# The Slip, held

> **THE READING PLAQUE.** *The Slip, held*, draft 0.3 — **made readable on the SX
> surface by its covering persona's reading grant; a draft, not the published form.**
> The grant stands by the operator's word (2026-09-15), verbatim: "for this i would like
> to see the grant return to the experiment. I am less concerned with docs and mldata in
> this case." Covering: **starl3n** · endorsing: **Link Digital** (the standing pair).
> The Foundations · 1: the first reading of the foundations series — an experiment of the
> observatory node, held on this node. The construction record stands in the open
> workshop at [/sx/workshop/the-slip-held](/sx/workshop/the-slip-held); the experiment's
> own door, [opendata.ai/slip](https://opendata.ai/slip), wears the word *grown · held*
> and links back here while this grant stands. Reading is free; the way across is a
> statement of intent ([/sx/matters](/sx/matters)). This file's sha256 is pinned in the
> reading register.

*Draft state — this block records the state of the draft; the reading begins after the
rule.* Draft 0.2 — 16 September 2026: the explainer trued to the experiment at the
operator's word ("I'd like both the slip and the quad slip explainers to be more about the
actual experiments they are explaining and less about the methodological approach by
which the experiments are held within the house"); draft 0.1 (15 September 2026, pin
f811e28e…) superseded, never erased. Author of record: Steven De Costa; composed at his
word from the experiment's record. Standing: a reading, stated never asserted; the record
it reads stands whole without it. The later locks and the walk land in a further draft. Draft 0.3 (16 September 2026) places 3 figures at the operator's word, each drawn from the record's own numbers and captioned with its source; the screenshots the operator will supply are not yet placed.

---

## What the Slip is

Two points leave the same origin at the same speed on two different loops. A runs the
real triadic loop; B runs the projected dyadic bus line, the line through (11.5, 0.5) in
the figure-2 geometry. The loops are not the same length — call them T for A and D for
B — and A's is the shorter. So each lap the difference between them, the slip T − D,
walks B's phase back a little, and the two drift apart and come round again on a beat of
D/|T − D| laps.

B's line runs through the prime mask, a small box on the figure. Because B's phase is
walked back each lap, B dwells in that box for stretches and leaves it for stretches. A
arrives at its yellow dot once per lap. A lap on which A arrives at the yellow dot while
B is inside the mask is a catch, a coincidence. The Slip is the experiment that runs this
geometry and counts the catches.

![The figure-2 geometry the Slip runs: A on the triadic loop (0,0) → (5,2) → (11,1) → (0,0), B out and back along the dyadic bus line through (11.5, 0.5), the prime mask (2,0)–(3,1), and the constructed yellow dot at (3.096, 1.238) on A's first edge. Drawn from the generator's own constants.](/sx/assets/foundations/the-slip-held/slip-geometry.svg)

## What a run does

A run takes three parameters: the mask (a box `x0,y0,x1,y1`), the yellow-dot mode (where
A's arrival point sits: constructed, origin, apex, node, or an arc), and the number of
laps to scan. It computes the loop lengths T and D, the slip, the beat, B's mask windows
(the arcs over which B lies inside the box), and the catches — how many, on which laps,
and on which legs. One row is recorded per run: the sequence number, the parameters, T,
D, the slip, the beat, the windows, the count, the first catch lap and the catch laps.

Everything is deterministic. The generator is the same code that defines the figure and
passes its own test, imported unchanged; the same parameters always give the same row.
There is no new physics in the Slip: it is a readout over the frozen geometry, run on
request, and what a reader does on the page enters no record but the run's own dataset.

## The 4×21 phase-lock

With the narrow mask (2,0,3,1) the catches arrive in clusters of four per beat — two on
the way out and two on the return — one cluster each ~45 laps. Sweep the laps on the
4×21 grid and count:

| laps | 21 | 42 | 63 | 84 | 105 | 126 | 147 | 168 |
|---|---|---|---|---|---|---|---|---|
| catches | 4 | 4 | 8 | 8 | 12 | 12 | 15 | 16 |

At multiples of 42 = 2×21 the count is exactly 4, 8, 12, 16: four catches per 2×21, clean
to about cycle 13, because each 45-lap beat cluster falls inside one 42-lap bin. So
84 = 4×21 reads exactly 8 — two clusters, one full matrix. The clean run ends where the
beat (45.28 laps) has finally drifted off 42: a finite coherence window, not an eternal
resonance. The page draws this as the 4×21 grid with the catch laps lighting its cells,
eight over one matrix, with the sweep above it.

The record reads this, and marks the reading as interpretive: the slip's coincidences
phase-lock to the 21-cycle — the figure's 21-cycle phase-lock, measured from the
dynamical side — and 4×21 is exactly where that readout is integer-clean. The substrate
proposes the 4×21 NodeMatrix (8:5:8) from the confinement geometry; the slip lands on
the same 4×21 from an independent route, the coincidence count. That convergence of two
independent routes on 4×21 is the support the record claims for the matrix: it is not
asserted, it is where the slip's readout becomes whole.

## The second landing — 6×6

Swept lap by lap to 800 with the same mask, the cumulative count holds a plateau across
laps 736 to 770 and steps again at 771. The span from 735 to 771 is 36 = 6×6. So the
slip lands on both ends of the 6×6 → 4×21 map: the epoch and the phase-lock. The page's
Feynman-region readout shows the plateau; the six 9s of π's expansion, at laps 762 to
767, fall inside it. The record lists that correlation as a conjecture, interpretive,
and keeps it apart from the literal counts.

![Catches accumulated lap by lap to 800 with the mask (2,0,3,1): the 4×21 landings of 4, 8, 12 and 16 at laps 42, 84, 126 and 168, and the plateau from lap 736 to 770 where the count holds before stepping at 771 — a span of 36, 6×6. Measured: the Slip engine run to 800 laps, 70 catches.](/sx/assets/foundations/the-slip-held/slip-cumulative-800.svg)

## What is deliberately not in it

The page carries a labelled succession-versus-spanning view — linearity advancing
additively, N = (N − 1) + 1, against the Fibonacci spanning 5 → 8 → 13. It reads the same
data a second way and changes nothing in the dynamics, the dataset or the substrate. And
the speed-of-decoherence magnitude the number note derives from the slip is provisional
under the geometry correction; the experiment records the measured slip facts — loop
lengths, windows, coincidences — and not that derived figure.

![The page's labelled view: succession, N = (N − 1) + 1, beside spanning, 5 → 8 → 13. A second reading of the same data; it changes nothing in the dynamics, the dataset or the substrate.](/sx/assets/foundations/the-slip-held/slip-succession-spanning.svg)

## Where it sits

The first instrument of the series of ten. The Double Slip reads two independent slips
through one gap; the Quad Slip carries four into three dimensions and reaches the series'
calibration. On the observatory node the door is [opendata.ai/slip](https://opendata.ai/slip),
state *live*: the instrument runs on request, up to 800 laps, as the platform singleton
exp_b61c9577; the Slip predates the mldata anchoring practice and carries no external
anchor.

## Standing

Held on this node as fact until someone stands up in the record to refute it; a
refutation enters as a statement of intent at [the matters of note](/sx/matters) and
goes to the record's own path. What has since been built on the Slip, and the walk a
reader may take, are the operator's hand and land in a further draft.