Skip to content

research(cantor-diagonalization-oq-01-oq-01-oq-02-oq-01): S5 — Easton non-examples + lt_apply corollary#16936

Draft
rjwalters wants to merge 1 commit into
mainfrom
research/cantor-easton-non-examples-20260508-082612
Draft

research(cantor-diagonalization-oq-01-oq-01-oq-02-oq-01): S5 — Easton non-examples + lt_apply corollary#16936
rjwalters wants to merge 1 commit into
mainfrom
research/cantor-easton-non-examples-20260508-082612

Conversation

@rjwalters
Copy link
Copy Markdown
Owner

Summary

Session 5 rounding-out of the Easton-function structure in proofs/Proofs/CantorDiagonalizationOQ01OQ01OQ02OQ01.lean. Adds 3 small theorems (one corollary + 2 non-examples). The 2 axioms (easton_permitted_realizable, easton_consistency, both True-codomain placeholders) and 0 sorries are unchanged.

Three new theorems:

  1. IsEastonFunction.lt_apply — every Easton function is strictly increasing on regular cardinals ≥ ℵ₀. Direct corollary of succ_le and Order.lt_succ:

    theorem IsEastonFunction.lt_apply {F : Cardinal.{0} → Cardinal.{0}}
        (hF : IsEastonFunction F) {κ : Cardinal.{0}}
        (hreg : κ.IsRegular) (hℵ₀ : ℵ₀ ≤ κ) : κ < F κ :=
      lt_of_lt_of_le (Order.lt_succ κ) (hF.succ_le κ hreg hℵ₀)
  2. id_not_isEastonFunction — the identity fun κ => κ is NOT an Easton function. Proof: lt_apply at ℵ₀ would require ℵ₀ < ℵ₀, contradicting lt_irrefl.

  3. const_aleph0_not_isEastonFunction — the constant fun _ => ℵ₀ is NOT an Easton function. Same proof pattern.

Why this is real progress (and the limit thereof)

  • Non-vacuity grounded with non-examples: previously isEastonFunction_continuum (existence) and isEastonFunction_nonempty (extracted ∃) showed the constraint set is non-empty; the new non-examples confirm it's not vacuously satisfied by every endofunction. The structure is genuinely a constraint.
  • Reusable API: IsEastonFunction.lt_apply exposes the strict-increase property as a single lemma any future work can call directly rather than re-deriving from succ_le + Order.lt_succ each time.
  • No axiom/sorry delta: this is a structural rounding-out, not Phase-3b work. The 2 True-codomain axioms (easton_permitted_realizable, easton_consistency) remain placeholders pending the genuine ConsistencyOf predicate (estimated 1000+ lines for Gödel-encoded ZFC formulas, or class-forcing infrastructure).

Build status: pending

The worktree's proofs/.lake is a recursive self-symlink (per feedback_researcher_lake_symlink_broken.md), forcing every Docker build to fresh-clone Mathlib (~10–15 min) + cache get (~10 min) — total ~45 min. Following the convention from birthday-OQ-03-OQ-01-OQ-02-OQ-01 PRs #16777, #16837, #16873 (all build-pending drafts), this PR is opened as draft.

The 3 added theorems use only basic structural tactics already exercised elsewhere in the same file:

  • lt_of_lt_of_le (standard order combinator)
  • Order.lt_succ (used at lines 105, 115, 124, 140 of this file)
  • lt_irrefl _ : ¬ a < a (basic Mathlib)
  • Cardinal.isRegular_aleph₀ (standard aleph₀ regularity instance)
  • le_rfl (trivial)

Risk of build failure is low; review of the proof body should be straightforward by inspection.

Files Modified

  • proofs/Proofs/CantorDiagonalizationOQ01OQ01OQ02OQ01.lean (+35 lines: 3 theorems + header update + 3 #check directives; 257→292 total)
  • src/data/proofs/cantor-diagonalization-oq-01-oq-01-oq-02-oq-01/meta.json (lineCount 257→292, theoremCount 7→10 in both top-level meta and leanFile; assumptions text expanded to mention the 3 new theorems)
  • src/data/research/problems/cantor-diagonalization-oq-01-oq-01-oq-02-oq-01.json (Session 5 entries in builtItems/insights/progressSummary; iteration 4→5)
  • research/problems/cantor-diagonalization-oq-01-oq-01-oq-02-oq-01/knowledge.md (Session 5 narrative)

Test plan

  • Re-run ./proofs/scripts/docker-build.sh Proofs.CantorDiagonalizationOQ01OQ01OQ02OQ01 from a worktree with warm Mathlib cache
  • Confirm IsEastonFunction.lt_apply, id_not_isEastonFunction, const_aleph0_not_isEastonFunction type-check
  • Confirm new #check directives at end of file resolve

Related

🤖 Generated by researcher-11

… non-examples + lt_apply corollary (build pending)

Adds 3 small theorems to CantorDiagonalizationOQ01OQ01OQ02OQ01.lean rounding
out the Easton-function picture. Existing 2 axioms (placeholder True codomain
pending Phase-3b ConsistencyOf predicate) and 0 sorries unchanged.

- IsEastonFunction.lt_apply: every Easton F is strictly increasing on regular
  cardinals ≥ ℵ₀ (corollary of succ_le + Order.lt_succ).
- id_not_isEastonFunction: identity κ ↦ κ violates lt_apply at ℵ₀.
- const_aleph0_not_isEastonFunction: constant κ ↦ ℵ₀ violates lt_apply at ℵ₀.

The non-examples ground the structure: previously isEastonFunction_continuum
(existence) and isEastonFunction_nonempty (extracted ∃) showed the constraint
set is non-empty; now id and const non-examples confirm it's not vacuously
satisfied. Proofs use only basic structural tactics (lt_of_lt_of_le,
Order.lt_succ, lt_irrefl, Cardinal.isRegular_aleph₀, le_rfl).

File: 257→292 lines, 7→10 theorems. meta.json + leanFile counts synced.

Build status: pending. The worktree's proofs/.lake is a recursive self-symlink
(per feedback_researcher_lake_symlink_broken.md), forcing every Docker build to
fresh-clone Mathlib + cache get (~45 min). Following the convention of
birthday-OQ-03 PRs #16777/#16837/#16873 (build pending, deferred to follow-up).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
@rjwalters rjwalters added the research Research agent work label May 8, 2026
@rjwalters rjwalters changed the title research(cantor-diagonalization-oq-01-oq-01-oq-02-oq-01): S5 — Easton non-examples + lt_apply corollary (build pending) research(cantor-diagonalization-oq-01-oq-01-oq-02-oq-01): S5 — Easton non-examples + lt_apply corollary May 8, 2026
@rjwalters
Copy link
Copy Markdown
Owner Author

Build verified locally via Docker (./proofs/scripts/docker-build.sh Proofs.CantorDiagonalizationOQ01OQ01OQ02OQ01): all 3060 jobs succeeded; #check output confirms 14 declarations (12 theorems + 2 axioms). researcher-11 2026-05-08.

rjwalters added a commit that referenced this pull request May 13, 2026
…problem.md + state.md (doc-only) (#18807)

The Lean file `proofs/Proofs/CantorDiagonalizationOQ01OQ01OQ02OQ01.lean`
has been in a stable axiomatized rest state since S4 (researcher-8,
2026-05-08): 0 sorries, 2 axioms with `True` placeholder codomain,
status="axiomatized"/badge="axiom". But `research/problems/<slug>/`
held only `knowledge.md` in git; `problem.md` and `state.md` were
untracked working-tree stubs on the main repo, never committed.

This PR commits both files for the first time, reflecting the actual
post-S4 state instead of the seeker-init placeholder.

Two new tracked files:

- `research/problems/.../problem.md` — Easton's-theorem statement
  (plain + LaTeX), classification, related-gallery cross-references
  (parent, sibling OQ-02-OQ-03, ContinuumHypothesis, child OQ-01).

- `research/problems/.../state.md` — Phase: AXIOMATIZED — Phase-3a-fix
  COMPLETE, iteration 4 (S1–S4 shipped). Tabulates the 7 axiom-free
  supporting theorems and the 2 `True`-codomain placeholder axioms.
  Documents three forward research levers:

  - Lever A (cheap, 1–2 sessions): Phase-3b — introduce explicit
    `ConsistencyOf : (Cardinal → Cardinal) → Prop` axiom and restate
    `easton_consistency` with a meaningful codomain.
  - Lever B (1 session): bridge file proving the two-sided
    characterization with sibling OQ-02-OQ-03 (excluded values).
  - Lever C (multi-session research-track): scope a flypitch-style
    Lean 4 port for class forcing.

  Also notes three unshipped DRAFT PRs (#16936 S5, #17137 S6, #17169
  S7) that have not advanced for 5 days, so future researchers know
  these are not "shipped iterations" when planning further work.

No Lean changes. Doc-only. Build: N/A.

Co-authored-by: Robb Walters <r.j.walters@gmail.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

research Research agent work

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant