An invitation to think
with the evidence.

Where it began

On August 12, 2026, while working on belief networks and stochastic neural networks, I discovered the connection linking epistemology, physics, and probability and statistics together. I could barely sleep for three days following the discovery!

What began with detecting hypocrisy in belief maps and drawing from distributions in Beta networks grew into a surprising mathematical realization: when you refuse to throw away the volume of evidence behind a probability, the geometric apparatus of special relativity emerges naturally from Bayesian statistics.

Start from something small

Begin with a basic evidence notebook recording positive trials $\alpha$ and negative trials $\beta$. A Beta distribution describes our uncertainty about the underlying rate of success. A simple change of coordinates turns these parameters into a point inside a Minkowski light cone:

• Total evidence ($\tau =\alpha +\beta$) acts as forward-ticking epistemic time.
• Directional balance ($x=\alpha -\beta$) acts as spatial position, with normalized velocity $v=x/\tau$ bounded by the speed of light.
• Contradictory evidence creates epistemic rest mass ($m=2\sqrt{\alpha \beta }$), giving established convictions physical resistance to sudden acceleration.

These identities open vivid routes into information geometry, directional probability, and source lineage. They invite us to ask how autonomous intelligences, biological cells, and distributed networks navigate their own uncertainty.

Made for exploration

Start here follows the fundamentals from simple observations to light cones, rapidity, and multidimensional compass states. The field trip articles venture into biology, quantum optics, statistical mechanics, and artificial intelligence.

Read a chapter, test a derivation, move a slider, and see whether the explanation survives your next question.