REVIEW 1 major objections 5 minor
A short note on the dimension of the CC-geodesic Kakeya sets in the first Heisenberg group
T0 review · 1 major / 5 minor · reviewed 2026-07-30 · grok-4.5
Pith's one-line read CC-geodesic Kakeya sets in the first Heisenberg group have sharp Heisenberg dimension 3, not the expected 4.
desk verdict Clean negative answer to Lukyanenko’s question: full CC-geodesic Kakeya sets in H¹ still only force Heisenberg dimension 3, even when compact, and the constructions are elementary and checkable. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The surface Σ parametrized by Φ(r,φ)=(r cos φ, r sin φ, r²φ/2), together with the horizontal plane Π. Suitable left translates of every nonzero-curvature unit geodesic land inside Σ, while the zero-curvature segments sit in Π; their union is therefore a CC-geodesic Kakeya set of Heisenberg dimension 3.
What would settle it
Exhibit a CC-geodesic Kakeya set whose Heisenberg Hausdorff dimension is strictly less than 3, or show that the constructed surface Σ fails to contain a left translate of some nonzero-curvature unit geodesic.
Extended reading notes
Core claim
Every CC-geodesic Kakeya set in the first Heisenberg group has Heisenberg Hausdorff dimension at least 3, and this lower bound is attained by an explicit Borel set (and even by a compact one). Consequently the natural full-dimension-4 conjecture fails. Separately, for each fixed curvature κ in (0,2π] there exist compact curvature-κ Kakeya sets of Euclidean and Heisenberg dimension exactly 1.
Load-bearing premise
The dimension comparison that every smooth two-dimensional Euclidean surface inside the Heisenberg group has Heisenberg Hausdorff dimension exactly 3.
Editorial extensions
If this is right
- The CC-geodesic Kakeya conjecture in H¹ is false at the level of Hausdorff dimension; the sharp lower bound is 3 rather than 4.
- Sharpness holds already inside the compact class, so unboundedness is not needed to reach dimension 3.
- Full dimension 4 is still compatible with zero Lebesgue measure for CC-geodesic Kakeya sets.
- Fixing a single positive curvature collapses the problem dramatically: compact examples of dimension 1 exist.
- Horizontal-only Kakeya lower bounds already control the full geodesic family; extra curvature does not improve the constant.
Reading between the lines
- The fixed-curvature dimension-1 examples suggest that any attempt to restore a dimension-4 conjecture would need to quantify over a positive-measure set of curvatures, not a single κ.
- The same surface-plus-plane construction may adapt to higher-step or higher-dimensional Heisenberg groups once the geodesic equations are known explicitly.
- Because the sharpness example is a union of two Euclidean surfaces, quantitative Kakeya maximal inequalities in H¹ cannot improve beyond the horizontal case without new geometric input.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies CC-geodesic Kakeya sets in the first Heisenberg group H¹: Borel sets containing a left translate of every unit-speed minimising CC-geodesic segment of length 1 issuing from the identity. The natural analogue of the Kakeya conjecture would require Heisenberg Hausdorff dimension 4. The authors prove that every such set has dim_H^{H¹} ≥ 3 (inherited from the horizontal case K_CC ⊂ K({0})), and that 3 is sharp, both in general (Theorem 1.1, via the explicit union Σ ∪ Π of two smooth Euclidean surfaces) and among compact sets (Theorem 1.2, via a compact packing E_c = D² ∪ ⋃_j E_j of geodesic arcs). They also produce a CC-geodesic Kakeya set of full dimension 4 and Lebesgue measure zero (Corollary 1.3), and, for each fixed curvature κ ∈ (0, 2π], a compact curvature-κ Kakeya set of Euclidean and Heisenberg Hausdorff dimension 1 (Theorem 1.4).
Significance. The work answers Lukyanenko’s open question and shows that the natural full-dimension prediction fails: nonzero-curvature geodesics do not raise the sharp lower bound above the horizontal value 3. Sharpness is obtained by fully explicit left-translation constructions (Lemma 3.1 and the screw-motion identity (4.3)) whose dimension upper bounds rest only on standard comparison theorems for C¹ surfaces and Lipschitz images in H¹. The fixed-curvature near-miss at dimension 1 is a clean and striking counterpart. The arguments are elementary, self-contained, and free of circularity; the contribution is solid and appropriate for a serious journal in geometric measure theory / harmonic analysis.
major comments (1)
- [Abstract] Abstract vs. Theorems 1.1–1.2: the abstract states that “their sharp lower bound is 4”, while the body (Introduction, Theorem 1.1, Theorem 1.2, and the constructions) consistently proves that the sharp lower bound is 3 and that the prediction of 4 fails. This is a load-bearing misstatement of the main result and must be corrected to “3” before publication; as written, the abstract contradicts the theorems.
minor comments (5)
- [Abstract] In the abstract, insert the missing space in “is4” (and likewise check “dimension4”, “measure0”, etc., for consistent spacing).
- [Section 1] Page 1 / Introduction: the sentence “A fact that at first glance supports this conjecture is that, if one uses only horizontal unit segments… then the sharp lower bound… is 3” is slightly awkward; a brief clarifying clause that horizontal segments are the k = 0 members of the family would help non-specialist readers.
- [Lemma 3.1] Lemma 3.1: the two cases k > 0 and k < 0 differ only by a constant phase shift of π in the t-coordinate; a one-line remark that the same computation covers both signs (or a unified formula) would shorten the write-up.
- [Section 4.3] Proof of Theorem 1.2: the removable-singularity claim for F_j at κ = 0 is correct by Taylor expansion, but citing the explicit limiting expressions (or noting that the coordinate functions extend C^∞) would make the Lipschitz bound immediate without appealing to a neighbourhood of the rectangle.
- [References] References: the arXiv number and year of the Wang–Zahl Kakeya paper [26] should be double-checked against the final published version if available at proof stage.
Circularity Check
No significant circularity: explicit geometric constructions with external dimension comparisons
full rationale
This is a self-contained existence-and-dimension paper in geometric measure theory. The lower bound dim ≥ 3 for CC-geodesic Kakeya sets is inherited from the strictly weaker horizontal Kakeya class K({0}) via the inclusion K_CC ⊂ K({0}), which was already proved by Liu and by Fässler–Pinamonti–Wald (independent authors). Sharpness is obtained by writing down explicit left translates (Lemma 3.1 for the non-compact surface Σ ∪ Π; the screw-motion identity (4.3) and length check for the compact union E_c) and then invoking standard external comparison theorems (Balogh–Durand-Cartagena–Fässler–Mattila–Tyson; Falconer) that C¹ Euclidean surfaces and Lipschitz images of 2-dimensional parameter domains have Heisenberg Hausdorff dimension 3. Nothing is fitted, nothing is defined in terms of the target dimension, and there is no load-bearing self-citation chain. Ordinary dependence on the paper’s own definitions of K_CC and K({κ}) is definitional bookkeeping, not circular derivation.
Assumptions & free parameters
assumptions (5)
- domain assumption Unit-speed minimizing CC-geodesics from the identity are exactly the family γ_{θ,k} for θ∈[0,2π), k∈[−2π,2π], with cut time 2π/|k| (Monti; Le Donne notes).
- domain assumption Every E in the horizontal Kakeya class K({0}) (or Liu’s open-segment class) satisfies dim_{H¹}_H(E)≥3.
- domain assumption Every C¹ Euclidean surface in H¹ has Heisenberg Hausdorff dimension 3; more generally the Balogh et al. dimension-comparison principle relating Euclidean and Heisenberg dimensions.
- standard math Hausdorff dimension is non-increasing under Lipschitz maps, and bi-Lipschitz equivalent metrics (d_K ~ d_CC) yield the same Hausdorff dimension.
- standard math L³ is a bi-invariant Haar measure on H¹ ≅ R³; smooth 2-surfaces and the constructed Cantor products are L³-null as claimed.
invented entities (2)
-
CC-geodesic Kakeya set (Definition 2.2 / class K_CC)
independent evidence
-
Curvature-κ Kakeya set (Definition 4.1)
independent evidence
Cite this review
Pith. "Pith review of A short note on the dimension of the CC-geodesic Kakeya sets in the first Heisenberg group." pith.science (2026). https://pith.science/paper/TRUZPCE7
@misc{pith2026260726906,
author = {Pith},
title = {Pith review of: A short note on the dimension of the CC-geodesic Kakeya sets in the first Heisenberg group},
year = {2026},
howpublished = {\url{https://pith.science/paper/TRUZPCE7}},
note = {Machine review of arXiv:2607.26906}
}
abstract
We study CC-geodesic Kakeya sets in the first Heisenberg group, namely Borel sets $E$ such that, for every unit-speed CC-geodesic segment of length \(1\) issuing from the identity, some left translate of the segment is contained in $E$. The natural analogue of the Kakeya conjecture would predict full Heisenberg Hausdorff dimension 4 for such sets. We show that this prediction fails: the sharp lower bound for their Heisenberg Hausdorff dimension is 3, and it remains sharp even among compact CC-geodesic Kakeya sets. By adjoining a Lebesgue-null set of full Heisenberg Hausdorff dimension, we obtain a CC-geodesic Kakeya set of full Heisenberg Hausdorff dimension 4 and zero Lebesgue measure. Finally, when one prescribes only geodesic segments with one fixed nonzero curvature parameter $\kappa$, rather than segments of all curvatures, the condition is weaker. For every $\kappa\in(0,2\pi]$, we construct a compact curvature-$\kappa$ Kakeya set of zero Lebesgue measure whose Euclidean and Heisenberg Hausdorff dimensions are both equal to $1$.
Reviewed July 30, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.