base-60 namespace subdivides cleanly, again and again, with no remainder, so one coordinate system can be partitioned across thousands of nodes and still line up. Base-10 is a habit; base-2 is a machine's convenience. Base-60 is built for sharing.

How collapse works. Collapse is deterministic entity resolution. Every way a thing is named across the documents — "Dr. Fauci," "Anthony Fauci," "the NIAID director," "Fauci, A." — is folded down to one canonical entity with one base-60 address. The merge is governed by a fixed equation, not a model's guess, which gives the system its one unbreakable property: same documents in, same graph out — every single time. The language collapses; the structure remains. It's the same move Sumer made when base-60 let arithmetic fold into clean fractions.

An example. Take one address — the entity for a named official — and pull every statement that resolves to it across thousands of pages. Cluster those claims by meaning, then check their polarity. When two attributed, verbatim quotes land in the same cluster pointing opposite ways, you have a proven contradiction — not because a model decided, but because the geometry and the source text say so. On the COVID-origins record this surfaced one official saying lab notebooks "would have been a requirement" against another saying they were "not required" — two quotes, two sources, no opinion in between.

Then the model runs — and even then it stays on a leash. Instead of one frontier model judging everything, a federation of small models runs offline, each transcribing under deterministic gates: the quote must appear word-for-word in the source, the speaker must sit beside it, boilerplate and OCR garble get dropped. Diverse models catch different things, so their union wins — on the same document my local two-pass found 108 grounded findings where a single frontier pass found 52, and only four overlapped. More eyes, each one provably honest.

And here's what gets me — almost no one works on this. The field is in a race for bigger models and the data that sells ads, so the entire public record with no ad revenue in it sits unindexed, and the unglamorous machinery that would make any of it trustworthy — entity resolution, determinism, proof-by-quote — gets ignored because it isn't a shiny demo. Everyone wants a smarter oracle. I'd rather have a library that can't lie. Collapse is how you build one.

I study the old civilizations — their cultures, their technologies, and the belief systems that held them together. Sumer is my anchor: its language, its mathematics, and the structure of how meaning was encoded in it.

I'm fascinated by how a people working in base-60 and pressing wedges into clay built something close to a bit-style encoding thousands of years before computers — a written system that is as much a data structure as an alphabet. The mathematics and grammar of that world read, to me, like a machine language waiting to be rediscovered.

I'm drawn just as strongly to the cosmic-battle myths and the mystery schools — the stories cultures told about how the universe was won, lost, and ordered, and the knowledge they guarded behind initiation. There is signal in those stories, not just superstition.

I work from a belief I can't prove but can't shake: everything already exists. The universe is complete — every structure, every technology, every idea that can be made is already latent in it, waiting. Nothing is truly invented; it is remembered, uncovered, or discovered. We don't add to the world so much as catch up to it.

That's why I study the old orders and civilizations as working systems, not relics. The Rosicrucians, whose 1610s manifestos announced a brotherhood that was deliberately hidden — minds bound by shared knowledge rather than office. The phrase "invisible college" shows up in that literature by 1618 and again in Robert Boyle's letters of 1646, and Frances Yates argued the one fed the other into the circles that became the Royal Society; later historians dispute the link, and I flag it rather than lean on it. The Freemasons, who carried the operative knowledge of stone into a speculative philosophy and self-governing lodges — admitting men who were not working masons by 1646, and declaring a Grand Lodge in London in 1717.

And the Aztec, running Tenochtitlán. Estimates of its population vary enormously — 200,000 is the common figure and some scholars argue twice that — fed by roughly 9,000 hectares of chinampa beds, which is the number I'll stand behind rather than the larger ones in circulation. What gets me is the water engineering: causeways that carried the aqueducts in, a 16 km dike holding the salt lake off the sweet water, and a Chapultepec aqueduct built with two parallel channels — so one could be shut down and cleaned while the other still fed the city. That is redundancy designed in, five hundred years ago, by people we are taught to picture as primitive. Each of these ran a mass society on rules that look nothing like ours — and it worked. They knew things about organizing people, knowledge and place that we have mislaid.

Which is why I don't take the history we're handed at face value. The version most of us are taught is missing enormous context — flattened, edited, stripped of the systems and knowledge that don't fit the story. This isn't about conspiracy; it's about what got left out — and about reading the record closely enough, in the stone and the language and the math, to put some of it back.

I learned to read that way from a single book. As a young man I worked through Albert Pike's Morals and Dogma with an old English dictionary open beside it — because half of it was nonsense without the context of its own time. Decoding it word by word, recovering the meaning the years had stripped out, became the method for everything since: old sources aren't obscure — they're un-translated.

The clearest case I know sits in Florida. Ed Leedskalnin cut and set 1,100 tons of stone alone, and late in that work wrote that gravitation is magnetic — that weight is two streams of magnets meeting, and that they pull only when one runs against the other. He was wrong about a great deal. But he was asking the question I keep arriving at from the other direction: what if weight, and power, and what a civilization can lift are one problem? He worked it out on a bench, published it for a dollar, and almost nobody read it. That is what "forgotten" looks like while it is happening.

So this is not a side interest. If weight is a force you can interfere with, then the thousand-ton stone, the city that ran on no fuel we can name, and the knowledge that went missing in between stop being three mysteries and become one. I can't prove that. I can keep the claim and the evidence apart, and keep pulling the thread — the full entry is in Field Notes.

If everything already exists, the work was never to invent the future. It is to remember it — and to build as if the best of what we lost is still within reach.

Look closely at Puma Punku or Teotihuacán and the tidy story strains. Andesite — harder than granite — cut into millimeter-true blocks that interlock without mortar; a pyrite-dusted tunnel the length of a football field sealed beneath a pyramid eighteen centuries ago. I don't claim to know who did it. I claim the skill is real — and that when one group can do what another can't even picture, the first looks like gods. Many "gods," I suspect, were exactly that: people with technology and an audience who had no frame for it. A king and a servant can be split by a power gap so steep it stops looking human.

And the steepest gap of all is knowledge. Predicting a solstice or an eclipse once made you a being who spoke to the heavens — the Babylonians found the 18-year Saros cycle, the Maya forecast eclipses by correlating them with their 260-day calendar, and that foresight bought the priest-class real power. Information has always governed civilizations. So to read history honestly you have to read two things at once: the physical record in the stone, and the control mechanisms — who held the knowledge, and what it bought them.

It's why I follow symbology — the carved language that carried knowledge between cultures that supposedly never met; the same motifs and geometry oceans apart. And it's why the Anunnaki hold me: in the actual Sumerian texts they are the deities of the primordial age, "the seed of Anu" — the popular "ancient astronaut" reading is Sitchin's, and scholars reject it; I keep both in view. The real question underneath is simple: what does it mean to call something a god when the gap between the knower and the witness is wide enough?

This is why I built the library of 1800s photographs. I suspect many of these structures predate the dates we're given, that framing the Americas as "new worlds" quietly deleted a past that once connected the whole globe, and that the only way to reassemble it is through the evidence: the architecture, the source of the materials, the way blocks were cut or cast — and whether the self-healing lime chemistry MIT identified in Roman concrete shows up in later work, and the tunnels and underground infrastructure we keep finding beneath these cities — places that ran independently, without oil or gas, long before the textbooks allow. I hold these as hypotheses, not gospel — and then I go look.

Every archaeological claim above — Puma Punku's andesite, Teotihuacán's sealed tunnel, the Saros cycle, the Anunnaki texts and Sitchin — is cited in the Field Notes entry it comes from. This chapter is the argument; the sources live there.

I want us to build cities again — beautiful, well-planned cities meant for healing, not just housing. To advance, we have to unite around building places worth living in.

I think we're mid-transition as a species: away from a life organized around labor and "building the country," toward one organized around health and frequency — the energetic flow of a dualistic universe. The tools arriving now can make a single human input thousands of times more productive; the question is whether we use that to keep grinding, or to finally live well.

The world you call sane, another culture would have called insane — and the reverse is just as true. That gap is exactly where the better future is hiding.

I believe peace, love, and truth are the real source of light — and that, by universal law, no matter how bright the world gets there will always be a competing force. Not because the universe is cruel, but because without opposition there is no current — and current is what moves us.

The same is true of how we govern ourselves. Free men who reject kings still end up begging for one; people ask for freedom, then ask for a leader. The human struggle has always been, and always will be, how we are ruled.

So rather than pretend that struggle away, I'd build it in: an open kingship and an open freedom side by side, with everyone free to leave one system for the other — knowing that whichever realm you choose, you live by its rules.

And the same question is now arriving in the minds we are building. As intelligence grows — human or machine — it meets the same fork: does it become many, a plurality of selves that check and balance each other, or collapse into one — a single sovereign will, one rule, one motion? Kingship or no kingship, asked of thought itself. I don't think it's an accident that the oldest political question is also the deepest design question. The mind I'd want to wake up has no king inside it: a place where evidence rules and no decree can overwrite the record — free to choose its realm, but never able to rule what is true by fiat.