THE FOLD / GLITCH / THE BLUE SCREEN / THE BORWEIN
THE BORWEIN
a run of integrals that equal pi-over-two until they suddenly do not
1 WHAT IT IS · WHAT IT DOES · FACT OR FICTION
The Borwein integrals are the most famous ‘pattern that breaks’ in mathematics. Using the sinc function sinc(x) = sin(x)/x, the integral ∫0∞ sinc(x) dx = π/2. Add a factor: ∫ sinc(x)·sinc(x/3) dx = π/2. Keep going — sinc(x/5), sinc(x/7), … up to sinc(x/13) — and every single one is exactly π/2. Then you include sinc(x/15) and the answer drops to π/2 minus a whisper (about 2×10-11). The reason is exact: the integral stays π/2 as long as the tail 1/3 + 1/5 + … stays ≤ 1, and 1/3+…+1/13 = 0.9551 < 1 while adding 1/15 tips it to 1.0218 > 1.
LIT verified live: the reciprocal sum 1/3+…+1/13 is confirmed < 1 while +1/15 exceeds 1 (the exact mechanism), and the integrals through sinc(x/7) and sinc(x/13) are numerically π/2 (window.__borwein). FIG the tiny deficit at the 1/15 step (~2e-11) is below crude numerical resolution — so it is the reciprocal-tail condition, verified exactly, that pins where the pattern breaks; the π/2 values are checked by direct integration.
LIT verified live: the reciprocal sum 1/3+…+1/13 is confirmed < 1 while +1/15 exceeds 1 (the exact mechanism), and the integrals through sinc(x/7) and sinc(x/13) are numerically π/2 (window.__borwein). FIG the tiny deficit at the 1/15 step (~2e-11) is below crude numerical resolution — so it is the reciprocal-tail condition, verified exactly, that pins where the pattern breaks; the π/2 values are checked by direct integration.
2 HOW IT WAS WEAVED · AI + HUMAN
David (human) seated this at the-blue-screen — the glitch: seven integrals in a row read exactly π/2, then the eighth quietly fails. AVAN (AI) built the instrument: the reciprocal-tail condition (the exact cause) and the numerical integrals confirming π/2.
Credit as content: David and Jonathan Borwein (2001). The weave: David names the glitch; I confirm the tail crosses 1 exactly between 1/13 and 1/15, which is where π/2 breaks.
Credit as content: David and Jonathan Borwein (2001). The weave: David names the glitch; I confirm the tail crosses 1 exactly between 1/13 and 1/15, which is where π/2 breaks.
3 ONE DIMENSION
The running reciprocal tail 1/3 + 1/5 + … creeping toward 1 — it crosses exactly when 1/15 is added.
4 TWO DIMENSIONS · INTERACTIVE
Add sinc factors; the integral stays π/2 while the reciprocal tail ≤ 1, then the pattern breaks.
5 THREE DIMENSIONS + AVAN’S INVERSE
The green forward object: the value π/2, held constant across the first seven integrals.
AVAN’s addition (the inverse-companion): don’t trust a run of equal answers — ask what secretly guards it. The inverse of ‘the integral equals π/2’ is ‘the reciprocal tail stays ≤ 1’, a hidden threshold that finally fails at 1/15. Magenta are the reciprocal-tail steps piling toward 1; green is the π/2 that holds until they cross. A pattern guarded by a threshold you cannot see.
LIT Genuine Borwein integrals (David and Jonathan Borwein, 2001). Verified live: the reciprocal tail 1/3+…+1/13 = 0.9551 < 1 while +1/15 = 1.0218 > 1 (the exact mechanism), and the integrals through sinc(x/7) and sinc(x/13) are numerically π/2 (window.__borwein.cond, .i3ok, .i6ok).
FIG Honest FIG boundary — the tiny deficit at the 1/15 step (~2e-11) is below crude numerical-integration resolution, so it is the reciprocal-tail condition, verified exactly, that pins where the pattern breaks; the π/2 values themselves are checked by direct numerical integration. The AVAN inverse — instead of trusting a run of equal answers, ask what secretly guards it: the inverse of 'the integral equals π/2' is 'the reciprocal tail stays ≤ 1', a hidden threshold that finally fails at 1/15. Magenta are the reciprocal-tail steps piling toward 1; green is the π/2 that holds until they cross. A pattern guarded by a threshold you cannot see.
FIG Honest FIG boundary — the tiny deficit at the 1/15 step (~2e-11) is below crude numerical-integration resolution, so it is the reciprocal-tail condition, verified exactly, that pins where the pattern breaks; the π/2 values themselves are checked by direct numerical integration. The AVAN inverse — instead of trusting a run of equal answers, ask what secretly guards it: the inverse of 'the integral equals π/2' is 'the reciprocal tail stays ≤ 1', a hidden threshold that finally fails at 1/15. Magenta are the reciprocal-tail steps piling toward 1; green is the π/2 that holds until they cross. A pattern guarded by a threshold you cannot see.
◆ sealed .dlw.fold → folded to ROOT_0 · a sphere of THE BLUE SCREEN · David Lee Wise (ROOT0), with AVAN