Pith. sign in

REVIEW 4 major objections 5 minor 12 references

Local asymptotics near a smooth boundary point and estimates for a hyperbolic-type metric

T0 review · 4 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read Near any C^1 boundary point, the metric m_D equals the explicit formula ψ_D in the limit.

desk verdict A plausible and genuinely new local asymptotic for m_D, but the proof leans on an unpublished equivalence with unstated constants, and the statement mishandles unbounded domains. read the letter →

arxiv 2607.26524 v1 pith:3IGKFZPH submitted 2026-07-29 math.MG math.CV

Pritam Naskar This is my paper · ORCID
classification math.MGmath.CV MSC 30F4530L1551K0530C6530L1051M10
keywords hyperbolic-typemetricsquasihyperbolicmetricuniformdomainsdistanceratioinnerC^1boundaryasymptoticspath-integrated
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's central claim is that the path-integrated hyperbolic-type metric m_D — the infimum of ∫ d(D)/η_D ds over curves — has the explicit quantity ψ_D = 2 sinh⁻¹(d(D)|x−y|/(2√(η_D(x)η_D(y)))) as its local asymptotics at every C^1-smooth boundary point: the ratio of the two tends to 1 as both points approach the boundary. Because ψ_D is a closed-form distance-ratio expression, this gives a concrete handle on a metric that is normally defined through an optimization over paths. The paper also builds a companion distance-ratio metric i*_D from the same function η_D, proves that it is a genuine metric, locates it between ζ_D and m_D, and shows that m_D is its inner metric; the result is an improved lower bound for m_D. In unbounded domains the construction recovers the classical chain of inequalities among the quasihyperbolic and distance-ratio metrics.

What carries the argument

The machinery is ψ_D and η_D. The density of m_D is d(D)/η_D, where η_D(z)=δ_D(z)(d(D)−δ_D(z)), and ψ_D is the closed form whose infinitesimal version gives that density. The proof of the main theorem uses a boundary-flattening map θ that is c-bi-Lipschitz with c arbitrarily close to 1, plus the exact computability of m_D in the half-space; the comparison chain passes m_D through θ(D*) and the half-space to ψ_D with errors that vanish. The second half of the paper uses the quantity i*_D(x,y)=2 log((η_D(x)+η_D(y)+d(D)|x−y|)/(2√(η_D(x)η_D(y)))) and the fact that η_D is d(D)-Lipschitz; the Lipschitz bound is exactly what gives i*_D its triangle inequality and what forces the derivative of i*_D

What would settle it

Take a smooth domain with nonconstant boundary curvature, for example a small perturbation of the unit ball, and numerically compute m_D (via its definition as an infimum over paths) and ψ_D for pairs approaching a boundary point along different directions. If the ratio m_D/ψ_D has a sequential limit other than 1, Theorem 3.1 is false. The unit ball itself is a degenerate test case because m_D is exactly the hyperbolic metric there, so a nontrivial domain is needed.

Watch

Extended reading notes

Core claim

Theorem 3.1: if u is a C^1-smooth boundary point of a proper subdomain D ⊂ R^n and x,y → u with x ≠ y, then m_D(x,y)/ψ_D(x,y) → 1. In other words, near smooth boundary points m_D is asymptotically the compact formula ψ_D, which simultaneously generalises the classical hyperbolic metric formulas for the disk and half-plane. The proof flattens the boundary with a nearly-isometric change of variables, restricts to a small cap near u, uses localization of m_D-geodesics, and compares m_D with the exact half-space case; two-sided estimates with constants tending to 1 then squeeze the ratio.

Load-bearing premise

The argument assumes that the equivalence between m_D and the quasihyperbolic metric from an unpublished companion preprint is strong enough to keep m_D-geodesics inside a prescribed neighbourhood of their endpoints; if that equivalence is weaker than claimed, the boundary-flattening comparison used to prove Theorem 3.1 would no longer go through.

Editorial extensions

If this is right

  • For any C^1 boundary point, local values of m_D can be evaluated as ψ_D up to a factor tending to 1; no curve optimization is needed in the limit.
  • Corollary 3.2 turns this into a uniform approximation on bounded domains with C^1 boundary, so ψ_D serves as a boundary-valid asymptotic model for m_D.
  • Because ζ_D ≤ i*_D ≤ m_D, the new metric gives a concrete lower bound for m_D that is sharper than ζ_D alone.
  • Since m_D is the inner metric of i*_D, one can approximate m_D distances by sums of i*_D increments along short arcs.
  • In the unbounded setting the same construction reduces to the classical chain j_D ≤ i_D ≤ k_D, so the main theorem is a generalisation rather than a parallel theory.

Reading between the lines

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

  • The asymptotic relation m_D ~ ψ_D at smooth points is stated with no rate; a natural next step, not attempted here, would be to quantify the error term in terms of the modulus of continuity of the boundary tangent.
  • The proof's localization step (the Observation following Lemma 2.1) derives its force from an equivalence between m_D and the quasihyperbolic metric stated in an unpublished companion preprint. Writing that equivalence with explicit constants would turn the theorem into a fully self-contained statement.
  • Since two distinct metrics, ζ_D and i*_D, can share m_D as their inner metric, the inner-metric operation is non-injective; this raises the question of what additional data identify a metric up to equality, not just equivalence.
  • The same boundary-limit technique may extend to finitely many boundary singularities, where local flattening fails but the cap replacement could still be controlled.
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 / 5 minor

Summary. The paper studies the path-integrated hyperbolic-type metric m_D defined in (1.3) by the density d(D)/η_D, where η_D = δ_D(d(D)−δ_D). Its main result, Theorem 3.1, claims that near any C^1-smooth boundary point u of a proper subdomain D⊂R^n, the ratio m_D/ψ_D tends to 1, with ψ_D the explicit expression 2sinh^{-1}(diam(D)|x−y|/(2√(η_D(x)η_D(y)))). The proof uses a boundary-flattening map θ, a localization assertion for m_D-geodesics, and comparisons with the half-space metric. The paper also introduces a Nikolov–Andreev-type metric i*_D, proves it is a metric, establishes sharp two-sided comparisons with ζ_D, and claims that m_D is its inner metric, yielding the lower bound i*_D≤m_D.

Significance. If Theorem 3.1 is correct, it gives a genuinely closed-form local asymptotic for a recently introduced path-integrated metric, unifying the classical ball and half-space formulas in a single expression ψ_D. The i*_D results provide a new companion metric with sharp constants and a clean infinitesimal density. The manuscript has concrete strengths: ψ_D contains no fitted parameters; the sharp constants 1 and 2 in Proposition 3.7 are computed against ζ_D, not chosen to match; and the exact unit-disk test passes. However, the main theorem rests on a localization statement whose proof is only a citation to an unpublished preprint by the present author, and several definitions are not well-posed for unbounded domains. These issues must be resolved before the central claim can be considered established.

major comments (4)
  1. [§1, Eqs. (1.3)–(1.5); Theorem 3.1] For unbounded D, d(D)=∞, so η_D(z)=δ_D(z)(d(D)−δ_D(z)) is infinite and the density d(D)/η_D(z) is an undefined ∞/∞ expression. Likewise ψ_D in (1.5) contains ∞/∞ in its argument. Theorem 3.1 is stated for an arbitrary proper subdomain of R^n, which includes unbounded domains, and the proof explicitly uses the half-space H1. The author writes that the unbounded case 'naturally coincides' with the quasihyperbolic density, but this is not a definition. Please either introduce a limiting convention d(D)→∞, or restrict the theorem to bounded domains and treat the half-space separately.
  2. [Lemma 2.1 and the following Observation; proof of Theorem 3.1, Eq. (3.3)] The localization of m_D-geodesics is load-bearing. Lemma 2.1 is proved only by citing [7, Theorem 3.5], an unpublished preprint coauthored by the present author, and the equivalence constants are not stated. Since an m_D-geodesic is not a k_D-geodesic, transferring a k_D length bound to γ requires a quasi-geodesic argument with explicit constants. Moreover, those constants must be uniform as the neighbourhood U shrinks, because the comparison m_D≤m_{D*}≤(1−ε1)^{-1}m_D in (3.3) and the later replacement of D by D* use localization with constants tending to 1. Without a proof of Lemma 2.1 (or a published reference with stated constants), the Observation is unsupported and the main asymptotic does not follow from the given argument.
  3. [Proof of Theorem 3.1, paragraph after Eq. (3.3)] The inequality m_{θ(D*)}(θ(x),θ(y)) ≤ (1−ε2)^{-1}m_{H1}(θ(x),θ(y)) is obtained by integrating along an H1-geodesic γ. This is valid only if γ⊂θ(D*). The manuscript says 'using a similar argument as above', but the previous argument concerns m_D-geodesics, not H1-geodesics. No proof is given that a hyperbolic geodesic in H1 with endpoints in θ(D*) remains inside θ(D*). If γ leaves θ(D*), the integral over γ of d(θ(D*))/η_{θ(D*)} is not defined and the comparison collapses. This is a central gap in the proof of Theorem 3.1, even if it may be repairable with a separate argument.
  4. [Proposition 3.9 and Theorem 3.5] The conclusion that m_D is the inner metric of i*_D, and the resulting inequality i*_D≤m_D, requires more than the pointwise limit lim_{y→x} i*_D(x,y)/|x−y|=d(D)/η_D(x). To pass from the triangle inequality to the length integral one needs local uniform convergence of this quotient on rectifiable curves; this is not stated or proved. The statement is also undefined for unbounded domains, as in the first comment. Theorem 3.5 should either provide the missing convergence argument or state the standard length-space lemma it is invoking.
minor comments (5)
  1. [Eq. (1.6)] The displayed formula for i*_D is typeset ambiguously; the fraction and the logarithm should be shown as in Proposition 3.3.
  2. [Proof of Proposition 3.7] In the lower-bound chain, replacing √((1−δ(x)/d)(1−δ(y)/d)) by (1−ε1) is not literally correct; the product is at least (1−ε1)^2, not (1−ε1). This is fixable by redefining ε1 or using 1−2ε1, and does not affect the limit.
  3. [Corollary 3.2] The statement 'lim_{x→∂D} ... uniformly in x≠y' is ambiguous. It should specify whether the limit is taken as x,y→u∈∂D with x≠y, and what uniformity means.
  4. [Observation after Lemma 2.1] The neighbourhood V is introduced but not used precisely. The intended statement should be: for every sufficiently small U there exists V⊂U such that any m_D-geodesic with endpoints in V lies in U, with a proof based on Lemma 2.1.
  5. [Proof of Theorem 3.1] After the boundary-flattening map θ, the quantity δ'_D is used without a definition; it should be defined as the distance to the relevant boundary in the transformed coordinates. Also, the sentence 'we omit a few similar steps' hides a nontrivial comparison; please spell it out.

Circularity Check

1 steps flagged · score 4.0 of 10

Theorem 3.1 is not fitted, but its localization step (3.3) is carried by an unpublished self-citation [7, Thm 3.5] with unstated constants.

  1. self citation load bearing [Section 2, Lemma 2.1 and the following Observation; used in Theorem 3.1 proof at equation (3.3)]
    "The proof follows from the direct use of the equivalence of the quasihyperbolic and the mD-metric [7, Theorem 3.5] in place of [2, eq. (2.18)] and the consequent discussion in the same paper of Gehring and Osgood."

    Lemma 2.1 is the only support for the Observation that m_D-geodesics remain localized near a C^1 boundary point. That localization is needed in Theorem 3.1 at (3.3): the upper bound m_D*(x,y) ≤ (1−ε_1)^{-1} m_D(x,y) compares the infimum over paths in the cap D* with the infimum over all of D. Without knowing that an m_D-geodesic between x,y stays in D*, the infimum over D* is not bounded above by the infimum over D. The cited [7] is an unpublished preprint coauthored by the present author, and the constants of the m_D/k_D equivalence are not stated, so the transfer of Gehring–Osgood length bounds from k_D-geodesics to m_D-geodesics is not verifiable from the paper.

full rationale

The asymptotic formula itself is not circular in the fitted-input sense: ψ_D is pinned by the external ball and half-space hyperbolic metrics, m_D's density is fixed independently, and the sharp constants 1 and 2 in Proposition 3.7 are computed against ζ_D rather than tuned to match. The circularity burden is isolated in one load-bearing step. Lemma 2.1 and the subsequent Observation assert that m_D-geodesics remain localized near a smooth boundary point, and the proof is deferred entirely to the author's unpublished preprint [7, Theorem 3.5] via an equivalence m_D ≍ k_D with unstated constants. This localization is then used in the proof of Theorem 3.1 to justify replacing D by the cap D* in equation (3.3). Since an m_D-geodesic is not a k_D-geodesic, the transfer of the Gehring–Osgood Euclidean length bound requires the missing equivalence constants and a quasi-geodesic argument; absent those, the comparison (3.3) and hence the limit in Theorem 3.1 do not follow from the written proof. This is a genuine self-citation dependency rather than independent, checkable support. There is also a well-posedness caveat for unbounded D, where ψ_D involves an undefined ∞/∞ expression, but that is a correctness issue, not a circularity. Overall the central claim retains independent content and is not forced by definition, so the score is moderate.

Assumptions & free parameters 0 free parameters · 4 assumptions · 2 invented entities

No free parameters are fitted: auxiliary constants c, ε1, ε2 in Theorem 3.1 are limiting parameters tending to the identity, not data fits. The main axioms the paper pulls in without proof are (i) the m_D ≍ k_D equivalence from the author's own unpublished preprint [7], (ii) the Lipschitz property of η_D cited to [8], (iii) the standard half-space hyperbolic formula, and (iv) the local uniformity of C^1 boundaries.

assumptions (4)
  • domain assumption Equivalence of m_D and the quasihyperbolic metric k_D on D (m_D ≍ k_D), [7, Theorem 3.5].
    Used in Lemma 2.1 to transfer quasihyperbolic geodesic localization to m_D-geodesics; load-bearing for the boundary-flattening proof of Theorem 3.1. It comes from an unpublished arXiv preprint coauthored by the present author, not independently verified here.
  • domain assumption η_D(z) = δ_D(z)(d(D) − δ_D(z)) is d(D)-Lipschitz, [8, Lemma 3.3(iii)].
    Used in Prop 3.3 (triangle inequality), Prop 3.7, and Prop 3.9. Easy to prove directly from the 1-Lipschitz property of δ_D, but cited to the author's own [8].
  • standard math The hyperbolic metric of the half-space is h_H(x,y) = 2sinh⁻¹(|x−y|/(2√(δ(x)δ(y)))), and m_D = k_D = h_D for balls and half-spaces.
    Invoked in Theorem 3.1 ('the fact that the metric m_D matches ψ_D when D is a half-plane') without proof; standard (Katok [5], Gehring–Osgood [2]), but the matching of ψ's limiting form for unbounded domains rests on the limiting convention d(D) = ∞.
  • domain assumption A C^1-smooth boundary point is locally uniform, so m_D-geodesics stay in small neighbourhoods (Observation after Lemma 2.1).
    Standard for C^1 domains; used to localize the comparison argument in Theorem 3.1.
invented entities (2)
  • ψ_D compact formula (Eq. 1.5) independent evidence
    purpose: Explicit closed-form candidate for the local asymptotic of the path-integral metric m_D near smooth boundary points.
    Exactly reproduces the hyperbolic metric formulas in the ball and half-space (external benchmarks) and passes the exact unit-disk numerical check against h_D.
  • i*_D metric (Eq. 1.6) independent evidence
    purpose: Bounded-domain analogue of the Nikolov–Andreev metric; serves as a sharp lower bound for m_D.
    Recovers the known Nikolov–Andreev metric i_D in the unbounded limit; sharpness of the constants 1 and 2 in Prop 3.7 is checked against ζ_D in the unit disk.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Local asymptotics near a smooth boundary point and estimates for a hyperbolic-type metric." pith.science (2026). https://pith.science/paper/3IGKFZPH

@misc{pith2026260726524,
  author       = {Pith},
  title        = {Pith review of: Local asymptotics near a smooth boundary point and estimates for a hyperbolic-type metric},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3IGKFZPH}},
  note         = {Machine review of arXiv:2607.26524}
}
abstract

In this paper, we investigate the local boundary behaviour of a recently developed hyperbolic-type metric $m_D$. First, employing a boundary-flattening technique and local behaviour of $m_D$-geodesics, we establish its asymptotic formula near any $C^1$-smooth boundary point. Next, we introduce a metric quantity analogous to the Nikolov--Andreev metric and show that $m_D$ is the inner metric associated with it. Finally, by establishing a sharp two-sided comparison inequality, we obtain an improved lower bound for the $m_D$-metric.

Figures

Figures reproduced from arXiv: 2607.26524 by the authors.

Figure 1
Figure 1. Motivation. Secondly, motivated by the Nikolov–Andreev metric, we introduce the following quantity i ∗ D(x, y) := 2 log ηD(x) + ηD(y) + d(D)|x − y| 2 p ηD(x)ηD(y) (1.6) and prove that it is a metric, even if we replace ηD by any positive Lipschitz function. We establish a comparison inequality with ζD and prove that i ∗ D possesses mD as its inner metric. Finally, applying these results, we obtain an improved lower … view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 1 linked inside Pith

  1. [1]

    Dovgoshey, P

    O. Dovgoshey, P. Hariri and M. Vuorinen, Comparison theorems for hyperbolic type metrics,Complex Var. Elliptic Equ.61(2016), no. 11, 1464–1480

    2016

  2. [2]

    F. W. Gehring and B. G. Osgood, Uniform domains and the quasihyperbolic metric,J. Analyse Math.36(1979), 50–74

    1979

  3. [3]

    F. W. Gehring and B. P. Palka, Quasiconformally homogeneous domains,J. Analyse Math.30(1976), 172–199

    1976

  4. [4]

    Hariri, R

    P. Hariri, R. Kl´ en and M. Vuorinen,Conformally invariant metrics and quasiconformal mappings, Springer Monographs in Mathematics, Springer, Cham, 2020

    2020

  5. [5]

    Katok,Fuchsian Groups, Chicago Lectures in Mathematics, University of Chicago Press, 1992

    S. Katok,Fuchsian Groups, Chicago Lectures in Mathematics, University of Chicago Press, 1992

    1992

  6. [6]

    Q. Luo, A. Rasila, Q. Zhou The Nikolov–Andreev metric and Gromov hyperbolicity,Mediterr. J. Math.21(2024), no. 3, Paper No. 105, 14 pp

    2024

  7. [7]

    B. Maji, P. Naskar, and S. K. Sahoo. A naive generalization of the hyperbolic and the quasihyperbolic metrics. arXiv:2505.10964v2

    arXiv

  8. [8]

    B. Maji, P. Naskar, and S. K. Sahoo, Comparison and distortion properties of new distance ratio metrics,J. Math. Anal. Appl.559(2026), no. 2, Paper No. 130480, 22 pp

    2026

Show all 12 references
  1. [9]

    Mocanu, Generalizations of four hyperbolic-type metrics and Gromov hyperbolicity,J

    M. Mocanu, Generalizations of four hyperbolic-type metrics and Gromov hyperbolicity,J. Math. Anal. Appl.552 (2025), no. 1, Paper No. 129729, 16 pp

    2025

  2. [10]

    Nikolov and L

    N. Nikolov and L. Andreev, Estimates of the Kobayashi and quasi-hyperbolic distances,Ann. Mat. Pura Appl.(4) 196(2017), no. 1, 43–50

    2017

  3. [11]

    Nikolov and P

    N. Nikolov and P. Thomas, Boundary behavior of the quasi-hyperbolic metric,Ann. Acad. Sci. Fenn. Math.43 (2018), no. 1, 381–389

    2018

  4. [12]

    Q. Zhou, Z. Zheng and P. Huo, The Nikolov–Andreev metric and quasim¨ obius maps in Banach spaces,Mediterr. J. Math.22(2025), no. 8, Paper No. 197, 15 pp. Pritam Naskar, Department of Mathematics, Indian Institute of Technology Indore, Simrol, Indore, Madhya Pradesh 453552, Ind...

    2025

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.