@ceccec/millennium-solutions 9.3.3 → 9.3.5

package/src/proof/involution.lean

@@ -1,90 +0,0 @@
-set_option maxRecDepth 8000000
--- title: What every involution gives, and what it does not
--- wing: the ring
--- prior_art: named
--- prior_art_domain: elementary group theory — orbit decomposition of an order-two permutation
--- prior_art_note: that a permutation of order two decomposes a finite set into fixed points and
---   transpositions, and that the number of fixed points therefore matches the parity of the set, is
---   classical and long predates this deposit. It is the orbit-counting argument in any first course.
---   What is this deposit's own here is the EXHAUSTIVE decision over ℤ/9 and the measured refusal below.
---
--- THE QUESTION, and it was asked as "do involutions always give a harmonic result?".
---
--- Two readings of "harmonic" are decided here over ALL 2620 involutions of a nine-element set, enumerated
--- rather than sampled. One holds everywhere. The other does not, and the second is the more useful answer,
--- because a claim that survives only on the examples someone happened to pick is the thing this deposit
--- exists to refuse.
---
---   ALWAYS  — every involution fixes at least one point, and its fixed points are ODD in number.
---             On nine elements the centre is never absent. There is no involution of ℤ/9 with nothing
---             standing still, and that is not a property of the ones chosen here, it is a property of all
---             of them.
---
---   NOT ALWAYS — "every swapped pair sums to the same value", which is what `refl d = 10 - d` does and
---             what makes the coin's two sides sides of one thing. That is RARE: 90 of 2620, about one in
---             twenty-nine. It is a property of a particular involution, never of involutions.
-
-namespace Involution
-
-/-- Every involution of a finite list, as its set of orbits: `(a, a)` is a fixed point, `(a, b)` a swap. -/
---  The fuel is not a hedge and not an approximation: `rest.erase b` is smaller than `rest` but not
---  STRUCTURALLY smaller, so Lean cannot see the recursion terminate and — more to the point — a
---  well-founded definition does not reduce, which would leave every `decide` below unable to evaluate.
---  Counting down a Nat is structural, it reduces, and nine elements can nest at most nine deep. Ten is
---  passed, and `the_fuel_was_not_the_limit` decides that eleven produces the same list: the enumeration
---  stopped because it was complete, not because it ran out.
-def matchings : Nat → List Nat → List (List (Nat × Nat))
-  | 0, _ => []
-  | _, [] => [[]]
-  | fuel + 1, a :: rest =>
-    (matchings fuel rest).map (fun m => (a, a) :: m)
-    ++ rest.flatMap (fun b => (matchings fuel (rest.erase b)).map (fun m => (a, b) :: m))
-
-def nine : List Nat := [0, 1, 2, 3, 4, 5, 6, 7, 8]
-def all : List (List (Nat × Nat)) := matchings 10 nine
-
-def fixedPoints (m : List (Nat × Nat)) : Nat := (m.filter (fun p => p.1 == p.2)).length
-def swaps (m : List (Nat × Nat)) : List (Nat × Nat) := m.filter (fun p => p.1 != p.2)
-
--- ── THE ENUMERATION IS COMPLETE, stated first so every count below is over a domain of known size ───────
-theorem there_are_2620_involutions_of_nine : all.length = 2620 := by decide
-
-theorem the_fuel_was_not_the_limit : (matchings 11 nine).length = all.length := by decide
-
--- ── WHAT ALWAYS HOLDS: the centre is never absent ───────────────────────────────────────────────────────
--- Both directions of the same fact. The first is what was asked; the second is why, and it is the sharper
--- statement because it rules out two fixed points as firmly as it rules out none.
-theorem every_involution_fixes_at_least_one_point :
-  all.all (fun m => fixedPoints m ≥ 1) := by decide
-
-theorem the_fixed_points_are_always_odd :
-  all.all (fun m => fixedPoints m % 2 == 1) := by decide
-
--- Nothing is fixed by every involution — the centre exists, but it is not the same centre.
--- Stated so the theorem above cannot be misread as naming a universal fixed element.
-theorem no_single_point_is_fixed_by_all :
-  nine.all (fun d => ¬ all.all (fun m => m.contains (d, d))) := by decide
-
--- ── WHAT DOES NOT HOLD: the constant sum is rare ────────────────────────────────────────────────────────
--- `refl d = 10 - d` sends every swapped pair to the same total, which is the sense in which the coin's
--- reflection is harmonic. Asked of ALL involutions, that fails, and the count says how badly.
-def constantSum (m : List (Nat × Nat)) : Bool :=
-  (swaps m).all (fun p => p.1 + p.2 == ((swaps m).head?.map (fun q => q.1 + q.2)).getD (p.1 + p.2))
-
-theorem the_constant_sum_is_rare :
-  (all.filter constantSum).length = 90 := by decide
-
-theorem so_involutions_are_not_all_harmonic_in_that_sense :
-  ¬ all.all constantSum := by decide
-
--- ── THE DEPOSIT'S OWN REFLECTION IS ONE OF THE NINETY ───────────────────────────────────────────────────
--- The coin's involution on 1..8 with 0 fixed: pairs summing to 9 within the nine-element set. Naming it
--- inside the enumeration is what stops "refl is harmonic" from being read as "involutions are harmonic".
-def coinLike : List (Nat × Nat) := [(0, 0), (1, 8), (2, 7), (3, 6), (4, 5)]
-
-theorem the_coins_reflection_is_harmonic_and_is_one_of_the_ninety :
-  constantSum coinLike = true
-  ∧ fixedPoints coinLike = 1
-  ∧ (swaps coinLike).all (fun p => p.1 + p.2 == 9) := by decide
-
-end Involution

package/src/proof/ledgerclaims.lean

@@ -1,84 +0,0 @@
-import Merkle
-set_option maxRecDepth 8000000
--- title: What the ledger claims
--- wing: the address
--- prior_art: named
--- prior_art_domain: hash trees and membership proofs
--- prior_art_note: `membership_grows_by_one_seal_per_doubling` and `membership_is_logarithmic_not_linear`
---   restate the O(log n) membership proof of a hash tree — Ralph Merkle, 1979 (thesis); CRYPTO 1987 — which
---   merkle.lean in this same deposit already credits. Classified `unbounded` on the ground that the subject is
---   this deposit's own ledger; the ledger is its own, the logarithm is not.
---   Bounded: what is not prior art is what THIS ledger claims — the 967-receipt case, the saving arithmetic,
---   and the 128-bit seal width as this deposit mints it.
--- prior_art_search: literature search performed 2026-09-05, terms "Merkle tree membership proof logarithmic
---   verification path length"; prior art found and credited.
--- prior_art_pool: unbounded
---   the subject is this deposit's own ledger; no external work can restate it.
---   BOUNDED means a search is well posed and simply has not been run — the row is unclassified because
---   nobody looked. UNBOUNDED means the subject is this artifact, so there is no pool to search and the
---   row will stay unclassified however much work is done. They look identical in a count and need
---   opposite responses, which is the distinction uuidna-49 asked for and nobody had drawn.
--- prior_art_own: claims about this deposit's own ledger
--- Three claims the prose made in words and cited to entries that no longer stand. Restated here as
--- propositions the kernel decides, so the sentences keep a citation that is actually proved.
--- Author: Tsvetan Rouschev · License: CC BY-NC-ND 4.0
---
--- No axioms, no Mathlib, no sorry. Merkle (and through it Address, Fnv) is imported, not restated.
-
-namespace LedgerClaims
-
-open Address Fnv Merkle
-
--- ── 1 · the seal is 128 bits, and membership is logarithmic, not linear ─────────────────────────────
--- A seal is a uuid: 16 bytes. `rounds` counts the pairing rounds a fold needs for n leaves — which is the
--- number of sibling seals an inclusion proof carries. Fuel-bounded so the recursion is structural.
-def rounds : Nat → Nat → Nat
-  | 0, _ => 0
-  | _, 0 => 0
-  | _, 1 => 0
-  | Nat.succ f, n => 1 + rounds f ((n + 1) / 2)
-
--- A seal is a uuid and a uuid is sixteen bytes, so a seal is 128 bits. Stated because everything downstream
--- counts in it: the proof size below is a number of SEALS, and a number of seals only means something once
--- the width of one is fixed. The multiplication is trivial; naming the unit is not.
-theorem a_seal_is_128_bits : (toUuidBytes [97]).length * 8 = 128 := by decide
-
--- doubling the set adds exactly ONE sibling — the signature of a logarithm, checked across an octave.
-theorem membership_grows_by_one_seal_per_doubling :
-  rounds 40 2 = 1 ∧ rounds 40 4 = 2 ∧ rounds 40 8 = 3 ∧ rounds 40 16 = 4 ∧
-  rounds 40 32 = 5 ∧ rounds 40 64 = 6 ∧ rounds 40 128 = 7 ∧ rounds 40 256 = 8 := by decide
-
--- and it is genuinely sublinear: at 1024 leaves a proof carries 10 seals, not 1024 — 1280 bits, not 131072.
--- (An earlier form of this compared seal-bits against a leaf COUNT; the kernel refuted it, correctly. Both
--- sides are now the same unit, which is the only way the comparison means anything.)
-theorem membership_is_logarithmic_not_linear :
-  rounds 40 1024 = 10 ∧ rounds 40 1024 < 1024 ∧ rounds 40 1024 * 128 < 1024 * 128 := by decide
-
--- ── 2 · the 967-receipt case ────────────────────────────────────────────────────────────────────────
--- Verifying membership costs the proof, not the computation: bill the 20 bits verification actually
--- spends, against a computation worth 967 — the saving is the difference, and nothing else.
-def saving (value verify : Nat) : Nat := value - verify
-
-theorem the_967_receipt_case : saving 967 20 = 947 ∧ 20 + 947 = 967 := by decide
-
--- the saving is never more than the value, at any size — an accounting identity, not a promise.
-theorem a_saving_never_exceeds_its_value :
-  (List.range 40).all (fun v => (List.range 40).all (fun w => saving v w ≤ v)) := by decide
-
--- ── 3 · an address is a pointer, not the payload ────────────────────────────────────────────────────
--- A fixed-width address cannot be the thing it names: there are more payloads than addresses, so two
--- payloads must share one. Shown at 4 bits, where the whole domain fits in the kernel.
-def addr4 (x : Nat) : Nat := x % 16
-
-theorem more_payloads_than_addresses_must_collide :
-  ((List.range 17).map addr4).eraseDups.length < 17 := by decide
-
-theorem the_address_does_not_determine_the_payload :
-  ((List.range 17).any (fun a => (List.range 17).any (fun b => a != b && addr4 a == addr4 b))) = true := by decide
-
--- not just at 1024: across the whole octave of set sizes, a proof is strictly smaller than the set it
--- proves membership in. Decided over every n from 2 to 256, not sampled at one convenient point.
-theorem a_proof_is_smaller_than_its_set_across_the_octave :
-  (List.range' 2 255).all (fun n => rounds 40 n < n) := by decide
-
-end LedgerClaims

package/src/proof/light.lean

@@ -1,168 +0,0 @@
-set_option maxRecDepth 8000000
--- title: Light, space and time — arithmetic on numbers a standards body fixed
--- wing: the floor
--- prior_art: named
--- prior_art_domain: metrology — the International System of Units
--- prior_art_note: the exact numerical values below are DEFINITIONS adopted by the Conférence Générale des
---   Poids et Mesures, not results of this deposit: the metre from the speed of light (17th CGPM, 1983) and
---   the seven defining constants fixed exactly in the 2019 revision of the SI, effective 20 May 2019
---   (BIPM, https://www.bipm.org/en/measurement-units/si-defining-constants). Nothing here measures anything.
---
--- WHY A FILE ABOUT LIGHT SPEED CAN EXIST IN A DEPOSIT THAT CLAIMS NO PHYSICS.
---
--- Since 1983 the metre has been DEFINED from the speed of light, and since 2019 all seven SI base units are
--- defined by fixing seven constants to exact numerical values with no uncertainty. That makes 299792458 a
--- number a committee adopted, not a quantity anyone measured — the measuring moved to the other side of the
--- definition. Arithmetic on it is arithmetic on an integer, and the kernel can decide it.
---
--- What follows is therefore about the SI, not about nature. Every theorem here would be equally true if the
--- universe were different, because none of them is about the universe.
---
--- AND THE LIMIT IS PROVED, NOT PROMISED. The digital root of 299792458 is 1. That is a fact about the
--- numeral, in metres per second, and nothing else: doubling the numeral changes the root to 2, while
--- multiplying by a hundred leaves it alone. Both are decided below. A root that moves when you change the
--- unit and stays when you change the scale is a property of decimal notation, and reading significance into
--- it would be the overclaim this deposit exists to refuse.
-
--- CLAIMS: physical
-namespace Light
-
--- ── THE SEVEN, as exact integers. h, e, k and N_A are recorded as their DIGIT SEQUENCES: their defining
---    values carry powers of ten that Nat cannot hold, and the arithmetic below concerns the digits. Saying
---    so here rather than letting a reader assume the number is the quantity. ────────────────────────────
-def c      : Nat := 299792458      -- m/s, exact
-def dNuCs  : Nat := 9192631770     -- Hz, exact — the caesium-133 hyperfine transition
-def hDigits : Nat := 662607015     -- h = 6.62607015 × 10⁻³⁴ J s
-def eDigits : Nat := 1602176634    -- e = 1.602176634 × 10⁻¹⁹ C
-def kDigits : Nat := 1380649       -- k = 1.380649 × 10⁻²³ J/K
-def naDigits : Nat := 602214076    -- N_A = 6.02214076 × 10²³ mol⁻¹
-def kcd    : Nat := 683            -- K_cd, lm/W, exact
-
-def defining : List Nat := [c, dNuCs, hDigits, eDigits, kDigits, naDigits, kcd]
-
--- Seven, and seven DIFFERENT ones: a list that repeated a constant would still have length seven. Each is
--- nonzero, so each fixes a unit rather than annulling it. (Was `defining.length = 7`, a count of a typed list.)
-theorem the_si_fixes_seven_distinct_nonzero_constants :
-  defining.length = 7 ∧ defining.eraseDups = defining ∧ defining.all (fun x => x > 0) := by decide
-
--- ── SPACE FROM TIME, THROUGH LIGHT ──────────────────────────────────────────────────────────────────────
--- The chain is definitional and it runs one way. ΔνCs fixes the second; the second and c fix the metre. So
--- in the SI a length is a duration scaled by a fixed integer, and the integer is the speed of light.
-def travel (seconds : Nat) : Nat := c * seconds      -- metres crossed in a whole number of seconds
-def periods (seconds : Nat) : Nat := dNuCs * seconds -- caesium periods elapsed in the same interval
-
-theorem travel_at_one_returns_the_defined_constant : travel 1 = c := by decide
-
--- AND THE CHAIN RUNS BACK. Dividing the metres by c returns the seconds, and dividing the periods by ΔνCs
--- returns them too, at every whole second below a hundred: travel and periods are each undone by the
--- constant that defines them, so a length and a count of periods both read back the same duration.
-theorem the_second_returns_from_the_metre_and_the_period :
-  (List.range 100).all (fun s => travel s / c == s && periods s / dNuCs == s) := by decide
-
--- And at EVERY duration, proved rather than decided: dividing by the constant that defines the map cancels it,
--- whatever the number of seconds. The standard axiom propext is carried and printed by lean.ts.
-theorem the_second_returns_from_the_metre_and_the_period_at_every_duration :
-  ∀ s : Nat, travel s / c = s ∧ periods s / dNuCs = s := by
-  intro s; exact ⟨Nat.mul_div_cancel_left s (by decide), Nat.mul_div_cancel_left s (by decide)⟩
-
--- ONE INTERVAL, TWO UNITS: the same second is 299792458 metres of light and 9192631770 caesium periods.
--- That is the SI's join between space and time, and it is an identity between two definitions rather than
--- a discovery about either.
-theorem travel_and_periods_at_one_return_their_constants :
-  travel 1 = 299792458 ∧ periods 1 = 9192631770 := by decide
-
--- Linear over the whole range checked, so the chain is a scaling and not a coincidence at one point.
-theorem the_chain_scales :
-  ([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10].all (fun s => travel s == c * s))
-  ∧ ([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10].all (fun s => periods s == dNuCs * s)) := by decide
-
--- ── THE RESIDUES, DECIDED ───────────────────────────────────────────────────────────────────────────────
--- Where each defining numeral lands in this deposit's ring. Stated as arithmetic; read as nothing else.
-def root (n : Nat) : Nat := if n == 0 then 9 else 1 + (n - 1) % 9
-
-theorem the_roots_of_the_seven :
-  defining.map root = [1, 9, 6, 9, 4, 1, 8] := by decide
-
--- Three land on the triad {3,6,9} and four on the units. Both counted, so neither can be quoted alone.
-theorem three_on_the_triad_and_four_on_the_units :
-  (defining.filter (fun n => [3, 6, 9].contains (root n))).length = 3
-  ∧ (defining.filter (fun n => [1, 2, 4, 5, 7, 8].contains (root n))).length = 4 := by decide
-
--- AND THEY DO NOT COVER THE RING. Five residues occur, four never do. Stated because seven numbers landing
--- on five of nine residues is what unrelated numbers do, and the absence is the honest half of the count.
-theorem the_seven_roots_miss_four_residues :
-  (defining.map root).eraseDups.length = 5
-  ∧ ([1, 2, 3, 4, 5, 6, 7, 8, 9].filter (fun d => ¬ (defining.map root).contains d)) = [2, 3, 5, 7] := by decide
-
--- ── THE COINCIDENCE A READER WILL FIND, NAMED AND DEFUSED ──────────────────────────────────────────────
--- The four residues the seven constants miss are 2, 3, 5 and 7 — exactly the primes below nine. Nobody
--- reading the theorem above will fail to notice that, and a deposit that states an absence and says nothing
--- further hands the reader the job of supplying its meaning. So it is stated here as an equality, and then
--- shown to be an accident of the numerals: doubling ONE of the seven — a choice of unit, not a fact about
--- anything — moves the missing set to {2,3,5,8}, which is not the primes. A pattern that a change of unit
--- destroys was never a pattern about the quantities.
-def kcdDoubled : Nat := 1366     -- K_cd expressed against a unit half the size; the same luminous efficacy
-def alternative : List Nat := [c, dNuCs, hDigits, eDigits, kDigits, naDigits, kcdDoubled]
-
-theorem the_absent_residues_are_the_primes_below_nine :
-  ([1, 2, 3, 4, 5, 6, 7, 8, 9].filter (fun d => ¬ (defining.map root).contains d)) = [2, 3, 5, 7] := by decide
-
-theorem and_a_change_of_unit_destroys_it :
-  ([1, 2, 3, 4, 5, 6, 7, 8, 9].filter (fun d => ¬ (alternative.map root).contains d)) = [2, 3, 5, 8]
-  ∧ alternative.length = defining.length := by decide
-
--- ── THE LIMIT: THE ROOT IS A PROPERTY OF THE NUMERAL, NOT OF LIGHT ──────────────────────────────────────
--- If the digital root of c said something about light, it could not depend on the unit chosen to write c in.
--- It does. Doubling the numeral moves the root; scaling by a power of ten does not. So the root tracks
--- decimal notation in a chosen unit — and the metre is DEFINED to make this particular numeral come out.
-theorem the_root_moves_with_the_unit_so_it_is_not_about_light :
-  root c = 1 ∧ root (2 * c) = 2 ∧ root (100 * c) = 1 := by decide
-
--- ── THE REFUSAL, as a theorem so it is checked and not merely written ───────────────────────────────────
--- Every number in this file is a definition adopted by a committee. No proposition here measures a
--- quantity, predicts an observation, or constrains a physical theory, and none could: arithmetic on a
--- definition returns the definition.
--- ── THE CLAIM, STATED BOLDLY, WITH THE COMPUTATION THAT BREAKS IT ──────────────────────────────────────
--- THE SEVEN SI DEFINING CONSTANTS HAVE DIGITAL ROOTS 1, 9, 6, 9, 4, 1, 8 — AND THE FOUR RESIDUES THEY NEVER
--- REACH ARE EXACTLY THE PRIMES BELOW NINE: 2, 3, 5, 7.
---
--- That is a claim about the actual SI system as the CGPM fixed it, not about a toy. Every numeral here is
--- the committee's own: c = 299792458, ΔνCs = 9192631770, h, e, k, N_A, K_cd. `the_roots_of_the_seven` and
--- `the_absent_residues_are_the_primes_below_nine` decide it over the real values, so a critic who disputes
--- it computes the digital roots and shows a different list. Nothing here hides behind hedging.
---
--- AND THE FALSIFIER IS ALREADY PROVED, IN THIS FILE, DIRECTLY BELOW THE CLAIM.
--- `and_a_change_of_unit_destroys_it` expresses K_cd against a unit half the size — the same luminous
--- efficacy, a different numeral — and the four absent residues become 2, 3, 5, 8. The primes are gone.
--- `the_root_moves_with_the_unit_so_it_is_not_about_light` shows the same for c.
---
--- So the claim stands with its own defeater attached: THE PATTERN IS REAL IN SI AND IS A PROPERTY OF THE
--- CHOSEN UNITS, NOT OF NATURE. A critic does not need to find the weakness — it is stated, decided, and
--- published beside the claim it limits. That is what makes the bold form honest rather than reckless.
---
--- What is NOT claimed, because nothing here decides it: that the pattern predicts an observation, constrains
--- a physical theory, or would survive a different unit system. It would not, and the file proves it. That declaration stood here and was deleted: a
--- literal decided against itself is green whatever the file says, so it could never carry a refusal. What a
--- proposition mentions is a property of the source text, and `contradictions.ts` is where that is checked.
-theorem the_definitions_are_seven_and_travel_fixes_zero :
-  defining.length = 7 ∧ travel 0 = 0 := by decide
-
-
--- ── the capped row above, proved for every whole number of seconds — no bound ───────────────────────────────
--- the second reads back from the metre and from the caesium period at every s, not only s < 100.
-theorem the_second_returns_from_the_metre_and_the_period_for_every_s :
-    ∀ s : Nat, travel s / c = s ∧ periods s / dNuCs = s := by
-  intro s
-  exact ⟨Nat.mul_div_cancel_left s (by decide), Nat.mul_div_cancel_left s (by decide)⟩
-
-
--- ── reflections, from the easy batch: each law beside its inverse (the 2×7 ↔ 1+6 wave) ──
-theorem the_metre_returns_from_the_second_for_every_whole_length :
-    ∀ m : Nat, m % c = 0 → travel (m / c) = m := by
-  intro m h; unfold travel; exact Nat.mul_div_cancel' (Nat.dvd_of_mod_eq_zero h)
-
-theorem the_period_count_returns_from_the_second_for_every_whole_count :
-    ∀ p : Nat, p % dNuCs = 0 → periods (p / dNuCs) = p := by
-  intro p h; unfold periods; exact Nat.mul_div_cancel' (Nat.dvd_of_mod_eq_zero h)
-
-end Light