The conic geometry of rectangles inscribed in lines

Bruce Olberding; Elaine A. Walker

REVIEW 1 major objections 4 minor 7 references

The conic geometry of rectangles inscribed in lines

T0 review · 1 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read The paper establishes that the locus of centers of rectangles inscribed in two pairs of lines is a hyperbola exactly when neither pair is parallel, at most one pair is orthogonal, and the pairs are not translations of each other; and…

desk verdict Solid cone-geometry paper with a correct core classification and a real gap in the 'faithful parameterization' claim that needs one more scale parameter. read the letter →

arxiv 1908.05413 v2 pith:KL5GXEAZ submitted 2019-08-15 math.MG

Bruce Olberding , Elaine A. Walker This is my paper · ORCID

classification math.MG MSC 51N1011E10

keywords inscribedrectanglesrectanglelocushyperbolicallyrotatedconesHR-conespairsoflinesconicsectionssemialgebraicparameterizationcenterloci

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper develops a geometric way to describe all rectangles whose four vertices lie on given lines. It reduces the problem to two pairs of lines—one diagonal of the rectangle joins the lines of one pair, the other diagonal joins the lines of the other pair—and encodes each pair of lines as a surface in three-dimensional space called a hyperbolically rotated cone. The locus of rectangle centers is then the projection of the intersection of two such cones. The paper proves this locus is a hyperbola exactly when neither pair is parallel, at most one pair is orthogonal, and the pairs are not translations of each other; otherwise it is a line, a point, a line with an open segment missing, empty, or the whole plane. It also proves every hyperbola in the plane arises this way, and that all line-pairings producing a given hyperbola are parameterized by a semialgebraic surface in $\mathbb{R}^4$. The value is a uniform, non-computational framework for a problem that has previously been treated by direct equations or case analysis.

What carries the argument

The machinery has three pieces. First is the reduction, called the permutation trick, that any rectangle inscribed in four lines is captured by choosing which two pairs of lines carry its diagonals; this lets the paper work with two pairs of lines. Second is the HR-cone: an elliptical cone with equation $z^2=(x-a)^TA(x-a)$ where $A$ is positive definite with determinant $1$; a pair of intersecting lines defines such a cone whose apex is the crossing point and whose level curves are ellipses of area $\pi z^2$. Third is the intersection lemma: the rectangle locus for two pairs is exactly the projection to the plane of the intersection of the two HR-cones. The determinant inequality for distinct positive-definite determinant-one matrices, $\det(A-B)<0$, is what forces the projected intersection to be a hyperbola and supplies its center and asymptote directions.

What would settle it

Choose a generic quadruple of lines, solve the polynomial system for all rectangles with vertices on the four lines, and compare the set of centers with the hyperbola predicted by Theorem 4.5 for the relevant diagonal pairing; any center outside the predicted hyperbola, including a degenerate or exscribed rectangle, would refute the completeness of the two-pair reduction.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that rectangles inscribed in two pairs of lines are governed by the intersection of two HR-cones. If the pairs meet at points $a$ and $b$ and have cone matrices $A$ and $B$, the rectangle locus is the set of points $x$ satisfying $(x-a)^TA(x-a)=(x-b)^TB(x-b)$. Because for distinct determinant-one positive-definite matrices the difference $C=A-B$ is invertible with $\det(C)<0$, this equation can be completed to the equation of a hyperbola centered at $C^{-1}(Aa-Bb)$. Hence the locus is a hyperbola precisely in the generic case: neither pair parallel, no more than one pair orthogonal, and the pairs not translations of each other. The same cone-intersection description yields a converse: for any prescribed hyperbola, there exist pairs of HR-cones whose intersection projects to it, and the set of all such pairs is faithfully parameterized by a semialgebraic surface in $\mathbb{R}^4$. The classification of rectangle loci is therefore complete, and every hyperbola is realizable.

Load-bearing premise

The load-bearing premise is the reduction that every rectangle inscribed in four lines is found by assigning its two diagonals to two pairs of the lines; if a rectangle were missed because its vertices lie on the lines in an order not covered by this assignment, or because it is degenerate or 'exscribed' in the polygon version, then the classification would not describe all inscribed rectangles.

Editorial extensions

If this is right

For a generic quadrilateral, the centers of all inscribed rectangles can be described as a finite union of hyperbolas, lines, and points, one locus per diagonal pairing of the four side-lines.
Every hyperbola in the plane occurs as a rectangle locus, so there is no hidden algebraic restriction on the shape of center loci.
The pairs of line configurations that share the same rectangle-locus hyperbola form a semialgebraic surface in $\mathbb{R}^4$, meaning the same locus can arise from many geometrically different line pairs.
Two HR-cones intersect in a curve whose projection is a hyperbola, giving a purely geometric way to build hyperbolas as shadows of cone intersections.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

One extension not pursued in the paper is to impose the extra condition that the inscribed rectangle be a square; within this cone framework that would require intersecting the HR-cones with an algebraic side-length condition, a testable computation for the 21 loci of a quadrilateral.
Because the locus only depends on the difference matrix $A-B$ and the vector $Aa-Bb$, the map from line configurations to hyperbolas has positive-dimensional fibers; this suggests that only a low-dimensional summary of the configuration is recoverable from the locus alone.
The same cone method could be adapted to rectangles inscribed in arrangements of higher-dimensional flats, where the resulting surfaces are higher-dimensional cones and the locus is a shadow of their intersection; whether the shadow remains conic in that setting is an open question.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

The thing to know: the paper has a genuinely useful idea—encoding line pairs as HR-cones, so rectangle-center loci become projections of cone intersections—and the main classification/converse is close to right. But Theorem 5.3's "faithful parameterization" is not faithful. The proof fixes the difference matrix A-B=C and misses pairs with A-B=αC, so it undercounts the solution set by a dimension. That needs fixing before I would trust the semialgebraic-counting claim.

What is new: the HR-cone framework itself, the general classification in Theorem 4.5, and the converse that every hyperbola is realized. The proofs are mostly self-contained, and the algebra works; I checked Lemma 3.5 and the cone-completion argument. The paper is honest about Schwartz's prior hyperbola theorem and builds on it. The sign typo at the end of Theorem 5.3 ("A = C − B" should be "A = C + B") is minor and obvious.

Soft spots, in order of size. First, the parameterization gap. After normalizing H to x^T C x=1, the proof only analyzes A,B with A-B=C. But a pair with A-B=αC and the appropriate constant term projects to the same H; such pairs exist (e.g., C=diag(1,−1), B=diag(√2−1,√2+1), A=B+2C works for α=2). So the claimed two-dimensional surface is one dimension too small. The converse "every hyperbola occurs" still holds; the faithful-parameterization assertion does not. Second, degenerate hyperbolas are handled loosely. Example 5.2 has all the hypotheses of Theorem 4.5 but yields a pair of crossing lines, not a nondegenerate hyperbola; the theorem's iff statement needs a nonzero constant condition, and Corollary 4.7's list omits that case. This is correctable, but it is a real imprecision. Third, the reduction to two pairs of lines and the "exscribed" rectangles mentioned in Section 6 deserve one clarifying sentence; I do not think that undermines the inscribed-rectangle interpretation.

Who it is for: people working on geometric configurations and conic encodings, especially the Olberding-Walker line of work. It deserves a serious referee: the core is solid, the central converse is salvageable, and the cone machinery is worth having. I would not cite the parameterization as stated, but I would ask the authors to fix the scale issue and tighten the degenerate-case taxonomy before publication.

Referee Report

1 major / 4 minor

Summary. The paper develops a geometric framework for the locus of centers of rectangles inscribed in two pairs of lines. For each pair of intersecting lines it constructs an HR-cone, a real elliptical cone z^2=(x-a)^T A(x-a) with det(A)=1; parallel pairs give vertical planes with a missing midsection. The rectangle locus is shown, via Lemma 4.2, to be the projection of the intersection of the two surfaces defined by the two pairs. Theorem 4.5 classifies when this locus is a hyperbola: precisely when neither pair is parallel, at most one pair is orthogonal, and the pairs are not translates. Theorem 5.3 claims that every hyperbola in the plane arises as such a projection and that the set of all cone pairs producing a given hyperbola is faithfully parameterized by a semialgebraic surface in R^4. The paper also proves uniqueness of the defining lines for non-unit HR-cones, records a rotation invariance for orthogonal pairs, and gives examples of non-uniqueness of cone pairs with the same rectangle locus.

Significance. The geometric translation is natural, and Theorem 4.5 is a clean generalization of Schwartz's hyperbola theorem. The proofs are explicit, self-contained, and checkable; Lemma 3.5 and Theorem 3.6 are elegant. The construction in Theorem 5.3 that realizes every hyperbola from a cone pair is a valuable converse. However, the faithful-parameterization assertion, which is the paper's principal quantitative claim, is not correct as stated; the omitted scaling parameter changes the dimension of the claimed moduli space. This does not destroy the existence part of Theorem 5.3, but it requires a substantive revision of the theorem and its corollaries.

major comments (1)

[5, Theorem 5.3] The parameterization asserted in Theorem 5.3 is not faithful, because the proof only treats pairs of cone matrices satisfying A-B=C, where H is written as x^T C x=1. Lemma 4.4 shows that a pair with A-B=λC, Aa-Bb=0, and (-a^T A a + b^T B b)/λ=1 produces the same H for every nonzero λ, and such pairs exist for every λ. For C=diag(1,-1), λ=2, take u=√2-1, B=diag(u,1/u), and A=diag(u+2,1/u-2); these are positive definite determinant-one matrices with A-B=2C, and choosing b on the indefinite hyperbola b^T C A^{-1}B b=1 gives an HR-cone pair whose intersection projects to x^2-y^2=1. This pair satisfies none of equations (2)-(3), which enforce A-B=C. Consequently, the set of all cone pairs is not parameterized by the two-dimensional surface described; an additional continuous parameter λ is needed, so the true solution set has dimension at least three. The existence part of the theorem is unaffected, but the 'faithfully parameterized' claim is false as stated.

minor comments (4)

[5, proof of Theorem 5.3] In the final paragraph of the proof, 'With A = C - B' should read 'With A = C + B', since A-B=C.
[4, Corollary 4.8] Corollary 4.8 needs an explicit hypothesis that the two HR-cones have distinct cone matrices; if the cone matrices are equal but the apices differ, the projection is a line, and if the cones coincide, the projection is the whole plane.
[4, Theorem 4.5] The paper uses 'hyperbola' to include degenerate hyperbolas, since the right-hand side of the equation in Lemma 4.4 can vanish. This convention should be stated explicitly before Theorem 4.5, as standard terminology reserves 'hyperbola' for the nondegenerate conic.
[5, Proposition 5.1] 'Rectangular locus' should be 'rectangle locus' for consistency with Definition 4.1.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's results are proved from definitions via standard algebra, and the cited external/self references are not load-bearing for the central claims.

full rationale

The central derivation chain is self-contained. Theorem 3.3 derives the HR-cone equation directly from the geometry of two intersecting lines, and Lemma 4.4 obtains the rectangle locus equation by substituting the two cone equations into the projection condition of Lemma 4.2. Theorem 4.5 then follows from Lemma 3.5 (det(A-B)<0), with the converse and exceptional cases proved from the same cone/locus formalism rather than imported from Schwartz. Theorem 5.3 explicitly constructs, for a given hyperbola x^T C x = 1, positive definite determinant-one matrices A, B with A-B = C and vectors a, b satisfying Aa = Bb, so the claimed realization is proved rather than assumed. The paper's citation of Schwartz is for the initial 'permutation trick' reduction and as comparative context; the hyperbola theorem is independently proved in Theorem 4.5 with a converse. The only self-citation, [5], is invoked for degenerate-hyperbola conditions and as additional context, not as the basis of the main classification or parameterization. The skeptic's complaint about Theorem 5.3 omitting cone pairs with A-B = alpha C is a possible mathematical incompleteness or faithfulness concern, not a circularity: the proof does not define the target hyperbola in terms of the constructed parameter space, nor fit a parameter and rename it a prediction. Therefore there is no reduction of a claimed result to its own inputs, and no load-bearing self-citation chain. Score 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 1 invented entities

No free parameters are fit to data. The paper imports standard facts (spectral theorem, conic classification) and the permutation-trick reduction from Schwartz. The HR-cone is a fully defined construction with derived properties, not an ad hoc entity.

assumptions (3)

standard math Spectral theorem for real symmetric matrices; positive definite matrices have unique positive square roots.
Used throughout; for example, Lemma 3.5 diagonalizes A via Q^T Λ Q, and Theorem 3.6 uses the spectral decomposition A = R_α Λ_β R_{-α}.
standard math Standard classification of real conics by the determinant of the quadratic-form matrix (Kendig [1]).
Used in Theorem 3.6(1) and Theorem 4.5 to identify x^T C x = constant with a hyperbola when det(C) < 0.
domain assumption The 'permutation trick' reduction: rectangles inscribed in four lines are classified by the pairing of lines that contain opposite vertices (diagonals).
Invoked in Section 1 and used to focus the paper on two-pair rectangle loci; cited to Schwartz [7, Section 4.1] but not reproved.

invented entities (1)

HR-cone (hyperbolically rotated cone) independent evidence
purpose: Encodes a pair of intersecting lines as an elliptical cone with cone matrix of determinant 1; the surface records midpoints and half-lengths of segments joining the lines.
Fully defined by Equation (1) and the determinant condition. Theorem 3.3 proves the surface of a line pair is an HR-cone, and Theorem 3.6 proves the defining lines are recoverable. It is a construction with derived properties, not an unsubstantiated postulate.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The conic geometry of rectangles inscribed in lines." pith.science (2026). https://pith.science/paper/KL5GXEAZ

@misc{pith2026190805413,
  author       = {Pith},
  title        = {Pith review of: The conic geometry of rectangles inscribed in lines},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KL5GXEAZ}},
  note         = {Machine review of arXiv:1908.05413}
}

read the original abstract

We develop a circle of ideas involving pairs of lines in the plane, intersections of hyperbolically rotated elliptical cones and the locus of the centers of rectangles inscribed in lines in the plane.

Figures

Figures reproduced from arXiv: 1908.05413 by the authors.

**Figure 1.** The level curves of the two HR-cones defined by the pairs A, C and B, D. The rectangle locus, which is the projection of the intersection of the two cones to the plane, is the hyperbola that passes through the intersections of the level curves [PITH_FULL_IMAGE:figures/full_fig_p011_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

7 extracted references · 6 canonical work pages

[1]

Kendig, Conics, Math

K. Kendig, Conics, Math. Assoc. America, 2005

work page 2005
[2]

A survey on the square peg problem,

B. Matschke, “A survey on the square peg problem,” Notices Amer. Math. Soc., 61 (2014), 346–352

work page 2014
[3]

M. D. Meyerson, Balancing acts, Topology Proc. 6 (1981), 59–75

work page 1981
[4]

Olberding and E

B. Olberding and E. A. Walker, Supplementary graphics, http://ziaspace.com/math/

work page
[5]

Paths of rectangles inscribed in lines over fields

B. Olberding and E. A. Walker, Aspect ratio and slope of algebraic rectangles inscribed in lines over ﬁelds, submitted, arXiv:2006.14424

work page Pith review arXiv 2006
[6]

Schwartz, Four Lines and a Rectangle, Exp

R. Schwartz, Four Lines and a Rectangle, Exp. Math. (2020), DOI: 10.1080/10586458.2019. 1671923

arXiv 2020
[7]

Schwartz, A trichotomy for rectangles inscribed in Jordan loops

R. Schwartz, A trichotomy for rectangles inscribed in Jordan loops. Geom Dedicata 208 (2020), 177–196. Department of Mathematical Sciences, New Mexico State University, Las Cruces, NM 88003-8001 Email address: bruce@nmsu.edu 1801 Imperial Ridge, Las Cruces, NM 88011 Email address: miselaineeous@yahoo.com

work page 2020

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Kendig, Conics, Math

K. Kendig, Conics, Math. Assoc. America, 2005

work page 2005

[2] [2]

A survey on the square peg problem,

B. Matschke, “A survey on the square peg problem,” Notices Amer. Math. Soc., 61 (2014), 346–352

work page 2014

[3] [3]

M. D. Meyerson, Balancing acts, Topology Proc. 6 (1981), 59–75

work page 1981

[4] [4]

Olberding and E

B. Olberding and E. A. Walker, Supplementary graphics, http://ziaspace.com/math/

work page

[5] [5]

Paths of rectangles inscribed in lines over fields

B. Olberding and E. A. Walker, Aspect ratio and slope of algebraic rectangles inscribed in lines over ﬁelds, submitted, arXiv:2006.14424

work page Pith review arXiv 2006

[6] [6]

Schwartz, Four Lines and a Rectangle, Exp

R. Schwartz, Four Lines and a Rectangle, Exp. Math. (2020), DOI: 10.1080/10586458.2019. 1671923

arXiv 2020

[7] [7]

Schwartz, A trichotomy for rectangles inscribed in Jordan loops

R. Schwartz, A trichotomy for rectangles inscribed in Jordan loops. Geom Dedicata 208 (2020), 177–196. Department of Mathematical Sciences, New Mexico State University, Las Cruces, NM 88003-8001 Email address: bruce@nmsu.edu 1801 Imperial Ridge, Las Cruces, NM 88011 Email address: miselaineeous@yahoo.com

work page 2020