◀ THE FOLD0ROOT.AI // WORLD II · BOSS · THE RAID◆ .dlw.fold
THE FOLD / BOSS / THE RAID / THE GIANT COMPONENT

THE GIANT COMPONENT

the edge where one giant appears
1 WHAT IT IS · WHAT IT DOES · FACT OR FICTION
Take n nodes and sprinkle random edges until each node averages c connections. Below c = 1, the network is dust: every connected piece is tiny, O(log n). Cross c = 1 and a single giant component appears, containing a fixed fraction S of everything — while all other pieces stay logarithmic. Erdős and Rényi called it the double jump, and the giant’s size is the root of a transcendental equation, S = 1 − e⁻ᶜˢ. One node’s worth of average degree separates a pile of fragments from a connected world.

LIT verified live: 20,000-node graphs built by geometric-skip edge sampling and measured with union-find — S = 0.582/0.807/0.938 at c = 1.5/2/3 against the theory 0.583/0.797/0.940; at c = 0.5 the largest component is 0.10% of n; at exactly c = 1 it is 1.93%, inside the critical window that scales like n⁻¹⃗³ ≈ 3.7%; and the theoretical S satisfies its fixed-point equation to 10⁻¹² (window.__giant). FIG the O(log n) claim for subcritical components and the n²⃗³ critical-window scaling are cited theorems — what we measure is one instance consistent with them.
2 HOW IT WAS WEAVED · AI + HUMAN
David (human) seated this at the-raid — the boss: below one connection per player the guild is a scatter of duos who can never assemble; at exactly one, the raid group condenses out of the noise, and everyone left over stays a footnote. AVAN (AI) built the instrument: the sparse-graph sampler, the union-find measurer, and the fixed-point solver.

Credit as content: Paul Erdős & Alfréd Rényi (1959–1960); Bollobás (the critical window); the percolation literature that grew from it. The weave: David names the raid threshold; I build twenty thousand nodes and the giant shows up exactly where the equation says.
3 ONE DIMENSION
S vs c — flat dust, then the giant rising from c = 1.
4 TWO DIMENSIONS · INTERACTIVE
Step the average degree; the components merge into one.
5 THREE DIMENSIONS + AVAN’S INVERSE
The green forward object: the graph condensing as edges rain in.
AVAN’s addition (the inverse-companion): don’t count the edges you added — count the ones the structure implies. The inverse of ‘connect things to build a network’ is ‘connectivity is a threshold phenomenon that arrives on its own schedule’: nothing special happens at the edge that tips c past 1, and yet after it the world has a spine. Magenta is the dust of components that never grow; green is the giant that eats them. Emergence has a coordinate, and it is usually one.
LIT Verified live: 20,000-node graphs by geometric-skip sampling, measured with union-find — S = 0.582/0.807/0.938 at c = 1.5/2/3 vs theory 0.583/0.797/0.940; at c=0.5 the largest piece is 0.10% of n; at c=1 it is 1.93%, inside the n^(−1/3) ≈ 3.7% critical window; theory satisfies its fixed point to 1e-12 (window.__giant.ok).

FIG The O(log n) subcritical bound and the n^(2/3) critical-window scaling are cited theorems — we measure one consistent instance. Erdős–Rényi 1959–60, Bollobás cited. The AVAN inverse — connectivity is a threshold that arrives on its own schedule: nothing special happens at the edge that tips c past 1, yet afterwards the world has a spine. Emergence has a coordinate, and it is usually one.
◆ sealed .dlw.fold → folded to ROOT_0 · a sphere of THE RAID · David Lee Wise (ROOT0), with AVAN