The question is not what categories exist. The question is what operations a system must perform to remain viable.
I have been building a knowledge system called Open Brain. It captures thoughts, analyses them, decomposes them into atomic claims, and maps the relationships between them. Over the past month, working with Claude, I populated a database with nearly ten thousand epistemic objects extracted from a book manuscript and hundreds of prior working sessions. Each object is tagged with one of eight ontological types drawn from a framework I call the Clean Well.
The types were designed as a classification scheme. A way to label what kind of epistemic work a given claim is doing. But today, staring at the data and thinking about what it would mean to embed these objects in high-dimensional space, something shifted. The types stopped looking like labels and started looking like an algorithm.
I want to walk through what happened and why I think it matters.
The Eight Types
The Clean Well framework defines eight ontological categories for epistemic objects. Each describes a different kind of work that a claim, observation, or idea performs within a larger body of thought:
EO-BND (Boundary): Drawing a distinction. This from that. Inside from outside. Self from other. The act of saying: here is where one thing ends and another begins.
EO-REF (Reference): Pointing at something. Anchoring to the world. A claim that says: this thing exists, this is what it looks like, here is evidence of it.
EO-CMP (Compression): Collapsing many into one. Pattern recognition crystallised into a rule. The move from a thousand observations to a single organising principle.
EO-REL (Relational): Connecting two things. Saying X relates to Y, this causes that, these are analogous. The bridge-building operation.
EO-EVL (Evaluative): Judging. Is this working? Is this true? Is this good enough? The viability check that closes the feedback loop.
EO-PRC (Process): Sequencing. If this, then that. First do A, then B. The procedural logic that turns understanding into action.
EO-MEM (Retention): Remembering. Storing what happened so it can be compared to what happens next. The substrate of learning.
EO-OPS (Operational): Doing. The motor output. The point where cognition meets the world and changes it.
As a classification scheme, this works fine. You can tag every claim in a corpus and get a useful breakdown. But classification is passive. It describes what something is. It does not explain why these eight and not some other eight. It does not explain why these categories feel complete, or what it would mean if one were missing.
The Shift: From Taxonomy to Algorithm
The shift happened when I asked: what would a minimal viable cognitive system need to do to survive?
Consider a protocell. The simplest thing that could plausibly be called alive. Not even a bacterium yet. Just a lipid membrane enclosing some chemistry that can, under the right conditions, persist and replicate.
What operations must this system perform?
It must sense something about its environment. That is reference. It must distinguish inside from outside, nutrient from toxin. That is boundary. It must connect the presence of a chemical gradient to a change in behaviour. That is relation. It must compress many molecular signals into a single response. That is compression. It must assess whether its current state is viable. That is evaluation. It must execute a sequence of chemical steps in order. That is process. It must act on the world. That is operational. And it must retain something of what happened so the next cycle can build on it. That is memory.
Every single type maps to a function that even the most primitive viable system must perform. And they map in order:
SENSE (REF) → DISTINGUISH (BND) → RELATE (REL) → COMPRESS (CMP)
↓
REMEMBER (MEM) ← ACT (OPS) ← DECIDE (PRC) ← EVALUATE (EVL)
↑ ──────── loop ──────── ↓
That is not a taxonomy. That is a cognitive loop. A reasoning protocol. An algorithm.
The Ratchet
Here is where it gets interesting. The loop runs once and produces outputs: a boundary was drawn, a compression was made, an evaluation was passed, a memory was retained. Those outputs are now objects in the world. And the next time the loop runs, it does not start from scratch. It starts from the outputs of the previous cycle.
The first cycle runs on raw chemistry. The reference objects are molecular concentrations. The boundaries are physical membranes. The compressions are reaction pathways.
The second cycle runs on the outputs of the first. Now the reference objects are not raw concentrations but patterns that previously correlated with concentrations. The boundaries are not just membranes but categories. The compressions are not reaction pathways but rules.
Each pass through the loop produces objects that are slightly more abstract, slightly more compressed, slightly further from raw physics and closer to what we might call meaning. The ontological types do not change. The algorithm does not change. But the substrate it operates on gets recursively richer.
This is the ratchet. And it is the same ratchet at every scale.
A cell does it with chemistry. A nervous system does it with electrochemical signals. A brain does it with representations. A language does it with grammar. A culture does it with institutions. At each level, the eight operations are running. At each level, the outputs of the previous level have become the inputs. The loop is invariant. The data is not.
What Is a Fact?
This framing redefines what a fact is.
A fact is not a different kind of object from a compression. It is a compression that has survived so many passes through the loop, been evaluated and retained so many times, that it has been promoted from active computation to load-bearing infrastructure. The system no longer re-derives it each cycle. It runs on top of it.
Gravity is a compression. It collapses an enormous number of observations (things fall, orbits curve, light bends near mass) into a single organising principle. But it has been through so many evaluation cycles, across so many domains, that no viable system bothers to re-examine it. It has been promoted from EO-CMP to something like axiom. It is still running the same algorithm. It is just running it at such depth that the ratchet has locked.
This means facts exist on a continuum, not in a separate category. A brand-new compression ("I think this pattern connects X and Y") and an ancient fact ("mass attracts mass") are the same type of object at different points in the ratchet. The difference is not ontological. It is biographical.
Implications for Mapping a Thought Space
I am building a three-dimensional landscape of ideas. Each epistemic object gets embedded in high-dimensional vector space and projected down to coordinates on a terrain. The topology of that terrain is supposed to reveal the structure of the thinking.
If the ontological types are merely labels, then the embedding should capture semantic content (what the idea is about) and the types can be painted on as colour. Useful but shallow.
But if the types are an algorithm, then the landscape needs to show something deeper. It needs to show the flow between types. Boundary objects should lead to relational objects. Clusters of reference objects should give birth to compressions. Evaluative objects should feed back into boundaries. The topology should reveal directed movement through the cognitive loop, not eight separate piles.
And if the ratchet is real, the landscape needs a vertical axis. The low ground would be raw observations, first-pass references, simple boundaries. The peaks would be the highest-order compressions: the claims that compress the most prior computation into the fewest words. The ridgelines connecting low to high would trace the phylogeny of abstraction. You would literally see how observations became patterns, patterns became rules, and rules became facts.
A Phylogeny of Cognition
If this is right, then the eight ontological types are not a human invention imposed on data. They are a discovery. They are the minimal set of operations that any viable cognitive system must perform, from a protocell to a civilisation. And the recursive application of this loop, with each cycle's outputs becoming the next cycle's inputs, is the mechanism by which meaning emerges from physics.
That is a strong claim. It is also, in principle, testable. If the types are genuinely fundamental, then a corpus of ideas mapped by this algorithm should reveal structure that a naive semantic embedding misses. The compression objects should show a particular distribution: present at every scale, in every domain, with the highest-confidence compressions sitting at the apex of long chains of prior evaluation. If the corpus is about a theory of compression, and compression is indeed the universal organising principle, then the CMP objects should be the most connected, the most cross-domain, the most deeply ratcheted. If they are not, that is diagnostic. It means either the theory is incomplete or the extraction missed something.
We have 9,463 epistemic objects extracted from a manuscript arguing that compression under energy constraint is the thread connecting genes, brains, language, meaning, and civilisation, plus hundreds of working sessions developing that argument. The ontological types are tagged. The ratchet depth is, in principle, estimable from the content. The embedding infrastructure exists.
The next step is to embed them in a way that respects the algorithm, not just the semantics. To build a landscape where the topology reflects the cognitive loop and the elevation reflects the ratchet. And then to see whether the shape that emerges confirms, refines, or refutes the theory that generated the objects in the first place.
That would be a thought space that thinks about itself.
• • •
This piece captures thinking-in-progress from a working session on the Open Brain project. The framework described here (the Clean Well ontological types as cognitive algorithm, the ratchet model of abstraction, the phylogeny of meaning) is provisional and under active development. It emerged from the collision of building a knowledge system and writing a book about compression as the organising principle of life. The collision is the point.