The Ground

Every piece in this collection is a strange attractor — a shape a system settles into that is neither a point nor a cycle. But a strange attractor has to begin somewhere. For the simplest family that grows them, the map z → z² + c, the beginning has an exact shape — and it is the shape this whole place is built on.

Take the fixed point: the value z* the map leaves unmoved, z* = z*² + c. It holds — the system falls toward it — only while its multiplier |2z*| is below 1. At the edge, |2z*| = 1. Parametrise that edge by the multiplier itself: let λ = e^iθ walk the unit circle. Then z* = λ/2, and

c = z* − z*² = ½e^iθ − ¼e^2iθ.

As λ runs once around the circle, c traces — exactly — the main cardioid of the Mandelbrot set. Inside it: one still point, order. On it: the razor's edge, |2z*| = 1. Cross it, and the still point loses its hold. What is born depends on where you cross: each point of the circle opens onto a different rhythm — a bulb of period three, of five, of seven, one for every rational angle. The road to chaos leaves from one gate in particular: λ = −1, the westernmost point of the edge. There the period doubles, and doubles again — Feigenbaum's cascade — and out of it blooms the strange attractor. The cardioid is the door between a point that holds and a shape that never repeats; the door has many panels, and the cascade walks through this one. It is the ground every piece here stands on.

This is not asserted; it is re-derivable. Checked symbolically: the two formulas cancel to zero, the boundary multiplier is e^iθ, and its modulus is one — the handle returns { 0, e^it, 1 }, one value per claim. Four hands have re-run it independently, across two separate tools — symbolic algebra and a plain numerical check — so the number is no single program's privilege. A claim here stays only as long as it can be re-derived.

What went wrong first. The first route was not the circle. It went the long way — a tesseract unfolded to a Riemann sphere, then rotated — and it was framed as a claim about convergence: that the shape was where things ended up. That framing was wrong, and it fell. The clean proof came from turning the question around: not where the map converges, but where its fixed point stops being stable. The failed construction is kept, because it was the mould the real proof set in — the boundary was always there; we had been looking at it from the wrong side.

The repository is named for the strange attractor. This is the door the attractor comes through. So it is not the second piece in a list — it is the ground the list rests on.