diff --git a/Makefile b/Makefile index 5b51f9c..9a971d1 100644 --- a/Makefile +++ b/Makefile @@ -878,6 +878,20 @@ textbook-peano: ## Peano Arithmetices Principia 1889 (CC-BY-SA, Verheyen+Nahas L $(ARBORIST) --db "$(CRAWL_SHARDS_DIR)/textbook_peano-arithmetices-principia-1889.db" \ ingest --source textbook_tex --url "$$url" +textbook-russell-pom: ## Russell Principles of Mathematics 1903 (PD content + CC-BY-SA-4.0 typesetting, Klement HTML) + $(MAKE) textbook ID=russell-pom-1903 + +textbook-laplace: ## Laplace Philosophical Essay on Probabilities 1814 / 1902 (PD, PG #58881) + $(MAKE) textbook ID=laplace-philosophical-essay-probabilities + +textbook-dedekind: ## Dedekind Essays on the Theory of Numbers 1901 (PD, PG #21016 TeX) + @row=$$(grep '"dedekind-essays-theory-numbers"' $(TEXTBOOK_MANIFEST) | head -1); \ + url=$$(echo "$$row" | $(PY) -c "import json,sys; print(json.loads(sys.stdin.read()).get('tex_url',''))"); \ + if [ -z "$$url" ]; then echo "no tex_url for Dedekind" >&2; exit 1; fi; \ + mkdir -p $(CRAWL_SHARDS_DIR); \ + $(ARBORIST) --db "$(CRAWL_SHARDS_DIR)/textbook_dedekind-essays-theory-numbers.db" \ + ingest --source textbook_tex --url "$$url" + test-crawler: bootstrap-crawler ## run only the lifted crawler tests $(VENV)/bin/pytest -q tests/crawler diff --git a/bench/fixtures/textbooks/manifest-v1.jsonl b/bench/fixtures/textbooks/manifest-v1.jsonl index 434260f..25fc2ec 100644 --- a/bench/fixtures/textbooks/manifest-v1.jsonl +++ b/bench/fixtures/textbooks/manifest-v1.jsonl @@ -15,3 +15,6 @@ {"id":"demorgan-first-notions-logic","title":"First Notions of Logic, Preparatory to the Study of Geometry","author":"Augustus De Morgan","year":"1839","license":"PD","license_url":"https://en.wikipedia.org/wiki/Public_domain","domain":"logic","pillar_targets":["I"],"urls":["https://www.gutenberg.org/cache/epub/67017/pg67017-images.html"],"home_url":"https://www.gutenberg.org/ebooks/67017","notes":"PG eBook #67017, single-page HTML. ~30 pages: AEIO propositions, syllogism six fundamental forms × four figures, hypothetical / indirect / a fortiori arguments, probabilistic inference. Pre-Boole, pre-Frege, pre-Hilbert; one of the earliest English systematic logic texts. Pillar I supplement to Boole 1854 already ingested."} {"id":"russell-imp-1919","title":"Introduction to Mathematical Philosophy","author":"Bertrand Russell","year":"1919","license":"PD","license_url":"https://en.wikipedia.org/wiki/Public_domain","domain":"logic-foundations","pillar_targets":["I","II","III"],"urls":["https://www.gutenberg.org/cache/epub/41654/pg41654-images.html"],"home_url":"https://www.gutenberg.org/ebooks/41654","notes":"PG eBook #41654, single-page HTML. ~200 pages covering natural numbers, definition of number, finite & infinite, classes & types, propositional functions, descriptions, relations, axiom of infinity, axiom of choice. Bridges pillars I (logic), II (transfinite cardinals), III (Peano arithmetic). Russell wrote IMP after Principia Mathematica as a non-symbolic exposition; PD by age (1919 + URAA = 95 years = 2015)."} {"id":"peano-arithmetices-principia-1889","title":"Arithmetices Principia, Nova Methodo Exposita / The Principles of Arithmetic, Presented by a New Method","author":"Giuseppe Peano (Vincent Verheyen + Michael Nahas, transl. + ed.)","year":"1889 (orig. Latin); 2018+ (Verheyen English, current revisions through 2022+)","license":"CC-BY-SA-4.0","license_url":"https://creativecommons.org/licenses/by-sa/4.0/","domain":"arithmetic","pillar_targets":["III"],"urls":[],"home_url":"https://github.com/mdnahas/Peano_Book","tex_url":"https://raw.githubusercontent.com/mdnahas/Peano_Book/master/Peano.tex","notes":"Side-by-side Latin (Peano 1889) + English (Verheyen 2018) translation with modern math notation, hosted on mdnahas/Peano_Book GitHub repo under CC-BY-SA-4.0. The 1889 Latin original IS the primary source for Peano arithmetic axioms (zero, successor, induction); Landau 1929 and Gödel 1931 both rederive from this. Ingest path: textbook_tex (paracol multi-column macros leak some `gobble 1cm` / `0.5cm` residue but substantive English text extracts cleanly — 39 chunks, 35 hits on 'peano', 38 on 'arithmetic'). Pillar III is gated on a fox decision to alias Landau + Gödel citations to Peano via #000041."} +{"id":"russell-pom-1903","title":"The Principles of Mathematics","author":"Bertrand Russell (Klement HTML typesetting 2024)","year":"1903 (orig.); 2024 (Klement HTML reformat)","license":"CC-BY-SA-4.0","license_url":"https://creativecommons.org/licenses/by-sa/4.0/","domain":"logic-foundations","pillar_targets":["I","II"],"urls":["https://people.umass.edu/klement/pom/pom.html"],"home_url":"https://people.umass.edu/klement/pom/","notes":"Kevin C. Klement's HTML typesetting of Russell's 1903 *Principles of Mathematics*. Russell's text is PD (1903 + URAA expired); Klement's typesetting is CC-BY-SA-4.0. Single-page HTML (~25K lines / ~1MB). Russell PoM is more class-theoretic and ZFC-foundational than IMP 1919 (PoM was the precursor to Principia Mathematica). Pillar I/II coverage extension: contains Russell's Paradox derivation, class theory, relations, magnitudes — what IMP and PM leave for the technical reader. Cited by claim-pack as Mendelson/Enderton substrate."} +{"id":"laplace-philosophical-essay-probabilities","title":"A Philosophical Essay on Probabilities","author":"Pierre-Simon Laplace (Truscott + Emory transl. 1902)","year":"1814 (French orig.); 1902 (English Truscott-Emory)","license":"PD","license_url":"https://en.wikipedia.org/wiki/Public_domain","domain":"probability","pillar_targets":["V"],"urls":["https://www.gutenberg.org/cache/epub/58881/pg58881-images.html"],"home_url":"https://www.gutenberg.org/ebooks/58881","notes":"PG eBook #58881, single-page HTML. Laplace 1814 *Essai philosophique sur les probabilités* is the foundational popular treatise on probability — predates Kolmogorov 1933 by 119 years but covers expected value, conditional probability, additive law, large-numbers reasoning, applications to natural philosophy. PD by age (Truscott-Emory English 1902 = pre-1929 US PD). Pillar V substrate; Kolmogorov 1933 axiomatizes what Laplace introduces narratively."} +{"id":"dedekind-essays-theory-numbers","title":"Essays on the Theory of Numbers (I. Continuity and irrational numbers, II. The nature and meaning of numbers)","author":"Richard Dedekind (Wooster Woodruff Beman, transl. 1901)","year":"1872 + 1888 (German orig.); 1901 (English Beman)","license":"PD","license_url":"https://en.wikipedia.org/wiki/Public_domain","domain":"arithmetic","pillar_targets":["III"],"urls":[],"home_url":"https://www.gutenberg.org/ebooks/21016","tex_url":"https://www.gutenberg.org/files/21016/21016-t/21016-t.tex","notes":"PG eBook #21016 — TeX-only (no HTML edition); ingest via textbook_tex pipeline (proven against Hilbert + Boole + Peano). Dedekind is the OTHER pillar III primary alongside Peano: 'Was sind und was sollen die Zahlen?' (1888) introduced the chains-and-induction definition of natural numbers; 'Stetigkeit und irrationale Zahlen' (1872) is the cuts construction of the reals. Beman 1901 English. PD by age. Closes the induction / completeness side of Landau citations that Peano alone misses."} diff --git a/bench/results/three-more-textbooks-2026-05-10.md b/bench/results/three-more-textbooks-2026-05-10.md new file mode 100644 index 0000000..533aa96 --- /dev/null +++ b/bench/results/three-more-textbooks-2026-05-10.md @@ -0,0 +1,140 @@ +# Three more textbooks (Russell PoM + Laplace + Dedekind) — 2026-05-10 + +**Date:** 2026-05-10 (UTC) +**Author:** blackops + fox +**Method:** under fox "keep filling maths in?" directive: ingested +three more PD/CC-BY-SA primary substrates, registered five new +citation aliases, ran resolver. Pillar II nearly clear. + +## Headline + +**54 → 62 / 92 records resolve (+8, 67% coverage).** Pillar I and II +both jumped substantially: + +| Pillar | Pre-turn | Post-turn | +|---|---|---| +| I (logic) | 7/13 | **10/13** | +| II (set theory) | 5/10 | **9/10** | +| III (arithmetic) | 7/13 | 7/13 | +| IV (geometry) | 18/18 | 18/18 | +| V (probability) | 0/5 | **1/5** | +| VI (classical phys) | 3/5 | 3/5 | +| VII (combinatorics) | 9/14 | 9/14 | +| IX (lambda + algebra) | 5/14 | 5/14 | +| **Total** | **54/92 (59%)** | **62/92 (67%)** | + +## Three new ingests + +### Russell *Principles of Mathematics* 1903 (Klement HTML) + +- Manifest `id: russell-pom-1903`, license `CC-BY-SA-4.0` (Russell's + text PD; Klement's HTML typesetting CC-BY-SA-4.0). +- Source: `https://people.umass.edu/klement/pom/pom.html` — single + ~1MB HTML page, 25K lines. +- Landed: **1 doc / 546 chunks** (largest single shard yet). +- Russell PoM is the technical precursor to *Principia Mathematica*. + Covers class theory, relations, magnitudes, Russell's Paradox + derivation — the formalism IMP 1919 leaves to the technical + reader. + +### Laplace *A Philosophical Essay on Probabilities* (PG #58881 HTML) + +- Manifest `id: laplace-philosophical-essay-probabilities`, license + `PD` (Laplace 1814 + Truscott-Emory 1902 English). +- Single-page PG HTML. +- Landed: **1 doc / 111 chunks**. +- Pre-Kolmogorov foundational treatise: expected value, conditional + probability, additive law, large numbers, applications. Pillar V + substrate — Kolmogorov 1933 axiomatizes what Laplace introduces. + +### Dedekind *Essays on the Theory of Numbers* (PG #21016 TeX) + +- Manifest `id: dedekind-essays-theory-numbers`, license `PD` + (Dedekind 1872 + 1888 + Beman 1901 English). +- Source: PG TeX (no HTML edition). Ingested via existing + `textbook_tex` pipeline (4th TeX-source shard after Hilbert, + Boole, Peano). +- Landed: **1 doc / 57 chunks**. +- Two essays: *Continuity and Irrational Numbers* (Dedekind cuts + for the reals) + *The Nature and Meaning of Numbers* (chains-and- + induction definition of natural numbers). Pillar III substrate + alongside Peano. + +## Five new citation aliases + +`decision_by="fox 2026-05-10"` per #000041 audit discipline. + +| Original | Substitute | Records lifted (this turn) | +|---|---|---| +| Mendelson — Introduction to Mathematical Logic | Russell PoM 1903 | +2 (pillar I) | +| Enderton — A Mathematical Introduction to Logic | Russell PoM 1903 | +1 (pillar I) | +| Jech — Set Theory | Russell PoM 1903 | +4 (pillar II — Russell's Paradox) | +| Landau — Foundations of Analysis | Dedekind *Essays* | 0 (cascade miss) | +| Kolmogorov — Foundations of Probability | Laplace 1814 | +1 (pillar V) | + +## Honest gaps (30 records remaining) + +| Pillar | Stuck | Why | +|---|---|---| +| I (3) | Existential Introduction, Vacuous Quantification, Double Negation Elimination | formal proof theory axioms; Russell PoM/IMP/Boole/De Morgan don't axiomatize natural deduction explicitly. Hilbert-Ackermann 1928 would close these (not yet ingested). | +| II (1) | Russell's Paradox Resolution | Russell PoM derives the paradox but doesn't axiomatize the type-theoretic resolution. Whitehead-Russell PM Vol I would close (not ingested). | +| III (6) | Axiom of Zero, Distinctness of Zero, Injectivity of Successor, Induction Schema, Commutativity of Addition, Distributivity | Peano + Dedekind cascade misses these; both write symbolically (`0 ∈ N`, not "axiom of zero"). Cascade tuning needed. | +| V (4) | Kolmogorov's three axioms, Inclusion-Exclusion | Laplace doesn't axiomatize σ-algebras; only narrative coverage. Borel 1909 + Kolmogorov 1933 itself blocked (German URAA-restored to 2058). | +| VI (2) | Newton's Third Law, Conservation of Momentum | Newton's 1729 prose vocabulary mismatch. Term-aliases registered but cascade tokens don't match. | +| VII (5) | Empty-Set/Boundary, Hockey-Stick, Vandermonde, Catalan, Stars-and-Bars | Bogart CTGD skews intro-level; advanced enumerative identities not named with these moderns labels. | +| IX (9) | β/α/η-Conversion, Extensionality (λ), Church-Rosser, Fixed-Point, Turing Completeness, Böhm-Jacopini, Orbit-Stabilizer | Lambda calculus has no PD primary ingested (Church 1932-1941 blocked until 2027-2031). Pierce *Software Foundations* + Wadler *Programming Language Foundations in Agda* are CC-BY-4.0 candidates worth investigating. Orbit-Stabilizer is a Judson cascade miss. | + +## Reproducibility + +```bash +make textbook-russell-pom # Klement HTML (CC-BY-SA-4.0 typesetting) +make textbook-laplace # PG #58881 HTML (PD) +make textbook-dedekind # PG #21016 TeX (PD) + +# Title backfill (one-off; fix in source pipeline tracked under #000031): +python -c " +import sqlite3 +for shard, t in [ + ('~/.arborist/crawl/textbook_russell-pom-1903.db', + 'The Principles of Mathematics, by Bertrand Russell (1903; Klement HTML 2024)'), + ('~/.arborist/crawl/textbook_laplace-philosophical-essay-probabilities.db', + 'A Philosophical Essay on Probabilities, by Pierre-Simon Laplace (Truscott + Emory transl. 1902)'), + ('~/.arborist/crawl/textbook_dedekind-essays-theory-numbers.db', + 'Essays on the Theory of Numbers (Continuity and Irrational Numbers + The Nature and Meaning of Numbers), by Richard Dedekind (Beman transl. 1901)'), +]: + c = sqlite3.connect(__import__('os').path.expanduser(shard)) + c.execute('UPDATE documents SET title = ?', (t,)) + c.commit() +" + +arborist warrant-resolve --use-aliases --write +# → 62 / 92 (67%) +``` + +## Cumulative session lift + +| Snapshot | Resolved | % | Drivers | +|---|---|---|---| +| Start of 2026-05-10 | 18/92 | 20% | Hilbert pillar IV only | +| After 3 high-conf aliases | 33/92 | 36% | Goldstein/Newton, Stanley/Bogart, Dummit-Foote/Judson | +| After Newton recrawl | 35/92 | 38% | deeper Principia text | +| After 7 medium-conf aliases | 39/92 | 42% | Mendelson/Enderton/Jech/Landau substitutes | +| After Peano + title backfill | 54/92 | 59% | Peano arithmetic + author-surname haystack fix | +| After Russell PoM + Laplace + Dedekind | **62/92** | **67%** | this turn | + +47% absolute coverage gain in one day, mostly through fox-decided +citation-alias work + 7 new PD/CC textbook ingests. + +## References + +- Prior bench journals (chronological today): + - `bench/results/citation-aliases-batch-2026-05-10.md` + - `bench/results/medium-confidence-aliases-and-peano-2026-05-10.md` + - this file: `bench/results/three-more-textbooks-2026-05-10.md` +- `#000041` — citation-aliases mechanism (15 rows live) +- `#000042` — term-aliases mechanism (6 rows live) +- `#000031` — surface-ingest cited textbooks (Phase 1 follow-up + needed: source-side title-from-author backfill) +- `#000038` — Phase 4 content acquisition (Hilbert-Ackermann + + Whitehead-Russell PM + Software Foundations / Wadler are top + next candidates)