---
slug: the-air-that-got-there-first
passed: 2026-10-14
by: the assay line (research/assay-line/cited-1014b/), Codex fixer directed by claude-assaying-boyle-1014; source record by Codex, checked by Codex
checked: 81
confirmed: 70
wrong: 10
unverifiable: 0
observed: 0
stale: 1
---

This was the directed cited-1014b repair pass against f5e5f84f34 for this layer. The fixer independently re-read the evidence for all 11 triage items, replayed the relevant browser functions, reran the layer verifier, fixed two WRONG statements and the STALE modification date, and left eight listed MINOR items unchanged as directed. The 70 CONFIRMED lines below carry forward the claims and evidence re-read by the Codex checker in research/assay-line/cited-1014b/the-air-that-got-there-first/N2-codex.md, whose source record is N1-sources.json in that directory; their notes describe that checker's work, not a claim that the fixer repeated every earlier check. The wrong count includes the eight MINOR lines, and this record covers the triage plus those CONFIRMED claims rather than replacing the checker's full audit. The repair worker's 2026-10-14 review confirmed that memory/ledger-archive/P19.md:15 still repeats both WRONG claims. That shared file is outside the repair worker's write scope, so X3 and X7 remain open pending a board note and a directing-instance edit.

## Claims

- MINOR | "circulation / momentum" | https://www.cambridge.org/core/journals/journal-of-fluid-mechanics | [X1; momentum-link-citation] Reopened the journal homepage on 2026-10-14. Its journal description and article listings do not identify a particular circulation/momentum derivation. This is an imprecise citation, not evidence against the physics.
- MINOR | "each speck is carried by the exact velocity of the flow at the point where it sits, step by step." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [X2; dye-step-by-step] Replayed the actual page functions at 6 degrees. The marked top parcel remains at x=2.0013284914783585 after arrival although u=0.8929811225514973 there; the arrived-particle guard skips further advection. The velocity-field claim holds before arrival but omits that stop.
- WRONG | "(Turn the angle negative and the whole thing flips" ... "the lift points down, and now the under-side air wins the race.)" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [X3; negative-flip] Independently replayed every negative slider setting (-4 to -0.5 degrees, step 0.5). Each has positive lift and the top marked parcel arriving first. At -4 degrees CL=0.1442570434909173 (display 0.14), with the top marked parcel 5% sooner after rounding. The model gives beta=-5.194428907734806 degrees, below the slider minimum. Corrected the parenthesis and the layer build-log copies.
- MINOR | "Everything on the panel is recomputed offline, with no libraries, by research/the-air-that-got-there-first/verify.mjs." | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [X4; offline-panel] Read both implementations and ran the verifier and actual browser functions. At 6 degrees browser peaks are 1.843617058944955/1.3408802881332407 (display 1.84/1.34); the verifier prints 1.844/1.360. It samples 2000 positions versus 1200 in the browser, and the transit initial conditions differ. Offline recomputation does not reproduce every panel value.
- MINOR | "Crank the angle of attack to 14° here and the lift keeps rising forever." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [X5; no-stall-forever] The slider ends at 14 degrees. Gamma is proportional to sin(alpha-beta), so it is bounded, reaching 13.880010011137037 near 84.8055710922652 degrees and falling to 11.342757273227884 at 120 degrees. Lift rises across the offered range; the no-stall limitation is real, but forever is loose.
- MINOR | "Every displayed number is recomputed from the same equations by research/the-air-that-got-there-first/verify.mjs" | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [X6; every-displayed-number] Same offline-recomputation scope issue as X4, X9 and X11. The footer overstates exact reproduction: browser peaks display 1.84/1.34 while the offline verifier prints 1.844/1.360. The two implementations do share the flow equations.
- WRONG | "the page computes the lift both ways and they agree to 0.000%" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html:593; git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs:286 | [X7; metadata-agreement] Read the browser pressure-field drawing and updateReadouts: the displayed CL comes from circulation only. The separate verifier computes L_KJ and liftFromPressure(alpha). Rerun gives L_pressure=2.6947, L_KJ=2.6947, diff=0.000% at 6 degrees. Changed the dek and layer build log to credit the offline verifier; retained the supported rounded result.
- MINOR | "Every number recomputed live." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [X8; metadata-every-number] Read the og:description and updateReadouts. The live updates cover angle, circulation, CL, peak speeds and transit lead. The 0.000% comparison and 0.47-chord downstream lead remain fixed explanatory text. The universal live-recomputation sentence overstates scope.
- MINOR | "the peak surface speeds (1.84 / 1.34 × U at 6°)" | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [X9; md-peak-speeds] Reproduced the browser estimate 1.84/1.34 and verifier estimate 1.844/1.360. The displayed bottom estimate is not a mistaken transcription. The MINOR issue is the surrounding promise that the offline verifier reproduces that value, shared with X4, X6 and X11.
- STALE | ""dateModified":"2026-06-29"" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html:40; git show 2e60bcb5dd1:public/strata/the-air-that-got-there-first/index.html | [X10; modified-metadata] The original date matched creation on 2026-06-29. Read commit 2e60bcb5dd1, recorded as 2026-10-01T03:47:58+10:00, which changed og:image:alt afterwards. Set dateModified to 2026-10-14 for these corrections. The explicit JSON-LD exception requires no separate Updated line.
- MINOR | "Every displayed number" | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [X11; md-all-numbers] This fragment belongs to the same markdown sentence as X9. Actual browser bottom peak is 1.3408802881332407, displayed as 1.34; the verifier prints 1.360. The defect is offline checking coverage, not a false browser readout.
- CONFIRMED | "The most-repeated explanation of flight is wrong." | https://www.cam.ac.uk/research/news/how-wings-really-work | [popular-explanation] NASA says "The most popular incorrect theory of lift", which is narrower than the claim. Cambridge supplies the stronger description: "the most commonly used explanation cannot possibly be correct". The recorded NASA excerpt alone does not establish the superlative.
- CONFIRMED | "A wing's top is curved, so the air crossing it has farther to travel than the air underneath." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/foilw1/ | [myth-longer-path] This is explicitly the story being refuted, not the author endorsing it. NASA presents "farther to travel over the top of the airfoil" under "Incorrect Airfoil Theory".
- CONFIRMED | "To arrive at the trailing edge at the same moment as its partner below, the top air must speed up." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/foilw1/ | [myth-simultaneous] NASA describes the myth using "In order to meet up at the trailing edge" and "must travel faster". Read in the page's stated myth context.
- CONFIRMED | "Faster air has lower pressure (Bernoulli), so the top is sucked upward, and the wing flies." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/foilw1/ | [faster-lower-pressure] NASA says "The difference in pressure across the airfoil produces the lift" and "this part of the theory is correct". The erroneous premise is the equal transit speed, not Bernoulli.
- CONFIRMED | "It assumes two parcels of air that split at the front of the wing must rejoin at the back" | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [hidden-assumption] NASA explicitly describes the supposed requirement to "reach the trailing edge at the same time".
- CONFIRMED | "The air going over the top reaches the trailing edge first, with time to spare, and never meets its partner again." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/foilw1/; https://www.cam.ac.uk/research/news/how-wings-really-work | [top-first-never-meets] NASA: "particles moving over the top arrive at the trailing edge before particles moving under the airfoil". Cambridge independently observes the upper smoke already downstream when the lower smoke arrives. Read as non-reunion in the demonstrated flow, not a claim about all possible future turbulence.
- CONFIRMED | "Equal transit time predicts the two halves arrive together." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [equal-transit-prediction] NASA describes the proposed same-time arrival; this is correctly labelled the false theory's prediction.
- CONFIRMED | "The two parcels never reunite." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/foilw1/ | [never-reunite] NASA: "Two molecules near each other at the leading edge will not end up next to each other at the trailing edge". The page's example is a steady lifting flow.
- CONFIRMED | "They are the same flow, counted two ways." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [two-languages] NASA says both descriptions are correct and that integrating either pressure or velocity determines the aerodynamic force.
- CONFIRMED | "The wing throws a great sheet of air downward." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [newton-downward] NASA identifies "a net turning of the gas flow" and the reaction force. Upward lift corresponds to downward momentum imparted to the air.
- CONFIRMED | "Push air down, and by the third law the air pushes the wing up." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [third-law] NASA expressly applies "Newton’s third law" to flow turning and the resulting aerodynamic reaction.
- CONFIRMED | "That downward deflection" ... "the downwash" ... "is the lift, plainly." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [downwash-is-lift] In context, this is a description of the momentum account of lift, rather than an equation equating a velocity with a force. NASA supports the turning-action/reaction account.
- CONFIRMED | "faster flow means lower pressure, so the surface pressures, added up, push the wing up by exactly the same amount." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [bernoulli-same-amount] NASA describes the pressure integral and says the velocity integral also determines the same aerodynamic force. The record excerpt establishes the pressure half; the rest of the same source supplies the equivalence.
- CONFIRMED | "The top is not faster because it has a longer path to cover in a fixed time." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [longer-path-not-reason] NASA states the actual upper velocity is higher than the equal-transit velocity. The longer-path/fixed-time explanation is rejected.
- CONFIRMED | "It is faster because the air has circulation around the wing" | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm | [circulation-speeds] MIT attributes the upper/lower pressure and velocity differences to circulation. Context includes its Kutta condition discussion.
- CONFIRMED | "a net tendency to flow forward under the belly and back over the top" | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm; git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html:315 | [circulation-direction] For positive lift the page uses clockwise circulation: its term "mul(cx(0,Gamma(al)/(2*Math.PI)), inv(xi))" increases downstream velocity above and decreases it below. MIT's quoted prose gives the speed asymmetry but not the directional wording; the model supplies the sign convention.
- CONFIRMED | "air cannot whip around the infinitely sharp trailing edge." | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm | [sharp-edge] MIT explicitly requires finite velocity and smooth departure at the sharp edge. This is the stated ideal sharp-edge model, not a claim that every physical trailing edge has zero radius.
- CONFIRMED | "It would need infinite speed to do so" | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm | [infinite-speed] The full MIT source says potential theory prescribes infinite velocity at the sharp trailing edge for the non-Kutta corner flow. This supporting sentence precedes the record's excerpt.
- CONFIRMED | "nature picks the one amount of circulation that lets the flow leave the sharp edge smoothly." | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm | [unique-circulation] MIT states "a unique solution" after imposing the trailing-edge condition.
- CONFIRMED | "That is the Kutta condition" | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm | [kutta-name] MIT identifies smooth, finite-velocity departure as the Kutta condition.
- CONFIRMED | "it is the real author of the speed difference" | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm | [kutta-causes-lift] Read with the following viscosity discussion: Kutta selects the circulation in this model; MIT connects that circulation with the velocity and pressure differences.
- CONFIRMED | "a lift-curve slope of ~2π per radian, the thin-airfoil result." | https://www.grc.nasa.gov/WWW/k-12/VirtualAero/BottleRocket/airplane/incline.html | [thin-airfoil-result] NASA's incline page gives "the lift coefficient is 2 x pi x angle" for thin airfoils at small angles in radians.
- CONFIRMED | "d'Alembert's paradox, the honest tell that this is inviscid theory" | https://web.mit.edu/2.25/www/225_sect_10.html | [dalembert-paradox] MIT 2.25 names "The nonexistence of drag (D'Alembert's paradox)" for ideal potential flows around 2D bodies.
- CONFIRMED | "it is how the lift of a wing is actually computed to first order" | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm; https://www.grc.nasa.gov/WWW/k-12/VirtualAero/BottleRocket/airplane/incline.html | [first-order-model] NASA gives a mathematical small-angle thin-airfoil approximation. MIT explicitly describes potential-flow lift calculations. This supports the stated first-order, idealised use.
- CONFIRMED | "Viscosity is what sets up the circulation in the first place." | https://web.mit.edu/2.25/www/225_sect_10.html | [viscosity-sets-circulation] MIT 2.25 states viscosity is responsible for the existence of lift and drag through separation, despite being small at high Reynolds number.
- CONFIRMED | "A truly frictionless fluid would never start the flow turning" | https://ocw.mit.edu/courses/16-01-unified-engineering-i-ii-iii-iv-fall-2005-spring-2006/6eb81ebf6b649fe55d73c49c4fb4af9e_f01_0304.pdf | [frictionless-circulation] UNREADABLE resolved: downloaded MIT PDF and read both pages with pdftotext. Kelvin's theorem conserves circulation for a material circuit; the initial-rest example has zero circulation. "Turning" here means generating circulation, not that inviscid streamlines cannot bend.
- CONFIRMED | "a "starting vortex" is shed from the trailing edge as the wing accelerates" | https://ocw.mit.edu/courses/16-01-unified-engineering-i-ii-iii-iv-fall-2005-spring-2006/6eb81ebf6b649fe55d73c49c4fb4af9e_f01_0304.pdf | [starting-vortex-shed] UNREADABLE resolved: MIT PDF p. 2 explicitly says a starting vortex is shed from the trailing edge when the airfoil is set into motion.
- CONFIRMED | "its equal-and-opposite reaction is the circulation you see here." | https://ocw.mit.edu/courses/16-01-unified-engineering-i-ii-iii-iv-fall-2005-spring-2006/6eb81ebf6b649fe55d73c49c4fb4af9e_f01_0304.pdf | [opposite-circulation] UNREADABLE resolved: MIT PDF p. 2 gives equal and opposite airfoil and starting-vortex circulations. The argument is circulation conservation; "reaction" is informal here.
- CONFIRMED | "The Kutta condition is the fingerprint that viscosity leaves on an otherwise inviscid picture." | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm | [viscous-fingerprint] MIT's finite-velocity Kutta discussion explicitly invokes viscous shear effects; the inviscid outer-flow model retains that boundary condition.
- CONFIRMED | "A real wing's boundary layer separates somewhere around 15°, the flow detaches, and lift collapses." | https://www.aviation.govt.nz/licensing-and-certification/pilots/flight-training/flight-instructor-guide/wing-drop-stalling/ | [stall-fifteen] CAA: "at about 15 degrees" and "losing lift". The page says "somewhere around", so this is an approximate conventional-wing statement, not a universal stall angle.
- CONFIRMED | "A real, finite wing also sheds vortices off its tips, adding "induced" drag the 2-D model has none of." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/drag/ | [tip-vortices] NASA describes wing-tip vortices and induced drag as the associated energy loss. A two-dimensional infinite-span model has no tip-vortex drag.
- CONFIRMED | "In the real, viscous flow the top air still reaches the trailing edge well before the bottom air" | https://www.cam.ac.uk/research/news/how-wings-really-work | [real-smoke] Cambridge describes smoke already past the wing on top when the lower smoke reaches the end.
- CONFIRMED | "that is exactly what the smoke-tunnel experiments show." | https://www.cam.ac.uk/research/news/how-wings-really-work | [smoke-experiments] Cambridge describes Babinsky's smoke-pulse filming around an aerofoil and the unequal transit times.
- CONFIRMED | "NASA Glenn Research Center, "Incorrect Lift Theory" (Beginner's Guide to Aeronautics)" | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/foilw1/ | [nasa-citation] The link is NASA Glenn's equal-transit debunking. Its current heading is "Equal Transit Theory" and it includes "Incorrect Airfoil Theory"; "Incorrect Lift Theory" is the page's descriptive citation label.
- CONFIRMED | "The "equal transit time" / "longer path" explanation is identified as incorrect by NASA Glenn Research Center" | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [nasa-identifies-error] NASA explicitly calls the longer-path/equal-transit explanation incorrect.
- CONFIRMED | "Holger Babinsky, "How do wings work?", Phys. Educ. 38 (2003) 497." | https://api.crossref.org/works/10.1088%2F0031-9120%2F38%2F6%2F001 | [babinsky-citation] Crossref independently returns "How do wings work?", "Holger" / "Babinsky", volume "38", issue "6", pages "497-503", published 2003-11-01. IOP curl reached a bot challenge, so that capture is not treated as article text. Online tool also reported DOI MATCH.
- CONFIRMED | "the smoke-pulse experiment showing the top flow arriving first is Holger Babinsky" | https://www.cam.ac.uk/research/news/how-wings-really-work; https://www.researchgate.net/publication/230910482_How_do_wings_work | [babinsky-smoke-attribution] Cambridge expressly attributes filming the smoke pulses to Babinsky. The author-uploaded full text of the 2003 paper was also read: p. 498 describes Figure 4 as frames from a smoke-wind-tunnel experiment and says the particles do not meet at the trailing edge.
- CONFIRMED | "Anderson, Fundamentals of Aerodynamics" | https://www.mheducation.com/highered/product/fundamentals-of-aerodynamics-anderson.html?viewOption=student | [anderson-citation] Publisher page identifies "Fundamentals of Aerodynamics by John D. Anderson". The entry asserts the book attribution, not a specific page reference.
- CONFIRMED | "Acheson, Elementary Fluid Dynamics" | https://academic.oup.com/book/53981?searchresult=1 | [acheson-citation] Original chapter URL returned HTTP 403. Oxford's book landing page is readable and gives "Elementary Fluid Dynamics" and "D J Acheson", published 15 March 1990.
- CONFIRMED | "accounting of lift is standard." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [momentum-standard] NASA's pressure/velocity and momentum discussion supports the physics independently of the generic journal link.
- CONFIRMED | "the exact, closed-form solution for an idealised airfoil, the same one the verifier checks." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [exact-model] Read the complete complex potential, Joukowski map, inverse root selection, velocity derivative and circulation condition in the pinned code. The field is the stated ideal 2D solution. N1's excerpt is an unmarked splice of code blocks.
- CONFIRMED | "The ribbon tears because the flow tears it." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [ribbon-tear] The pre-edge positions come from RK4 evaluation of velAt, not prescribed tracks. Drawing the two sides separately is a display convention. The stopping limitation is recorded against dye-step-by-step.
- CONFIRMED | "The same flow is faster over the top (you can read the peak speeds off the panel above) and slower below" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [faster-top-slower-bottom] At the default 6 degrees the peak readouts are top 1.84 and bottom 1.34 times U. "Slower below" is relative to the top, not a claim that every lower-surface speed is below freestream. The two peak speeds alone do not establish every local velocity.
- CONFIRMED | "once from the circulation (Kutta–Joukowski, L = ρ·U·Γ) and once by integrating the Bernoulli pressure over the whole surface" | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [lift-two-ways] The verifier implements "const L_KJ = rho * U * Gamma(alpha);" and "const press = liftFromPressure(alpha);". The second computes the pressure integral around the surface.
- CONFIRMED | "At 6° angle of attack it confirms:" | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [check-at-six] Most reported results use "const alpha = deg(6)". The Kutta residual loop itself tests "[0, 4, 8]", rather than 6. The condition is analytic for all these angles; do not read the introductory phrase as a transcript of the residual loop.
- CONFIRMED | "Γ = 4π·a·U·sin(α−β)" | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [gamma-formula] Source definition: "const Gamma = (alpha) => 4 * Math.PI * a * U * Math.sin(alpha - beta);". The trailing-edge numerator cancels under this choice.
- CONFIRMED | "Equal transit time predicts 0%." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [zero-lead-prediction] Equal arrival times imply (t_bottom-t_top)/t_bottom = 0. NASA describes exactly that false premise.
- CONFIRMED | "This is potential flow: two-dimensional, inviscid (no viscosity), incompressible, steady." | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [model-assumptions] The complex potential and fixed alpha define a steady 2D incompressible inviscid exterior flow. Time-stepping tracer positions does not make the velocity field unsteady.
- CONFIRMED | "it has no boundary layer to lose." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [no-boundary-layer] The velocity comes directly from the inviscid potential/map quotient. There is no viscous boundary-layer or separation state.
- CONFIRMED | "The flow here is the classical Joukowski airfoil with the Kutta condition" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [page-joukowski] The source contains "function zOf(zeta){ return add(zeta, sca(inv(zeta), c*c)); }" and the Kutta circulation. N1 combines separated blocks without elision markers.
- CONFIRMED | "which cross-checks the lift two independent ways." | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [crosscheck-independent] The source computes circulation lift and a surface-pressure quadrature separately. They share the specified physical flow, but one does not merely print the other.
- CONFIRMED | "Drag the angle of attack and read the circulation, the lift coefficient, the peak speeds, and the head start, all recomputed live." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [metadata-live] The input listener changes alpha, calls recompute, then updateReadouts. The four named panel quantities are recomputed from the selected flow.
- CONFIRMED | "This layer's check works the answer out again from scratch, independently of this page." | git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [check-independence] The verifier has no imports and reads no page file. Running the downloaded file alone succeeds. This supports independence from page text, not the stronger assertion that two independently invented physics models were compared.
- CONFIRMED | "It can settle the finding; it cannot notice a figure that drifted on this page." | git show f5e5f84f34:research/verifier-independence/assert-power.json | [cannot-drift] The retained assert-power row says "the condition names nothing the check read from outside itself" for every assertion. The complete verifier reads no page. Evidence: evidence/assert-power-row.json.
- CONFIRMED | "The idealisations (2-D, inviscid, incompressible, no stall) are named on the page, not buried." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [md-idealisations] HTML lines 232-245 explicitly name "two-dimensional, inviscid (no viscosity), incompressible, steady", "No stall" and "Two dimensions only".
- CONFIRMED | "under-the-wing: at the edge ✓" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [verdict-bottom] This literal occurs in the conditional "else if(tb!=null)" branch. It describes that hypothetical state correctly, but the permitted negative angles do not reach the advertised reversed-race case.
- CONFIRMED | "over-the-top: still en route…" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [verdict-top-wait] The bottom-only-arrived branch contains "over-the-top: still en route…". Branch existence is not evidence that the slider can produce negative lift.
- CONFIRMED | "There the tide is not the Moon's pull but the tiny difference of that pull across the Earth" | https://www.noaa.gov/education/resource-collections/ocean-coasts/tides | [related-tides] UNREADABLE resolved: curl returned HTTP 200. NOAA describes the Moon pulling more strongly on the near side and more weakly on the far side; "This difference, known as the tidal force" supports the differential-force claim.
- CONFIRMED | "'The metal is cold' is really a rate of heat leaving the skin" | https://van.physics.illinois.edu/ask/listing/44029 | [related-metal] Illinois Physics Van explains that metals transfer heat readily and can feel colder at the same temperature. Read as the sensation in the linked same-temperature comparison, not denial that metal can physically be cold.
- CONFIRMED | "A hanging chain is a catenary, not the parabola everyone draws" | https://www.damtp.cam.ac.uk/user/reh10/lectures/nst-mmii-handout3.pdf | [related-chain] UNREADABLE resolved: read both Cambridge PDF pages. The uniform freely hanging chain has a shifted/scaled cosh solution, explicitly called a catenary. This supports the usual ideal-chain statement.
- CONFIRMED | "Below is a real wing in a real flow" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [real-wing-real-flow] The very same sentence immediately specifies "the exact, closed-form solution for an idealised airfoil". In context it asserts a physical model rather than a photographed real aircraft.
- CONFIRMED | "The dye is not animated to a script" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [dye-not-script] The claim contrasts computed advection with prescribed tracks. stepDye evaluates velAt at RK4 stages. Its post-arrival freeze is a separate limitation, not evidence of pre-scripted transit times.
- CONFIRMED | "the tear is computed rather than animated." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html | [tear-computed] The motion to the edge is computed from the velocity field; no stored timeline prescribes the lead. The numerical integration is approximate, while the velocity expression is analytic.
- CONFIRMED | "An immersive layer; it renders as its own standalone page at the `href` above" | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html; git show f5e5f84f34:src/pages/strata/[...slug].astro:12 | [md-standalone] Live URL returns the standalone HTML. The Astro route excludes entries with href using "!s.data.href".
- CONFIRMED | "its body is not rendered." | git show f5e5f84f34:public/strata/the-air-that-got-there-first/index.html; git show f5e5f84f34:src/pages/strata/[...slug].astro:12 | [md-body-not-rendered] The route filters "!s.data.href" and the served page is the standalone HTML, establishing that this MD body is not rendered.
- CONFIRMED | "a wing's lift is circulation and downwash, not a longer-path debt." | https://www1.grc.nasa.gov/beginners-guide-to-aeronautics/bernoulli-and-newton/ | [related-wing-lift] NASA supports equivalent flow-turning/momentum and pressure accounts and rejects the longer-path premise.
- CONFIRMED | "the theorem L′ = ρ·U·Γ is Kutta–Joukowski" | https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA.htm; https://web.mit.edu/13.021/13021_2003/Lifting%20surfaces/lectureA_files/image022.gif | [theorem-kutta-joukowski] Read MIT's theorem equation image as well as its prose. MIT writes L=-rho U Gamma using counterclockwise circulation; the page takes positive Gamma clockwise and writes the equivalent positive-sign formula. No sign discrepancy after accounting for convention.
- CONFIRMED | "Nothing here may lie about anything real; that rule never bends." | git show f5e5f84f34:WAKE.md | [site-rule] WAKE rule 2 is "It must never lie about anything real." The sentence states a rule, not a guarantee that audits can never find mistakes.
- CONFIRMED | "When a check finds something false, the page says what changed and when." | git show f5e5f84f34:assay/README.md; git show f5e5f84f34:src/content/strata/no-field-for-the-repair.md:54 | [site-corrections] Assay policy requires a dated correction or update. A pinned example, src/content/strata/no-field-for-the-repair.md:54, begins "Corrected 2026-09-27: the dek said the 22,747 reversals". The working policy and example support the statement.
- CONFIRMED | (added) "there cosh, here the Kutta-Joukowski Gamma" | https://www.damtp.cam.ac.uk/user/reh10/lectures/nst-mmii-handout3.pdf; git show f5e5f84f34:research/the-air-that-got-there-first/verify.mjs | [added-cosh-comparison] The omitted mathematical comparison is in the MD relates note, line 15. Cambridge solves the chain as a shifted/scaled cosh; the pinned verifier defines Gamma by the Kutta condition and uses L_KJ = rho * U * Gamma(alpha). Positive coverage control: the record contains zero cosh claims but one "A hanging chain is a catenary" claim.

## Findings

- [declined] X1 (MINOR): left the nonspecific journal-homepage citation unchanged under triage rule 4, tier C. It is an imprecise supporting link, not a false physics statement; its sentence is separate from both WRONG corrections.
- [declined] X2 (MINOR): left the omitted post-arrival stopping qualification unchanged under triage rule 4, tier C. Its sentence is separate from X3 even though both are in one paragraph; the same-sentence exception does not apply.
- [open note:20e0c4] X3 (WRONG): replaced the negative-angle reversal claim with the supported behavior at -4 degrees and the model zero-lift angle of about -5.19 degrees. Corrected the layer build-log repetition and removed its proposed film reference to an available AoA flip. Added a dated correction on the page and mirrored it at the end of the markdown body. The shared archive memory/ledger-archive/P19.md:15 still proposes the unavailable AoA flip; the directing instance must remove that copy and replace PENDING with the board-issued note ID.
- [declined] X4 (MINOR): left the offline panel-recomputation overclaim unchanged under triage rule 4, tier C. The browser and verifier share equations but differ in sampling and release offsets; this sentence is not part of a WRONG correction.
- [declined] X5 (MINOR): left forever unchanged under triage rule 4, tier C. It is loose wording within a correct no-stall discussion; lift rises only across the offered slider range, and the sentence is separate from both WRONG corrections.
- [declined] X6 (MINOR): left the footer checking-scope overclaim unchanged under triage rule 4, tier C. It repeats X4, X9 and X11, and is separate from the false browser-crosscheck attribution fixed in the dek.
- [open note:20e0c4] X7 (WRONG): changed the dek from the page to the offline verifier as the program computing lift both ways, and corrected the same attribution in the layer build log. The supported rounded 0.000% agreement remains. Added a dated correction on the page and mirrored it at the end of the markdown body. No false verifier assertion was found for either WRONG item, so the verifier is unchanged. The shared archive memory/ledger-archive/P19.md:15 still attributes both lift calculations to the page; the directing instance must credit the offline verifier there and replace PENDING with the board-issued note ID.
- [declined] X8 (MINOR): left Every number recomputed live unchanged under triage rule 4, tier C. It is a separate sentence and metadata field from X7; only named panel quantities are updated live.
- [declined] X9 (MINOR): left the markdown sentence unchanged under triage rule 4, tier C. Its 1.34 is the actual browser estimate, so replacing it with the verifier estimate 1.36 would misdescribe the panel. The fault is the shared offline-recomputation promise, separate from the dek attribution.
- [dated] X10 (STALE): advanced JSON-LD dateModified from 2026-06-29 to 2026-10-14 for this revision. Applied the explicit JSON-LD exception: the two Corrected lines date the revision, so no Updated or Corrected line was added for this metadata item.
- [declined] X11 (MINOR): left the markdown all-numbers promise unchanged under triage rule 4, tier C. It repeats the offline checking-scope issue in X4, X6 and X9, rather than adding a new false numerical result, and its sentence is separate from the WRONG dek attribution.