Pith. sign in

REVIEW 2 major objections 4 minor 25 references

This paper claims an exhaustive enumeration and classification of triangle-maximal pseudoline arrangements for all odd n through 27, with exact wiring-diagram, Euclidean-class, and projective-class counts plus symmetry groups for each class

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-03 11:02 UTC pith:3Q4UIOR2

load-bearing objection Genuinely useful enumeration machinery with a clean completeness proof for the perfect case; the 2-defective counts are real but rest on an explicitly unproved pruning that needs proof or removal before full trust. the 2 major comments →

arxiv 2607.29236 v1 pith:3Q4UIOR2 submitted 2026-07-31 math.CO cs.CG

Enumeration and Classification of Triangle-Maximal Pseudoline Arrangements

Roman Parpalak , Denis Utkin This is my paper
classification math.CO cs.CG MSC 52C3005B3505A15
keywords pseudoline arrangementstriangle-maximalreduced wordslongest permutationperfect arrangements2-defective arrangementsprojective equivalencesymmetry groups
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper aims to settle, for odd n, which simple pseudoline arrangements maximize the number of triangular faces, and exactly how many of them exist. It proves that optimal arrangements split into two families—perfect ones for n ≡ 3, 5 mod 6 and 2-defective ones for n ≡ 1 mod 6—and gives depth-first searches over reduced words for the longest permutation that enumerate each family completely. The perfect search emits exactly one reduced word per commutation class, and the 2-defective search emits at least one representative of every projective class, so the reported tables (e.g., 85,562,064 wiring diagrams in 56,646 projective classes at n = 27) are claimed to be exact. If correct, this gives the first complete classification of these extremal arrangements in the stated ranges, and shows a triangle-maximal arrangement exists for every odd n ≤ 89 except n = 11, with n = 91 the smallest open case.

Core claim

On the paper's own terms, the central discovery is that triangle-maximality has a word-level characterization that makes exhaustive enumeration tractable: after fixing a checkerboard coloring, an optimal arrangement is encoded by a reduced word in which odd-indexed generators open black faces, and the local constraint that every such face be a triangle forces the word into a composite form built from nested triples σgσg−1σg+1. Branching only on even-indexed generators then explores exactly the reduced words that correspond to a perfect arrangement, one per commutation class; for n ≡ 1 mod 6, the 2-defective search anchors one defect at the boundary, uses a defect jump to move defects, and is

What carries the argument

Reduced words for the longest permutation w0, packaged into composite generators Kg = σgσg−1σg+1 with special variants K±g for defects; an odd-majority parity filter; the O-matrix representation that turns commutation classes into crossing-order matrices; Euclidean rotation and reflection operations and projective change-of-chart operations that define E-classes and P-classes; and the defect jump, a local move that reconnects two non-adjacent sides of a defect to relocate it. The composite form makes the search tree branch only on even-indexed generators and yields a canonical representative per commutation class, while the O-matrix machinery turns classification into deduplication and allow

Load-bearing premise

The 2-defective search's coverage of every projective class rests on an intricate case analysis—the lemma that every boundary-defect diagram is either directly emitted or is the image of an emitted diagram under a single defect jump—and the paper explicitly leaves the forced-descent pruning in that search unproved; if that case analysis has a gap, the n = 13, 19, 25 counts and the existence range could be incomplete.

What would settle it

Run an independent enumerator for 2-defective arrangements at n = 13—for example, a brute-force backtracking search over O-matrices using only the defect budget, without the composite-generator alphabet—and compare the resulting projective classes with the paper's six; any additional P-class would falsify the completeness theorem for the 2-defective search.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • The enumeration tables through n = 27 for perfect arrangements and through n = 19 for 2-defective arrangements are claimed to be exact and reproducible from the published outputs.
  • A triangle-maximal simple pseudoline arrangement exists for every odd n ≤ 89 except n = 11, where none exists; n = 91 is the first odd size left open.
  • The counts agree with previously published projective and affine counts wherever a comparison is defined, providing an independent check at the level of individual wiring diagrams rather than totals alone.
  • Every projective class carries a symmetry group drawn from a short list (cyclic, dihedral, A4, S4, A5), and the orbit–stabilizer identity m(E)·|HE| = |GP| organizes the way Euclidean classes sit inside projective classes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the completeness theorem survives scrutiny, the same search framework could be adapted to near-optimal arrangements by extending the composite-generator alphabet to larger side-excess budgets; the structural lemmas suggest the search space would still be governed by the number of composite generators rather than the full word length.
  • The n = 91 gap is likely a search-parameter artifact rather than a sign of non-existence: the known doubling construction cannot reach it, but the paper's own first-hit method, run over a wider grid of defect anchors and branch orders, would plausibly find an example.
  • The O-matrix-to-word correspondence suggests the same classification pipeline—canonical representatives, deduplication, symmetry recovery—can be reused for other extremal families defined by local face constraints, not only the triangle-maximal ones.
  • Because the paper itself leaves one pruning step in the 2-defective search unproved, a skeptical reader can test completeness by rerunning the search with that pruning disabled; the paper reports identical output streams in that comparison, which is direct empirical evidence for the claimed coverage.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper develops exhaustive depth-first search algorithms for simple pseudoline arrangements of odd order that attain the black-face upper bound (2.4), encoded as reduced words for the longest permutation. Two families are treated: perfect arrangements for n ≡ 3, 5 (mod 6), and 2-defective arrangements for n ≡ 1 (mod 6). The perfect search branches only on even-indexed composite generators and proves (Cor. 4.11) that it emits exactly one reduced word per odd-majority commutation class of a perfect arrangement. The 2-defective search anchors one defect on the initial boundary, uses a special composite-generator alphabet and a "defect jump" operation, and proves (Thm. 6.13) that every projective class of 2-defective arrangements is represented in its output. Classification into commutation classes, Euclidean classes, and projective classes is performed through O-matrices and the operations of reflection, Euclidean rotation, and projective rotation; for each projective class the full symmetry group and orbit–stabilizer profile are recovered. The paper reports exhaustive counts up to n = 27 for perfect arrangements and up to n = 19 for 2-defective arrangements, plus the single-pentagon family at n = 25, partial first-hit results up to n = 93, and Corollary 7.1 stating existence of triangle-maximal arrangements for all odd n ≤ 89 except n = 11.

Significance. If the results are correct, this is a substantial contribution to the computational and structural study of triangle-maximal pseudoline arrangements. The paper provides exact enumeration counts and symmetry classifications, including the first exhaustive 2-defective counts, clean normal-form and orbit–stabilizer arguments for the perfect case and the P-class completeness of the 2-defective case, and a public repository with code and data. The cross-checks against the independent algorithms of BRS and Savchuk, and the separate reconstruction of the BRS algorithm, are valuable and cover the perfect case. The main gap is that the completeness of the 2-defective search relies on an explicitly omitted proof of the forced-descent pruning, with only empirical self-consistency checks in support; this affects the exactness of the n = 13, 19, 25 2-defective counts and the first-hit existence data for larger n.

major comments (2)
  1. [§6.4, Remark B.4, Prop. 6.11] The completeness of the 2-defective search is not fully proved. The implemented algorithm includes the forced-descent pruning of Appendix B in the ordinary tail after the special generator, and Remark B.4 explicitly states that a proof of this pruning is omitted. Proposition 6.11 verifies that the canonical representative of a 2-defective word satisfies the listed admissibility conditions (i)–(v), but it does not show that the forced-descent pruning cannot reject that representative. Since this pruning is active in the exhaustive runs that produce the n = 19 counts and the n = 25 pentagon-family counts, an incorrect pruning would silently remove valid 2-defective words and make the counts and Corollary 7.1 incomplete. The pruning on/off identity of output streams and the X-invariance checks in Appendix E are empirical consistency checks, not a proof. Please either supply the omitted proo
  2. [Theorem 6.13 / Appendix E] The exactness of the 2-defective counts lacks independent certification. For perfect arrangements, Table 3 provides comparisons with BRS and Savchuk, and Appendix F reconstructs the BRS algorithm; for 2-defective arrangements the only checks are internal: the #P/#D*/#E* invariance in X and the pruning on/off identity. These checks are consistent with completeness but cannot detect a systematic error in the shared code path—for example, a defect-jump trigger that misses the same class at every X. Because the 2-defective family supplies the n = 13/19/25 counts and the first-hit examples for n ≥ 31, an independent cross-enumeration at n = 13 or n = 19, e.g., by a SAT-based or O-matrix-based method, would materially raise confidence. As published, the abstract's claim that "every wiring diagram is reached" is stronger than Theorem 6.13, which establishes only P-class coverage for the 2-defec
minor comments (4)
  1. [Abstract] The sentence "Completeness of the search and classification is proved: every wiring diagram is reached" is too strong for the 2-defective search, whose completeness is at the level of projective classes (Theorem 6.13). Please qualify the abstract accordingly.
  2. [Table 1] The table header "Exhaustive enumeration counts" may mislead readers for n = 25, where only the single-pentagon family is enumerated and the two-quadrilateral family is deliberately skipped. The footnote is clear, but consider stating in the caption that the n = 25 column is not a full enumeration of 2-defective arrangements.
  3. [§7.5, Table 5] The first-hit parameters (X, g1, branch order) are selected per instance by a budgeted grid search, and the normalized times depend on an iteration-count model. For reproducibility of the existence claims, please include in the repository a verification script that checks that each emitted word is reduced, encodes a wiring diagram, and is 2-defective (or perfect).
  4. [Appendix E] Table E.2 is a useful illustration, but the reader must infer from the row sums that the two C1 classes contribute #D* = 16 and the four C2 classes contribute #D* = 8. A brief sentence spelling out this arithmetic would make the orbit–stabilizer interpretation of Remark 6.14 easier to verify.

Circularity Check

0 steps flagged

No significant circularity: the enumeration and classification derivations are self-contained; the unproved pruning is a completeness caveat, not a circular step.

full rationale

The paper's central derivations do not reduce to their inputs. The perfect search depth Nperf = (n^2 − 2n + 3)/6 is derived from word length and the composite-generator decomposition (Lemma 4.5), not fitted to the output counts. Completeness for the perfect search is argued from structural lemmas (Lemma 4.4, Lemma 4.5, Lemma 4.7, Theorem 4.10, Corollary 4.11) that rest on commutability, Newman's lemma, and the geometry of defect-free arrangements; no enumeration count is used as an assumption. The 2-defective search similarly derives Nsp = (n^2 − 2n + 7)/6 from the word length and the two allowed defect profiles (Lemma 6.4), and its completeness is stated at the P-class level via Lemma 6.10, Proposition 6.11, and Theorem 6.13. The defect profiles themselves follow from the side-counting identity in Lemma 2.7, independent of the search. Classification into E- and P-classes is based on explicit O-matrix operations (Section 5 and Appendix C) with a canonical representative computed independently of the counts. External comparisons with BRS and Savchuk (Table 3) and the reconstructed BRS algorithm provide independent checks for the perfect cases rather than being required to define the outputs. The paper explicitly leaves one 2-defective pruning unproved in Remark B.4 ('We omit it') and supports it by the empirical identity of output streams with the pruning enabled and disabled; this is a completeness gap or correctness risk, not circularity, because the pruning is not a fitted parameter renamed as a prediction and the central completeness theorem is not established by invoking that pruning. The Appendix E X-invariance checks likewise illustrate consistency of the 2-defective search but are not the argument for Theorem 6.13. No load-bearing self-citation or uniqueness theorem imported from the authors' own prior work is used to force the results. Accordingly, no circular step meeting the evidentiary standard is present.

Axiom & Free-Parameter Ledger

1 free parameters · 6 axioms · 0 invented entities

The enumeration rests on no fitted constants: the search depths Nperf = (n²−2n+3)/6 and Nsp = (n²−2n+7)/6 are derived from word length, the defect profile (one pentagon or two quadrilaterals) follows from the side-counting identity in §2.5, and uniqueness of canonical representatives follows from Newman's lemma. The perfect-search completeness proof, the dihedral reduction of O-matrix operations, and the P-class canonicalization are self-contained given standard arrangement/oriented-matroid background. No entity is postulated beyond internal bookkeeping objects (composite generators, defects, anchors), which the enumeration itself constructs rather than assumes. The single hand-tuned element is the first-hit hyperparameter grid in Table 5, which affects runtime only.

free parameters (1)
  • 2-defective first-hit search hyperparameters (skipped generator X, first composite generator g1, branch order) = per-n values in Table 5, e.g., n=31: X=5, g1=6, descending; n=37: X=11, g1=16, ascending
    Hand-chosen per n to minimize time-to-first-hit in the partial searches of §7.5. They alter traversal order only, not the set of reachable leaves, so they do not affect the exhaustive counts or the existence claims. Listed for completeness as the only hand-set numeric choices in the pipeline.
axioms (6)
  • standard math Every simple pseudoline arrangement is isomorphic to a wiring diagram, i.e., encodable by a reduced word for the longest permutation w0; commutation-equivalent words give the same diagram (Goodman 1980, cited as [17]; Lemma 5.2).
    Underlies the entire word-based search (§2.1, §3.1). Standard background in arrangement theory, accepted as an external theorem.
  • standard math The O-matrix triple condition characterizes which label matrices arise from rank-3 oriented matroids / pseudoline arrangements ([14, Theorem 5.2.10]); used in Lemma C.9 to show two encodings of one arrangement have identical or all-reversed rows.
    Load-bearing for the dihedral-reduction Lemma C.10 and hence for P-class canonicalization (Appendix D). The paper relies on the cited handbook chapter rather than reproving it.
  • domain assumption The segment-count upper bound a3(A) ≤ floor(n(n−2)/3) and the fact that a bounded segment borders at most one triangular face (Eq. (2.3) and the double-crossing argument following it; cf. Blanc [7]).
    Defines the perfect/2-defective dichotomy on which both searches are built. Proved in-paper by a short geometric argument, but the bound itself is classical.
  • domain assumption Faces of a pseudoline arrangement admit a checkerboard coloring; for odd n, an affine face and its antipodal counterpart carry opposite colors in the projective closure (§2.2).
    The defect framework (Definition 2.4), Lemma 2.2, and Lemma 2.7 all depend on this coloring structure.
  • domain assumption The parity-color convention 'odd generators open black faces, even generators open white faces' is WLOG for odd n by the reflection σi → σn−2−i (§4).
    A modeling convention that halves the 4n geometric redundancy. Justified as without loss of generality, but it fixes the orientation of the entire enumeration.
  • domain assumption The defect jump preserves simplicity and 2-defectiveness via the even-crossing Jordan-curve argument: every other pseudoline crosses the closed region boundary ∂R an even number of times (Lemma 6.8).
    The key topological premise for the defect-jump mechanism in §6.3; assumes standard Jordan-curve behavior of pseudolines in the projective plane.

pith-pipeline@v1.3.0-daily-deepseek · 50553 in / 27089 out tokens · 257011 ms · 2026-08-03T11:02:25.134867+00:00 · methodology

0 comments
read the original abstract

We describe algorithms for the exhaustive enumeration and classification of simple arrangements of $n$ pseudolines ($n$ odd) maximizing the number of triangular faces. The depth-first search enumerates reduced words for the longest permutation $w_0$ by branching only on the even-indexed generators, using pruning constraints imposed by the geometry of optimal arrangements. The approach handles both perfect arrangements with a regular triangular pattern and unavoidable deviations from it for $n \equiv 1 \pmod 6$. The output is classified into a hierarchy of equivalence classes: by commutation, by Euclidean transformations, and by projective transformations. For each projective class we recover its full symmetry group $G \subseteq S_{n+1}$ together with the orbit-stabilizer profile of its Euclidean subclasses. Completeness of the search and classification is proved: every wiring diagram is reached. We report full enumerations; e.g. for $n=27$, 85,562,064 wiring diagrams partitioned into 56,646 projective classes. For larger $n$ (up to $n=93$), where exhaustive enumeration is out of reach, we report partial (first-hit) results.

Figures

Figures reproduced from arXiv: 2607.29236 by Denis Utkin, Roman Parpalak.

Figure 1
Figure 1. Figure 1: From a labeled line arrangement to an allowable se￾quence and to the corresponding wiring diagram. Conversely, any reduced word for w0 can be drawn as a wiring diagram (also called a primitive sorting network [16]): the labels start in the order 0, 1, . . . , n − 1, and each generator σi crosses the two wires currently occupying positions i and i + 1. The obtained diagram is itself a simple pseudoline arra… view at source ↗
Figure 2
Figure 2. Figure 2: A 2-defective arrangement for n = 7 and a perfect arrangement for n = 9. 2.4. Black-face bound and defects. There are two checkerboard colorings, dif￾fering by a swap of colors. We choose the one in which the black faces are at least as numerous as the white faces. Then in a perfect arrangement the bounded triangles are black and the adjacent non-triangular faces are white ( [PITH_FULL_IMAGE:figures/full_… view at source ↗
Figure 3
Figure 3. Figure 3: Wiring diagram patterns for n = 9. (a) The forced odd prefix σ1σ3σ5σ7, the single σ0, and the trailing generators. (b) A composite generator K4 = σ4σ3σ5: the white σ4, with σ3 and σ5 closing two black triangles. (c) Two consecutive K4 form a white quadrilateral; its four black neighbors cannot all be triangles, since otherwise the highlighted wires would cross twice (Lemma 4.6). 4.1. Constraints from exter… view at source ↗
Figure 4
Figure 4. Figure 4: Examples of local merge patterns producing defects. (a) An external digon merges with an adjacent bounded triangle into a single external face, a quadrilateral defect in the projective closure (σ3 skipped). (b) Two adjacent bounded triangles merge into an internal quadrilateral defect (K − 2 K4). (c) Three black faces in a row merge into a pentagonal defect when both crossings sepa￾rating them are missing … view at source ↗
Figure 5
Figure 5. Figure 5: Local quadrilateral defect patterns: (a) K−-trapezoid K − 4 K2 K4 and K+-trapezoid K + 2 K4 K2; (b) rhombi K + 2 K0 and K − 2 K4 = K + 4 K2; (c) trailing external defect K + 2 . Proof. Let W be any reduced word encoding the wiring diagram D. By the argu￾ment of Lemma 4.1, commutations of individual σi (2.1) bring it into the form W = σ1σ3 · · · σcX · · · σn−2 R, [PITH_FULL_IMAGE:figures/full_fig_p021_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: An arrangement for n = 7 with two applications of K0. The composite generator K0 = σ0σ1 has only one black generator and admits no shortened special variant. When the external face between wires 0 and n − 1 is a quadrilateral defect, the single-use rule of Lemma 4.2 relaxes instead: Corollary 6.3 (K0 for a quadrilateral defect at (0, n−1)). In case (2) of Lemma 6.2, the two K0 encoding the quadrilateral de… view at source ↗
Figure 7
Figure 7. Figure 7: The exceptions allowed by Lemma 6.6, for X = 1: (a) KX+1 K + X+1 as a local pattern; (b) KX+1 KX+1 in the word K + 4 K2K2K0K4K2K4 emitted at n = 7 [PITH_FULL_IMAGE:figures/full_fig_p025_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: (a) The region R in the proof of Lemma 6.8. (b) A fragment for Lemma 6.9 with a K+ g whose omitted σh (h = g − 1) would cross L1, L2 held by the slots h, h + 1: the wires carry non￾adjacent sides of F. Here the σg of the later Kg moves L1 out of the pair, ending its side before the σh of that Kg closes F [PITH_FULL_IMAGE:figures/full_fig_p026_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Two 2-defective arrangements differing by a defect jump on the strip between wires 4 and 5: (a) an internal defect (K − 4 at the start) becomes (b) a trailing external one (K + 2 at the end). A pentagonal defect (c) and the two quadrilateral defects (d) after a defect jump. By Lemma 6.8 the two faces of A at p along the strip are black, which leaves three cases. If F is a quadrilateral and p is the shared … view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

25 extracted references · 3 linked inside Pith

  1. [1]

    Grünbaum

    B. Grünbaum. Arrangements and Spreads. Number 10 in CBMS Regional Conference Series in Mathematics. American Mathematical Society, Providence, RI, 1972

    1972

  2. [2]

    Harborth

    H. Harborth. Some simple arrangements of pseudolines with a maximum number of triangles. In Discrete Geometry and Convexity , volume 440 of Annals of the New York Academy of Sciences, pages 31–33. New York Academy of Sciences, 1985

    1985

  3. [3]

    Roudneff

    J.-P. Roudneff. On the number of triangles in simple arrangements of pseudolines in the real projective plane. Discrete Mathematics, 60:243–251, 1986

    1986

  4. [4]

    Füredi and I

    Z. Füredi and I. Palásti. Arrangements of lines with a large number of triangles.Proceedings of the American Mathematical Society , 92(4):561–566, 1984

    1984

  5. [5]

    Forge and J

    D. Forge and J. L. Ramírez Alfonsín. Straight line arrangements in the real projective plane. Discrete & Computational Geometry , 20(2):155–161, 1998

    1998

  6. [6]

    Bartholdi, J

    N. Bartholdi, J. Blanc, and S. Loisel. On simple arrangements of lines and pseudo-lines inP 2 and R2 with the maximum number of triangles. InSurveys on Discrete and Computational Geometry: Twenty Years Later , volume 453 ofContemporary Mathematics, pages 105–116. American Mathematical Society, 2008

    2008

  7. [7]

    J. Blanc. The best polynomial bounds for the number of triangles in a simple arrangement of n pseudo-lines. Geombinatorics, 21:5–17, 2011

    2011

  8. [8]

    Fujimura

    K. Fujimura. The Tokyo Puzzles . Charles Scribner’s Sons, New York, 1978. Edited by M. Gardner

    1978

  9. [9]

    Kobon triangles.https://oeis.org/A006066

    OEIS Foundation Inc. Kobon triangles.https://oeis.org/A006066. Accessed 2026-07

    2026

  10. [10]

    V. I. Arnold.Arnold’s Problems. Springer, 2004. English edition

    2004

  11. [11]

    Fejes Tóth

    L. Fejes Tóth. A combinatorial problem concerning oriented lines in the plane.The American Mathematical Monthly, 82(4):387–389, 1975

    1975

  12. [12]

    Felsner and K

    S. Felsner and K. Kriegel. Triangles in Euclidean arrangements.Discrete & Computational Geometry, 22(3):429–438, 1999

    1999

  13. [13]

    Parpalak and D

    R. Parpalak and D. Utkin. The 18· 2t + 1 triangle-maximal series of straight lines. arXiv:2604.22035, 2026

    Pith/arXiv arXiv 2026

  14. [14]

    Felsner and J

    S. Felsner and J. E. Goodman. Pseudoline arrangements. In J. E. Goodman, J. O’Rourke, and C. D. Tóth, editors, Handbook of Discrete and Computational Geometry , chapter 5, pages 125–157. CRC Press, Boca Raton, FL, 3rd edition, 2017

    2017

  15. [15]

    J. E. Goodman and R. Pollack. Allowable sequences and order types in discrete and compu- tational geometry. In J. Pach, editor,New Trends in Discrete and Computational Geometry , volume 10 ofAlgorithms and Combinatorics , pages 103–134. Springer, Berlin, 1993

    1993

  16. [16]

    D. E. Knuth.Axioms and Hulls , volume 606 ofLecture Notes in Computer Science . Springer, Berlin, 1992

    1992

  17. [17]

    J. E. Goodman. Proof of a conjecture of Burr, Grünbaum, and Sloane.Discrete Mathematics, 32(1):27–35, 1980

    1980

  18. [18]

    Bokowski, J.-P

    J. Bokowski, J.-P. Roudneff, and T.-K. Strempel. Cell decompositions of the projective plane with Petrie polygons of constant length.Discrete & Computational Geometry , 17(4):377–392, 1997

    1997

  19. [19]

    K. Wood. Optimal solution for the number of Kobon triangles with eleven lines.https:// github.com/Bombardlos/Kobon_Triangle_Workspace/blob/main/11_Description.pdf, 2024. Accessed 2026-07

    2024

  20. [20]

    P. Savchuk. Constructing optimal Kobon triangle arrangements via table encoding, SAT solving, and heuristic straightening. arXiv:2507.07951, 2025

    Pith/arXiv arXiv 2025

  21. [21]

    equivalence

    M. H. A. Newman. On theories with a combinatorial definition of “equivalence”.Annals of Mathematics, 43(2):223–243, 1942

    1942

  22. [22]

    Diekert and Y

    V. Diekert and Y. Métivier. Partial commutation and traces. In G. Rozenberg and A. Sa- lomaa, editors, Handbook of Formal Languages , volume 3, pages 457–533. Springer, Berlin, Heidelberg, 1997

    1997

  23. [23]

    J. E. Goodman and R. Pollack. Semispaces of configurations, cell complexes of arrangements. Journal of Combinatorial Theory, Series A , 37(3):257–293, 1984

    1984

  24. [24]

    G. Rote. NumPSLA — an experimental research tool for pseudoline arrangements and order types. arXiv:2503.02336, 2025. Full version; abridged version at EuroCG 2025

    Pith/arXiv arXiv 2025

  25. [25]

    Parpalak and D

    R. Parpalak and D. Utkin. Triangle-maximal pseudoline arrangements: code and data.https: //github.com/parpalak/pseudoline-algorithms, commit e80f7a2, 2026

    2026