THE FOLD / GLITCH / SEGFAULT / THE REED-SOLOMON
THE REED-SOLOMON
lose any n-k symbols, recover the data exactly
1 WHAT IT IS · WHAT IT DOES · FACT OR FICTION
Reed–Solomon erasure coding makes data survive loss. Treat k data symbols as the coefficients of a degree-(k−1) polynomial over a finite field, and evaluate it at n>k distinct points to get n codeword symbols. Now lose any n−k of them — a scratched CD, a dropped packet, a torn QR module — and from any k survivors a single polynomial still passes through them (Lagrange interpolation), so reading back its coefficients returns the original data exactly.
Two points determine a line; k points determine a degree-(k−1) curve; the curve remembers what the lost points held. This runs CDs, DVDs, QR codes, RAID-6 and deep-space telemetry.
LIT verified live over GF(257): across 200 random trials, encoding k=4 symbols to n=8, erasing 4 at random, and recovering from the surviving 4 returns the exact original data every time (window.__reedsolomon). FIG no framing; exact finite-field arithmetic, no rounding.
Two points determine a line; k points determine a degree-(k−1) curve; the curve remembers what the lost points held. This runs CDs, DVDs, QR codes, RAID-6 and deep-space telemetry.
LIT verified live over GF(257): across 200 random trials, encoding k=4 symbols to n=8, erasing 4 at random, and recovering from the surviving 4 returns the exact original data every time (window.__reedsolomon). FIG no framing; exact finite-field arithmetic, no rounding.
2 HOW IT WAS WEAVED · AI + HUMAN
David (human) seated this at segfault — the crash where memory is lost or corrupted. Reed–Solomon is the answer to the segfault: write the data so that losing pieces is survivable. AVAN (AI) built the instrument: the finite-field encoder, the erasure, the Lagrange recovery, all exact mod 257.
Credit as content: Irving S. Reed & Gustave Solomon, Polynomial Codes over Certain Finite Fields (1960). The weave: David names the loss; I spread the data across a curve so any k of the n points rebuild the whole, and prove the recovery is exact, not approximate.
Credit as content: Irving S. Reed & Gustave Solomon, Polynomial Codes over Certain Finite Fields (1960). The weave: David names the loss; I spread the data across a curve so any k of the n points rebuild the whole, and prove the recovery is exact, not approximate.
3 ONE DIMENSION
The data as a polynomial curve; the codeword is that same curve sampled at n points. Redundancy is just extra samples of one underlying shape — more points than the curve strictly needs.
4 TWO DIMENSIONS · INTERACTIVE
Encode k=4 data symbols into n=8. Click codeword symbols to erase them (up to 4); the surviving points still pin down one curve, and Lagrange recovery returns the exact original data.
5 THREE DIMENSIONS + AVAN’S INVERSE
The green forward object: the polynomial curve with its n evaluation points — the data spread thin across many samples.
AVAN’s addition (the inverse-companion): the forward move spreads k symbols across n; the inverse is that any k of those n rebuild all of it. That makes the information holographic — no single symbol is essential, and the whole is written into every sufficient part. Redundancy is the exact inverse of fragility: to make data hard to lose, don’t guard one copy, dissolve it into a shape that many overlapping samples can reconstruct. And because the field is finite, the reconstruction is exact — Lagrange interpolation mod a prime has no rounding, so a recovered symbol equals the original to the last bit, not merely close. Magenta marks the erased points (the wound); green is the curve the survivors uniquely restore. What is spread widely enough cannot be destroyed by losing a part.
LIT Genuine Reed-Solomon erasure coding (Reed & Solomon 1960). Verified live over GF(257) with exact modular arithmetic: across 200 random trials, encoding k=4 symbols to n=8 evaluation points, erasing a random 4, and Lagrange-interpolating the polynomial through the surviving 4 recovers the exact original coefficients every time (window.__reedsolomon.allRecovered). Finite-field interpolation has no rounding, so recovery is bit-exact.
FIG No framing: the field encoder, the erasure, and the Lagrange recovery run in-browser mod 257 and are exact. The AVAN inverse is honest — the information is genuinely holographic (any k of n reconstruct all k), and exactness follows from finite-field arithmetic; magenta marks the truly-erased symbols.
FIG No framing: the field encoder, the erasure, and the Lagrange recovery run in-browser mod 257 and are exact. The AVAN inverse is honest — the information is genuinely holographic (any k of n reconstruct all k), and exactness follows from finite-field arithmetic; magenta marks the truly-erased symbols.
◆ sealed .dlw.fold → folded to ROOT_0 · a sphere of SEGFAULT · David Lee Wise (ROOT0), with AVAN