Pith. sign in

REVIEW 4 major objections 6 minor 20 references

Discrete Lorentz surfaces and s-embeddings II: maximal surfaces

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read There are s-embeddings whose Lorentz lifts are maximal surfaces already at the discrete level.

desk verdict Genuinely discrete maximal s-embeddings with a constant-X associated family; the main theorem has a fixable proof gap rather than a fatal flaw. read the letter →

arxiv 2411.19055 v1 pith:HRCTSEMU submitted 2024-11-28 math.DG math-phmath.MP

Niklas Christoph Affolter , Felix Dellinger , Christian Müller , Denis Polly , Nina Smeenk This is my paper · ORCID
classification math.DGmath-phmath.MP MSC 53A1052C2682B20
keywords s-embeddingsincircularnetsdiscretemaximalsurfacesLorentzspaceisothermicIsingmodelX-variablescirclepatterns
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

This paper answers positively a question from the statistical-mechanics literature: whether some s-embeddings—planar quad nets in which every face has an incircle and which encode Ising couplings—already lift to maximal surfaces in Lorentz space at the discrete level, with no thermodynamic limit needed. The construction singles out a class of isothermic s-embeddings called maximal congruences and shows they correspond to discrete S-maximal surfaces in Lorentz space, the Lorentz analogue of Euclidean discrete minimal surfaces built from circle patterns. Each such surface carries a one-parameter associated family of isometric discrete maximal surfaces, obtained by rotating the discrete differential around the normal axis. Projecting each member back to the plane yields a one-parameter family of s-embeddings, and the paper proves that the Ising coupling data of these s-embeddings—the X-variables—are independent of the family parameter.

What carries the argument

The central object is the null congruence: a triple in Lorentz space $\mathbb{R}^{2,1}$ consisting of timelike spheres at white vertices, null-spheres at black vertices, and isotropic lines on faces, whose orthogonal projection is an incircular net. Three successive specializations carry the argument: isothermic congruences, where the centers of the timelike spheres form a conjugate net; Koebe congruences, where the circles of the associated S-isothermic net (the sphere-circle contact discretization of isothermic surfaces) lie on the upper unit sphere and come from hyperbolic orthogonal circle patterns; and maximal congruences, where the Christoffel dual is a Koebe congruence. The Christoffel dual supplies the normal map and the discrete Weierstraß representation. The associated family is generated by a closed discrete differential obtained from elliptic Lorentz rotations around the axis from the origin to the face normal; the key quantitative fact is that all associated vertex spheres have the same radius $\sin\varphi$, which is what makes the associated null congruences exist and lets the X-variables be computed from points whose distances to the center are $\varphi$-independent.

What would settle it

Take a concrete hyperbolic orthogonal circle pattern in the Poincaré disk (for example a regular square grid), build the corresponding maximal congruence with the formulas of Section 8, construct the associated null congruences for $\varphi=0$ and $\varphi=\pi/2$, and compute the X-variables of their incircular-net projections; any deviation from equality between the two values of $\varphi$ would refute Theorem 11.2.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the Lorentz lift of an incircular net can be a maximal surface before any limit is taken. The route is: take a hyperbolic orthogonal circle pattern, viewed as a Koebe net; form its Christoffel dual, which defines a discrete S-maximal surface; then pass through the 2:1 correspondence between isothermic congruences and S-isothermic nets to obtain a null congruence, whose orthogonal projection is the desired maximal s-embedding. The paper gives a discrete Weierstraß representation with explicit coordinates in terms of the circle-pattern data. For each maximal surface it then constructs the associated family $h^{\varphi}$ by elliptic Lorentz rotations, derives associated contact congruences whose vertex spheres all have radius $\sin\varphi$, and from them associated null congruences that project to incircular nets. Theorem 11.2 shows the X-variables are $\varphi$-independent, so the whole associated family defines one and the same Ising model.

Load-bearing premise

The construction inherits the companion paper's framework wholesale—in particular the 2:1 correspondence between isothermic congruences and S-isothermic nets and the reading of X-variables as squared Lorentz distance ratios—and if that framework has unstated exceptional cases, the maximal-surface construction inherits the error.

Editorial extensions

If this is right

  • There are genuine discrete maximal surfaces in Lorentz space whose orthogonal projections are s-embeddings, so the correspondence between Ising s-embeddings and maximal surfaces is not only a continuum phenomenon.
  • Every maximal s-embedding carries a one-parameter family of s-embeddings with identical X-variables; the whole family describes the same Ising model.
  • The family passes through non-curvature-line conformal parametrizations, with the half-turn member recovering a rotation of the original surface; at $\varphi=\pi/2$ one obtains conformal asymptotic coordinates in analogy with the smooth theory.
  • Maximal s-embeddings are obtainable from hyperbolic orthogonal circle patterns by a variational principle, so boundary data determine them uniquely.

Reading between the lines

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

  • The paper does not say this, but $\varphi$-independence of X-variables means any Ising observable that depends only on these couplings is invariant under the associated-family flow; the family acts as a discrete symmetry of the statistical model.
  • The same Koebe-to-Christoffel route should transfer to the dimer-model side, where conical nets (t-embeddings) have an analogous Lorentz lift; one testable expectation is that their Lorentz-minimal discrete lifts also admit associated families with constant coupling data.
  • A direct numerical check on a small square-grid Koebe pattern would settle the mechanism: compute the X-variables for the associated incircular nets at $\varphi=0$ and $\varphi=\pi/2$ and compare; the paper predicts exact equality.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. Building on their companion paper [ADM+24], the authors construct a discrete analogue of maximal surfaces in Lorentz space within the framework of s-embeddings/incircular nets. They define maximal congruences as isothermic congruences whose Christoffel dual is a Koebe congruence, relate these to hyperbolic orthogonal circle patterns, and derive a discrete Weierstrass representation. The paper then defines an associated family of congruences parametrized by φ ∈ S1 and shows that each member yields, after a Laguerre offset, a null congruence and hence an incircular net. The central claim is that the X-variables—the Ising weights of the corresponding s-embeddings—are independent of φ. The paper thus answers Chelkak–Laslier–Russkikh's question in the positive, at least at the level of construction, provided the main theorems are fully established.

Significance. If the main claims are correct, this paper is a substantial bridge between discrete differential geometry and the statistical mechanics of s-embeddings. The identification of a class of isothermic s-embeddings that lift to discrete maximal surfaces, together with a one-parameter family of s-embeddings with constant Ising weights, is a genuinely new and interesting structure. The paper also provides explicit Weierstrass-type formulas and connects the construction to hyperbolic orthogonal circle patterns, which gives a variational route to existence from boundary data. The authors are careful to acknowledge open points and conjectures, and the reliance on [ADM+24] is legitimate for a sequel. However, the manuscript is not yet self-contained at the level needed for the main theorem: the proofs of Theorem 9.1 and Theorem 10.4 are only sketched, and Theorem 11.2 contains a missing phase argument that is load-bearing for the headline claim about Ising weights.

major comments (4)
  1. [§11, Theorem 11.2] The proof reduces the X-variable to the planar cross-ratio X◦(w) = −(Z1−P0*)(Z2−P0*)/(Z3−P0*)(Z4−P0*), and Lemma 11.1 establishes only that each distance |Zi−P0*| is independent of φ. A cross-ratio is not determined by the four moduli alone: writing Zi−P0* = r_i e^{iθ_i}, the phase combination θ1+θ2−θ3−θ4 enters X◦. The conclusion 'hence so are the X-variables' therefore requires a proof that the four points Zi(φ) are obtained from Zi(0) by a single rotation about P0*, or an equivalent control of the phases. The formulas for the projected isotropic-line directions in the proof of Lemma 11.1 suggest that such a common-rotation property may hold, but it is not stated or proved. Since Theorem 11.2 is the paper's main statistical-mechanics claim, this is a load-bearing gap.
  2. [§9, Theorem 9.1] The proof is one sentence: 'It suffices to check the closing condition around each black vertex, which corresponds to the fact that P φ is well-defined in the calculation above.' The closing of the associated discrete differential d⊙hφ◦ is the foundation for the entire associated-family construction. The informal discussion around Equation (9.4) is not a complete proof: one needs an explicit verification that the telescoping sums vanish in the Ti, n, and bi components, or a direct computation of the sum around each black vertex.
  3. [§10, Theorem 10.4] The proof chooses one of the two spheres at an initial black vertex b0 and then asserts that all other spheres can be chosen consistently. Since the two spheres per black vertex are defined face-locally by Lemma 10.3 and are generically different from face to face, one must prove that the choice propagates consistently around all cycles of the quad graph. Without that argument, the existence of the associated congruences (c1)φ and (c2)φ is not established. Remark 10.7 sketches a Lie-geometric route, but it does not supply the missing consistency proof.
  4. [§3 and §4, external dependencies] The paper imports Theorem 3.5 and Lemma 4.3 from [ADM+24] and uses them at essential points: Theorem 3.5 underlies the definition of maximal congruences and the passage between congruences and S-isothermic nets, while Lemma 4.3 is used in the proof of Theorem 11.2 to replace the Euclidean projection by an arbitrary spacelike plane. These are legitimate prior results, but the present manuscript provides no independent check or statement of the exact hypotheses needed. The main text should explicitly delineate which imported statements are load-bearing and flag that the results here inherit any assumptions from Part I.
minor comments (6)
  1. [§8] There is a typo in the third display paragraph: 'In clonclusion' should read 'In conclusion.'
  2. [§5, Remark 5.4] The characterization of isothermic incircular nets that come from Koebe congruences is deferred with 'This can be shown' and no proof or precise reference to Part I. Since Remark 7.3 builds on it, either supply the argument or cite the exact location in [ADM+24].
  3. [Definition 10.1] The notation for contact congruences omits the black-vertex component that appears in Definition 3.1 for null congruences; please clarify how the two definitions are compared, especially in Theorem 10.4.
  4. [§9, Equation (9.4)] The sign convention for the ± in the dual edge directions is used repeatedly but is stated only informally; a single unified statement of the horizontal/vertical sign rule would improve readability.
  5. [§7, after Theorem 7.2] The phrase 'the normal map' may suggest an exact equality, whereas the smooth analogue in Theorem 2.2 holds only up to scaling and translation; please clarify the normalization used in the discrete setting.
  6. [Remarks 10.6, 10.7, 11.5] Several statements are explicitly left open ('It is not clear', 'currently unclear'). This is honest, but the paper would benefit from a short paragraph distinguishing theorems, conditional results, and conjectures.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the associated-family construction and the X-variable theorem are derived, not assumed, and the authors' Part I results are legitimate parameter-free prior work.

full rationale

The derivation chain is: Koebe congruence (existence by the external variational principle [BS04]) -> Christoffel dual -> maximal congruence by Definition 7.1 (an analog of Theorem 2.2, not an assumption of the existence claim) -> associated family by integrating the closed rotated differential d⊙hφ (Theorem 9.1, proved in the text) -> associated null congruences via Lemma 10.3 and Corollary 10.5 (proved in the text) -> X-variable constancy via Lemma 11.1 (proved in the text). The only imported load-bearing ingredients are Theorem 3.5 (2:1 correspondence between isothermic congruences and S-isothermic nets) and Lemma 4.3 (X-variables as ratios of squared Lorentz distances), both from the authors' Part I [ADM+24]. These are parameter-free prior statements whose assumptions do not include the present target results; they are checkable independently of this paper, and they do not by themselves assert constancy of X-variables or the existence of maximal s-embeddings. No fitted quantity is renamed a prediction; no uniqueness theorem from the same authors is used to preclude alternatives; and the associated-family argument does not presuppose its conclusion. The proof of Theorem 11.2 may contain a completeness gap (the step from φ-independence of |Zi − P*0| to φ-independence of the complex cross-ratio does not explicitly justify that the Zi rotate together about P*0), but a missing rotation argument is a correctness issue, not a circularity, because the conclusion is not an input of the proof. Hence no circular step is identified.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central construction rests on an external existence theorem for circle patterns, the correspondence and Christoffel-dual machinery inherited from the companion paper [ADM+24] and from [BHS06], and a standard local-to-global integrability assumption. No free parameters are fitted to data; the only continuous parameter, φ, is a construction parameter for the associated family, not a fitted quantity.

assumptions (4)
  • domain assumption Hyperbolic orthogonal circle patterns can be constructed uniquely from boundary data via a convex variational principle.
    Invoked in Section 5 to define Koebe congruences and thus maximal surfaces; the paper relies on [BS04] for existence and uniqueness.
  • domain assumption Every isothermic congruence corresponds to a unique S-isothermic net, and every S-isothermic net corresponds to two isothermic congruences (Theorem 3.5).
    This is the central correspondence from the companion paper [ADM+24]; it is used throughout Sections 5, 7, and 10 without reproof.
  • domain assumption The Christoffel dual of an S-isothermic net exists and is itself an S-isothermic net.
    Used in Sections 6 and 7 to define maximal surfaces via the dual being Koebe; the closedness and integrability of the dual differential is taken from [BHS06].
  • domain assumption The local closing condition around each vertex implies the integrability of the associated differential on the whole quad graph.
    The proof of Theorem 9.1 is reduced to a local check; the reduction from local to global integrability is a standard fact for quad-graphs but is not spelled out.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Discrete Lorentz surfaces and s-embeddings II: maximal surfaces." pith.science (2026). https://pith.science/paper/HRCTSEMU

@misc{pith2026241119055,
  author       = {Pith},
  title        = {Pith review of: Discrete Lorentz surfaces and s-embeddings II: maximal surfaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HRCTSEMU}},
  note         = {Machine review of arXiv:2411.19055}
}
read the original abstract

S-embeddings were introduced by Chelkak as a tool to study the conformal invariance of the thermodynamic limit of the Ising model. Moreover, Chelkak, Laslier and Russkikh introduced a lift of s-embeddings to Lorentz space, and showed that in the limit the lift converges to a maximal surface. They posed the question whether there are s-embeddings that lift to maximal surfaces already at the discrete level, before taking the limit. We answer this question in the positive. In a previous paper we identified a subclass of s-embeddings--isothermic s-embeddings--that lift to (discrete) S-isothermic surfaces, which were introduced by Bobenko and Pinkall as a discretization of isothermic surfaces. In this paper we identify a special class of isothermic s-embeddings that correspond to discrete S-maximal surfaces, translating an approach of Bobenko, Hoffmann and Springborn introduced for discrete S-minimal surfaces in Euclidean space. Additionally, each S-maximal surface comes with a 1-parameter family of associated surfaces that are isometric. This enables us to obtain an associated family of s-embeddings for each maximal s-embedding. We show that the Ising weights are constant in the associated family.

Figures

Figures reproduced from arXiv: 2411.19055 by the authors.

Figure 1
Figure 1. The null-spheres c• (left) and time-like spheres c◦ (right) of a null congruence c as well as the isotropic lines c (red). Definition 3.3. A (spacelike Lorentz) isothermic congruence is a null congruence c such that ⊙c◦ is a spacelike conjugate net. ■ Isothermic congruences were introduced in [ADM+24], and are of particular use because they are in correspondence with the following known object from discrete differen… view at source ↗
Figure 2
Figure 2. Left: a circle of h• (blue) of an S-isothermic net h and the four adjacent timelike Lorentz spheres of h◦ (teal), as well as the isotropic lines of c 1 (green) and c 2 (red) of the two corresponding isothermic congruences c 1 , c 2 . Right: An isothermic incircular net p 1 (green) and the second incircular net p 2 (green) given by the other tangents. We begin with incircular nets, which are essentially quad nets in … view at source ↗
Figure 3
Figure 3. Local Christoffel dual h ∗ of an S-isothermic net h. Corresponding edges (black and brown) are parallel. Remark 5.3. Let h be a Koebe isothermic net. At each face f there is a unique tangent T(f) to the two adjacent circles of h•, which is also a tangent to U+. Around a white vertex w, the adjacent tangents intersect in the point ⊙h◦(w). Viewed in this way, as an edge-tangent conjugate net, ⊙h◦ is called a Koebe pol… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Left: the circles of a Koebe isothermic net. Right: the circles of a discrete maximal [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Local configuration in a Koebe isothermic net [PITH_FULL_IMAGE:figures/full_fig_p013_5.png]
Figure 6
Figure 6. Figure 6: The frames (Ti , n, bi) and (Ti , Ni , Bi) used in the calculations for the associated family. section we study how the discrete construction translates to maximal congruences in Lorentz space and the corresponding incircular nets in E2 . Let c be a Koebe congruence an…
Figure 7
Figure 7. Figure 7: Congruences in the associated family of a maximal surface, from top to bottom for [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]
Figure 8
Figure 8. Figure 8: Planar nets obtained from the associated congruences in the associated family of a [PITH_FULL_IMAGE:figures/full_fig_p020_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 19 canonical work pages

  1. [1]

    Discrete Lorentz surfaces and s-embeddings I: isothermic surfaces

    Niklas C. Affolter, Felix Dellinger, Christian Müller, Denis Polly, and Nina Smeenk. Discrete Lorentz surfaces and s-embeddings I: isothermic surfaces , 2024. Preprint, arXiv:2410.11575

    work page Pith review arXiv 2024

  2. [2]

    Bobenko and Tim Hoffmann

    Alexander I. Bobenko and Tim Hoffmann. S-conical cmc surfaces. towards a unified theory of discrete surfaces with constant mean curvature. In Alexander I. Bobenko, editor, Advances in Discrete Differential Geometry , pages 287--308. Springer Berlin Heidelberg, Berlin, Heidelberg, 2016

    work page 2016

  3. [3]

    Bobenko, Tim Hoffmann, and Boris A

    Alexander I. Bobenko, Tim Hoffmann, and Boris A. Springborn. Minimal surfaces from circle patterns: G eometry from combinatorics. Ann. Math. (2) , 164(1):231--264, 2006

    work page 2006

  4. [4]

    Constant mean curvature surfaces from ring patterns: Geometry from combinatorics

    Alexander I. Bobenko, Tim Hoffmann, and Nina Smeenk. Constant mean curvature surfaces from ring patterns: G eometry from combinatorics, 2024. Preprint, arXiv:2410.08915

    work page Pith review arXiv 2024

  5. [5]

    Bobenko and Ulrich Pinkall

    Alexander I. Bobenko and Ulrich Pinkall. Discretization of surfaces and integrable systems. In Alexander I. Bobenko and Ruedi Seiler, editors, Discrete integrable Geometry and Physics , pages 3--58. Clarendon Press, Oxford, 1999

    work page 1999

  6. [6]

    Bobenko, Helmut Pottmann, and Johannes Wallner

    Alexander I. Bobenko, Helmut Pottmann, and Johannes Wallner. A curvature theory for discrete surfaces based on mesh parallelity. Mathematische Annalen , 348(1):1--24, Sep 2010

    work page 2010

  7. [7]

    Bobenko and Boris A

    Alexander I. Bobenko and Boris A. Springborn. Variational principles for circle patterns and K oebe's theorem. Trans. Amer. Math. Soc. , 356(2):659--689, 2004

    work page 2004

  8. [8]

    Bobenko and Yuri B

    Alexander I. Bobenko and Yuri B. Suris. Discrete differential geometry: Integrable structure , volume 98 of Graduate studies in mathematics . American Math. Soc., 2008

    work page 2008

Show all 20 references
  1. [9]

    Planar I sing model at criticality: state-of-the-art and perspectives

    Dmitry Chelkak. Planar I sing model at criticality: state-of-the-art and perspectives. In Proceedings of the I nternational C ongress of M athematicians--- R io de J aneiro 2018. V ol. IV . I nvited lectures , pages 2801--2828. World Sci. Publ., Hackensack, NJ, 2018

    2018

  2. [10]

    Bipartite dimer model: perfect t-embeddings and Lorentz-minimal surfaces , 2021

    Dmitry Chelkak, Beno \^i t Laslier, and Marianna Russkikh. Bipartite dimer model: perfect t-embeddings and Lorentz-minimal surfaces , 2021. Preprint, arXiv:2109.06272

    2021 arXiv

  3. [11]

    Dimer model and holomorphic functions on t-embeddings of planar graphs

    Dmitry Chelkak, Beno \^i t Laslier, and Marianna Russkikh. Dimer model and holomorphic functions on t-embeddings of planar graphs. Proceedings of the London Mathematical Society , 126(5):1656--1739, 2023

    2023

  4. [12]

    Adam Doliwa and Paolo M. Santini. Multidimensional quadrilateral lattices are integrable. Physics Letters A , 233(4):365 -- 372, 1997

    1997

  5. [13]

    Dimers and circle patterns

    Richard Kenyon, Wai Yeung Lam, Sanjay Ramassamy, and Marianna Russkikh. Dimers and circle patterns . Annales Scientifiques de l' \'E cole Normale Sup \'e rieure , 55(3):863--901, 2022

    2022

  6. [14]

    Maximal surfaces in the 3-dimensional M inkowski space L^3

    Osamu Kobayashi. Maximal surfaces in the 3-dimensional M inkowski space L^3 . Tokyo Journal of Mathematics , 06(2):297--309, 1983

    1983

  7. [15]

    Geometric modeling with conical meshes and developable surfaces

    Yang Liu, Helmut Pottmann, Johannes Wallner, Yongliang Yang, and Wenping Wang. Geometric modeling with conical meshes and developable surfaces. ACM Transactions on Graphics , 25(3):681--689, 7 2006

    2006

  8. [16]

    Weierstrass-type representations

    Mason Pember. Weierstrass-type representations. Geometriae Dedicata , 204(1):299--309, 2020

    2020

  9. [17]

    Wackelige Kurvennetze bei einer infinitesimalen Fl \"a chenverbiegung

    Robert Sauer. Wackelige Kurvennetze bei einer infinitesimalen Fl \"a chenverbiegung . Mathematische Annalen , 108(1):673--693, 1933

    1933

  10. [18]

    Circle patterns with the combinatorics of the square grid

    Oded Schramm. Circle patterns with the combinatorics of the square grid. Duke Math. J. , 86(2):347--389, 1997

    1997

  11. [19]

    Wolfgang K. Schief. On the unification of classical and novel integrable surfaces. II . D ifference geometry. R. Soc. Lond. Proc. Ser. A Math. Phys. Eng. Sci. , 459(2030):373--391, 2003

    2003

  12. [20]

    Wolfgang K. Schief. On a maximum principle for minimal surfaces and their integrable discrete counterparts. Journal of Geometry and Physics , 56(9):1484--1495, 2006

    2006

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.