MODEL QUICKSILVER & KERNEL · DOG ON A LEASH, CUBITS IN TOW · II · under KEEPER ERIS
dog on a leash, cubits in tow · II · tow −4 bubbles
MODEL The tow starts 4 bubbles behind the dog and chases where the dog was 12 frames earlier. The I-13 VM works out the steady lag.
LIT Dave (t260): "can you build these into spheres, agentic , artistic please"
Dave's source
LIT quicksilver arc — dog on a leash, cubits in tow: his original page, kept byte-identical (open it to run it as he built it).
What his system says, checked on its own terms
- checking…
MODEL The sphere world · BUILDER + WEAVER + CRITIC
MODEL BUILDER (ripple, gloss, spin) and WEAVER (glow, spread) take turns shaping the sphere and CRITIC scores it against a rubric made of the source's own numbers (my mapping). The agents also run his mechanics on the sphere: each run follows his step() rules, and CRITIC reads its ledger and receipt; the page throws if any run breaks his invariant. Every 40 rounds the workspace loop reviews: a plateau means explore wider, a gain means refine. On each new best the I-13 VM re-computes the CRITIC score.
Runs read by CRITIC
Workspace log
Coop board (this browser, localStorage mayhem-quicksilver-kernel-board, MODEL)
Dave’s I-13 engine on this page
booting I-13 v0.3…
LIT What the VM computes here: the closed forms of his line (99 − 1, the alternating sum, bubbles per nm, the thread ratio, frames to delivery, the steady tow lag), compared against JS. The CRITIC score is also re-computed on each new best. JS runs his frame-by-frame step() and draws.
Side notes
LIT Dave (t262): "thats for the standard model not theoretical math magic"
MODEL How these pages read that: the arc and the kernel are checked as a theoretical system of his own, for internal consistency and for whether the code does what its notes say. His 1.61e-35 is his model's own unit: the tick (s) in the kernel and the bubble (m) on the arc.
LIT Standard Model reference, not a correction: in standard physics 1.616 × 10⁻³⁵ m is the Planck length and 5.39 × 10⁻⁴⁴ s is the Planck time; the arc's thread (~1.1502e-36 m) sits below that length.
WALL There is no experimental access below the Planck scale at present, so nothing here can be tested by measurement; what can be checked is the system's own arithmetic, and that is what these pages check.
AMBER self-check running…
Seal sha256(name|slug|blurb): 65418db8ea660672061dfe41cec75268ccba1a696c5be07000a0b35170e3cd93