On the Linearity of Extension Operators for Traces of Besov Spaces on Metric Measure Spaces: The Limiting Case

Aleksei Y. Chikalov

REVIEW 2 major objections 5 minor 19 references

On the Linearity of Extension Operators for Traces of Besov Spaces on Metric Measure Spaces: The Limiting Case

T0 review · 2 major / 5 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read The paper proves that, in the limiting case of the Besov trace problem on metric measure spaces, no bounded linear extension operator exists from the L_p trace space back to the Besov space, whenever the Hausdorff measure on the trace set h

desk verdict A genuine new result with a clever p=2 argument, but the theorem's meaning rests entirely on the trace identification imported from the author's companion preprint — referee should check that foundation. read the letter →

arxiv 2607.11604 v2 pith:4ABYCMNP submitted 2026-07-13 math.FA

Aleksei Y. Chikalov This is my paper · ORCID

classification math.FA MSC 46E3546B2546B4542B35

keywords BesovspacestracesmetricmeasureAhlfors–Davidregularsetshyperbolicfillingsextensionoperatorslimitingcasenonlinear

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

For Besov spaces on metric measure spaces, the limiting trace case (where the smoothness index equals the codimension divided by p) identifies the trace space with an L_p space on the boundary set. The paper establishes that, under this identification and mild regularity assumptions, any bounded operator extending L_p functions back into the Besov space must be nonlinear. This generalizes the classical Euclidean result of Burenkov and Gol'dman to arbitrary Ahlfors–David regular subsets of metric measure spaces. The proof uses hyperbolic fillings to embed the Besov space into ℓ₁(ℓ_p); for p ≠ 2 a Banach-space obstruction applies, while for p = 2 a new finite-scale argument is needed, showing that any hypothetical linear extension would yield finite-rank approximations to the identity on an infinite-dimensional space.

What carries the argument

Hyperbolic fillings: the space X is discretized into a graph of balls at dyadic scales, and Besov functions are mapped to their Poisson averages on the filling. This yields an isomorphic embedding of B^{θ/p}_{p,1}(X) into a closed subspace of ℓ₁(ℓ_p) via the discrete sequence space d b^{θ/p}_{p,1}(V). For p ≠ 2, the contradiction comes from the classical fact that ℓ₂ does not embed into ℓ₁(ℓ_p). For p = 2, a tail estimate (Lemma 4.2) shows that truncating the filling at scale N produces an operator approximating the identity in operator norm, which is impossible because the truncated operator has finite rank on an infinite-dimensional L₂(A).

What would settle it

The central claim would be falsified by exhibiting a p-admissible metric measure space X and an Ahlfors–David codimension-θ regular set E with H_θ|_A non-atomic on a set of positive finite measure for which a bounded linear extension operator exists. For p ≠ 2, the proof reduces this to the known non-embedding of ℓ₂ into ℓ₁(ℓ_p), so the only plausible loophole is the trace identification; checking whether the companion paper's identification holds for a specific candidate space is the sharpest concrete test.

Watch

Extended reading notes

Core claim

Theorem 1.2 states that if X is p-admissible, E is Ahlfors–David codimension-θ regular, and there is a measurable A ⊂ E with 0 < H_θ(A) < ∞ such that H_θ|_A is nonatomic, then there is no bounded linear extension operator Ext: L_p(E,H_θ|_E) → B^{θ/p}_{p,1}(X). Combined with the companion paper's trace identification and nonlinear extension, this completes the linearity question: in the limiting case, the extension must be nonlinear. The non-atomicity condition is essential, as it ensures the trace space contains a copy of ℓ₂; finite-dimensional trace spaces can admit linear extensions.

Load-bearing premise

The proof takes as given the trace-space identification from the companion paper (that the trace of B^{θ/p}_{p,1}(X) to E equals L_p(E,H_θ|_E) with equivalent norms); if that identification fails, the theorem's statement loses its meaning.

Editorial extensions

If this is right

In Euclidean space, the result covers all Ahlfors–David regular subsets of codimension θ < n, recovering and extending the classical nonlinearity phenomenon of Burenkov–Gol'dman.
For any p-admissible metric measure space satisfying the hypotheses, the endpoint Besov trace admits no bounded linear right-inverse; nonlinearity is intrinsic to the limiting case.
For p ≠ 2, the obstruction is purely Banach-space-theoretic: the endpoint trace space contains a copy of ℓ₂ that cannot be embedded into the target's discrete model ℓ₁(ℓ_p).
For p = 2, the obstruction is quantitative: any hypothetical linear extension would produce finite-rank approximations to the identity on an infinite-dimensional L₂ space, a contradiction.
The nonlinear extension operator constructed in the companion paper is optimal in the sense that no bounded linear alternative exists.

Reading between the lines

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

If the trace-space identification from the companion paper were to fail for some metric measure space, the main theorem's conclusion would not follow; the result is conditional on that identification.
The finite-scale argument for p = 2 suggests that any hypothetical linear extension would have to be 'spread' across all scales, hinting at a deeper connection to the failure of bounded averaging operators at the endpoint.
A natural testable extension is to ask whether the same nonlinearity holds for Besov spaces with q > 1 or for Triebel–Lizorkin spaces in the limiting case; the present method relies on q = 1 via the embedding into ℓ₁(ℓ_p).
The role of the non-atomicity assumption is to guarantee an infinite-dimensional Hilbert subspace; it may be possible to weaken it to 'infinite-dimensional trace space' while preserving the conclusion.

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

Dear colleague,

You should know about this paper: it's a genuine new result — the nonexistence of bounded linear extension operators for limiting Besov traces on Ahlfors–David regular subsets of metric measure spaces — and the p=2 case is handled by an argument that is actually new. But the statement is only as solid as the author's companion preprint [6], which supplies the trace identification B^{θ/p}_{p,1}(X)|_E ≅ L_p(E,H_θ|_E) as a black box. If that identification fails, the whole theorem loses its object.

What the paper does well: Theorem 1.2 extends the Euclidean results of Burenkov–Gol'dman and Gol'dman to a much more general setting. The p≠2 route is the expected one — embed into ℓ1(ℓp) and use the fact that ℓ2 doesn't embed there — but the p=2 case is where the real work sits. The finite-scale tail estimate (Lemma 4.2) and the finite-rank contradiction in Section 4 are carefully argued. Theorem 2.7, the hyperbolic filling equivalence, is proved in full, with the local inhomogeneous setting handled properly. The paper is honest about its assumptions, and the nonatomicity condition is genuinely mild.

Soft spots, in proportion: the main one is the black-box trace identification from [6]. Both Section 3 and Section 4 use boundedness of the trace and Tr∘Ext=Id; the theorem only exists if that identification holds. The author does not restate or prove the necessary trace estimates here. That is not a flaw in the internal logic, but it is a load-bearing external dependency. Minor: Lemma 2.4 is imported from Tyulenev [19] with a citation, which is fine; the packing estimate is standard. Remark 3.5 gives informal but useful sufficiency conditions for nonatomicity.

Who is this for? Specialists in trace theory on metric spaces and Banach-space geometry. It deserves a serious referee; the right move is to have the referee check the companion preprint or ask the author to include the trace identification proof. I would accept it conditional on that verification.

Referee Report

2 major / 5 minor

Summary. The paper proves a nonexistence result for bounded linear extension operators in the limiting Besov trace problem on metric measure spaces. Let (X,d,µ) be p-admissible, E⊂X Ahlfors–David codimension-θ regular with θ∈(0,p), and suppose H_θ restricted to some measurable A⊂E is nonatomic with 0<H_θ(A)<∞. The main theorem (Theorem 1.2) asserts that there is no bounded linear Ext: L_p(E,H_θ|_E) → B^{θ/p}_{p,1}(X). For p≠2, the proof embeds B^{θ/p}_{p,1}(X) into ℓ_1(ℓ_p) via hyperbolic fillings and uses the non-embeddability of ℓ_2 into ℓ_1(ℓ_p). For p=2, the argument uses a weak-compactness tail estimate and a finite-scale reconstruction operator to force a finite-rank approximation of the identity on an infinite-dimensional L_2-space. The paper also proves an equivalence between the oscillation and hyperbolic-filling definitions of Besov spaces (Theorem 2.7).

Significance. If the quoted trace identification from the companion preprint [6] is accepted, the result is significant: it extends the classical Burenkov–Gol'dman nonlinearity phenomenon to general p-admissible metric measure spaces and gives a unified treatment of the cases p≠2 and p=2. The p=2 argument is notably subtle, since ℓ_2 does embed into ℓ_1(ℓ_2); the finite-scale contradiction is a genuine addition. The paper gives careful, largely self-contained proofs of the hyperbolic-filling equivalence and of the Banach-space obstruction Theorem 3.2, and it explicitly identifies the role of the nonatomicity assumption. The main weakness is the heavy reliance on the author's own companion preprint [6] for the identification of the trace space, which is the object of the main theorem.

major comments (2)

[§1.1, §3, §4] The statement and both proofs depend on the trace identification B^{θ/p}_{p,1}(X)|_E ≅ L_p(E,H_θ|_E) and on the boundedness of the trace operator, imported from [6, Thm 1.2, Cor 1.5] without proof or even a precise statement of the norm equivalence. The bounded-below estimate (3.24) and the identity R_A Tr Ext I_A = Id in §4 (around (4.30)–(4.31)) both require Tr to be bounded and surjective onto L_p(E,H_θ|_E). If the identification fails, the theorem has no object. Please either prove the needed trace theorem, state it as an explicit standing hypothesis, or restate it with the exact norm-equivalence constants.
[Lemma 4.2, Step 1 (after (4.15))] The proof asserts 'By the same argument, S_0u∈B^s_{p,1}(X)' without giving the argument. This assertion is needed for the convergence of the telescoping series and for the identity S(Pf)=f used to derive the tail estimate (4.9), which is the key estimate in Theorem 4.3. Please supply a proof that S_0u belongs to B, for example by estimating the L_p norm and the local Lipschitz constants of the partition S_0u via (2.13), using the level-0 discrete derivative.

minor comments (5)

[Abstract] The abstract states the trace-space identification as a fact without attribution; add a reference to [6] there or in a footnote.
[Lemma 2.4] The packing estimate is quoted from [19] without proof. A one-sentence indication of the proof would improve self-containedness.
[§4, (4.26)] There is a typo: 'where where' should be 'where'.
[Lemma 4.2, (4.14)] Inequality (2.13) is applied with δ=2^{-n}; for n=0 this gives δ=1, which is outside the stated range δ∈(0,1). The endpoint case should be treated separately.
[Theorem 4.3, (4.27)–(4.28)] The relation between the row-indexing in Π_N and the levels in J is not explicit; the shift between N and N+1 matters for the tail estimate. Please clarify the convention.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation: the main nonexistence proof is independent of its inputs, though it relies on the author's companion trace theorem [6] as a theorem dependency.

full rationale

The paper's central claim is that no bounded linear extension operator L_p(E,H_theta|_E) -> B^{theta/p}_{p,1}(X) exists. The proof by contradiction assumes such an Ext and uses Tr∘Ext = Id together with boundedness of the trace, the latter imported from [6, Corollary 1.5]. This is a normal theorem dependency, not a circular reduction: [6] identifies the trace space and provides a nonlinear extension, but it does not assert and is not used to assert the nonexistence of a linear extension. In the p≠2 case, the contradiction is a Banach-space obstruction (ℓ2 does not embed into ℓ1(ℓp)), with the non-embeddability cited to [16] and the rest of the argument proved in the paper. In the p=2 case, the contradiction comes from finite-rank operators F_N approximating the identity on an infinite-dimensional L2 space, using the weak-compactness/tail estimate of Lemma 4.1 and the hyperbolic-filling reconstruction Lemma 4.2; these are proved in the paper, modulo standard oscillation-to-discrete estimates cited from [6]. No equation in the proof is defined in terms of the target result, no fitted quantity is relabelled as a prediction, and the self-citations to [6] are to a parameter-free trace theorem with stated assumptions that do not include the target statement. The paper is not fully self-contained because the trace-space identification and some norm estimates are taken from the companion preprint, but that is a verification/dependency issue, not circularity.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The paper introduces no fitted constants or new postulated entities. The hyperbolic filling is a technical device, not an invented physical entity. The main external dependencies are the author's own prior trace theorem [6] and standard textbook results.

assumptions (7)

domain assumption Trace identification: B^{θ/p}_{p,1}(X)|_E ≅ L_p(E,H_θ|_E) with equivalent norms; trace operator Tr is bounded
Quoted from [6, Thm 1.2, Cor 1.5] and used as a black box in the statements and proofs of both Theorem 3.6 and Theorem 4.3. Not proved or independently verified in this paper.
domain assumption Packing estimate (Lemma 2.4)
Imported from [19, Prop 2.12]; bounds the degree of the hyperbolic filling graph and guarantees finiteness of vertices meeting a bounded set. No proof given in this paper.
domain assumption Besov norm equivalence (2.12) and Lipschitz estimate (2.13)
Taken from [6, Rem 2.14, 2.16]; used in the proof of Theorem 2.7 and Lemma 4.2 to pass between oscillation and discrete norms.
standard math ℓ_2 does not embed isomorphically into ℓ_p for p≠2
Classical result cited as [16, Prop 2.a.2]; used in Theorem 3.2 to show Π_N T cannot be bounded below on an infinite-dimensional subspace.
standard math Weak compactness criterion in Bochner L_1 spaces
Cited as [7, Ch. IV, Sec. 2, Thm 1]; used in Lemma 4.1 to conclude uniform integrability of U(K).
standard math Lebesgue differentiation theorem for uniformly locally doubling measures
Used in Theorem 2.7 and Lemma 4.2 to assert that a.e. point is a Lebesgue point and that ball averages converge.
standard math Khintchine's inequality
Used in Lemma 3.3 to identify the Rademacher span in L_p(A) with ℓ_2; standard.

how reviews work

0 comments

Cite this review

Pith. "Pith review of On the Linearity of Extension Operators for Traces of Besov Spaces on Metric Measure Spaces: The Limiting Case." pith.science (2026). https://pith.science/paper/4ABYCMNP

@misc{pith2026260711604,
  author       = {Pith},
  title        = {Pith review of: On the Linearity of Extension Operators for Traces of Besov Spaces on Metric Measure Spaces: The Limiting Case},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4ABYCMNP}},
  note         = {Machine review of arXiv:2607.11604}
}

abstract

Let $p\in[1,\infty)$, and let $X=(X,d,\mu)$ be a metric measure space such that $\mu$ is uniformly locally doubling and $X$ supports a local weak $(1,p)$-Poincar\'e inequality. Given $\theta\in(0,p)$ and an Ahlfors--David codimension-$\theta$ regular set $E\subset X$, the trace-space of the Besov space $B^{\theta/p}_{p,1}(X)$ to $E$ can be identified with $L_p(E,\mathcal{H}_{\theta}\lfloor_E)$. If there exists a measurable set $A\subset E$ with $0<\mathcal H_\theta(A)<\infty$ such that $\mathcal H_\theta\lfloor_A$ is nonatomic, we prove that there is no bounded linear extension operator $\operatorname{Ext}:L_p(E,\mathcal{H}_{\theta}\lfloor_E) \to B^{\theta/p}_{p,1}(X)$.

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 1 linked inside Pith

[6]

Chikalov

Aleksei Y. Chikalov. Traces of Besov spaces to regular subsets of metric measure spaces: the limiting case,
[19]

Tyulenev

Alexander I. Tyulenev. Traces of Sobolev spaces to irregular subsets of metric measure spaces.Sb. Math., 214(9):1241–1320, 2023. doi: 10.4213/sm9893e. Steklov Mathematical Institute of Russian Academy of Sciences Email address:chikalov.a@phystech.edu

doi:10.4213/sm9893e 2023
[1]

O. V. Besov. Investigation of a family of function spaces in connection with theorems of imbedding and extension.Translations of the American Mathematical Society. Series 2, 40:85–126, 1964. doi: 10.1090/trans2/040/03

doi:10.1090/trans2/040/03 1964
[2]

Sobolev spaces and hyperbolic fillings.Journal f¨ ur die Reine und Ange- wandte Mathematik, 737:161–187, 2018

Mario Bonk and Eero Saksman. Sobolev spaces and hyperbolic fillings.Journal f¨ ur die Reine und Ange- wandte Mathematik, 737:161–187, 2018. doi: 10.1515/crelle-2015-0036

doi:10.1515/crelle-2015-0036 2018
[3]

Triebel–Lizorkin spaces on metric spaces via hyperbolic fillings.Indiana Univ

Mario Bonk, Eero Saksman, and Tom´ as Soto. Triebel–Lizorkin spaces on metric spaces via hyperbolic fillings.Indiana Univ. Math. J., 67(4):1625–1663, 2018. doi: 10.1512/iumj.2018.67.7282

doi:10.1512/iumj.2018.67.7282 2018
[4]

V. I. Burenkov and M. L. Gol’dman. On the extension of functions fromL p.Trudy Matematicheskogo Instituta Imeni V. A. Steklova, 150:31–51, 1979

1979
[5]

On the mutually non isomorphicℓ p(ℓq) spaces.Math

Pilar Cembranos and Jos´ e Mendoza. On the mutually non isomorphicℓ p(ℓq) spaces.Math. Nachr., 284(16):2013–2023, 2011. doi: 10.1002/mana.201010056

doi:10.1002/mana.201010056 2013
[7]

American Mathematical Society, Providence, RI, 1977

Joseph Diestel and John Jerry Uhl, Jr.Vector Measures, volume 15 ofMathematical Surveys and Mono- graphs. American Mathematical Society, Providence, RI, 1977. AMS, Mathematical Surveys and Mono- graphs, Vol. 15

1977

Show all 19 references

[8]

Emilio Gagliardo. Caratterizzazioni delle tracce sulla frontiera relative ad alcune classi di funzioni inn variabili.Rendiconti del Seminario Matematico della Universit` a di Padova, 27:284–305, 1957

1957
[9]

Interpolation properties of Besov spaces defined on metric spaces.Math

Amiran Gogatishvili, Pekka Koskela, and Nageswari Shanmugalingam. Interpolation properties of Besov spaces defined on metric spaces.Math. Nachr., 283(2):215–231, 2010. doi: 10.1002/mana.200810242

2010 doi
[10]

Characterizations of Besov and Triebel–Lizorkin spaces on metric measure spaces.Forum Math., 25(4):787–819, 2013

Amiran Gogatishvili, Pekka Koskela, and Yuan Zhou. Characterizations of Besov and Triebel–Lizorkin spaces on metric measure spaces.Forum Math., 25(4):787–819, 2013. doi: 10.1515/form.2011.135

2013 doi
[11]

M. L. Gol’dman. On the extension of functions fromL p(Rm) to a space of higher dimension.Math. Notes, 25(4):266–270, 1979. Russian original: Mat. Zametki 25 (1979), no. 4, 513–520; doi: 10.1007/BF01688477

1979 doi
[12]

A theory of Besov and Triebel–Lizorkin spaces on metric measure spaces modeled on Carnot–Carath´ eodory spaces.Abstract and Applied Analysis, 2008:Article ID 893409, 250 pp., 2008

Yongsheng Han, Detlef M¨ uller, and Dachun Yang. A theory of Besov and Triebel–Lizorkin spaces on metric measure spaces modeled on Carnot–Carath´ eodory spaces.Abstract and Applied Analysis, 2008:Article ID 893409, 250 pp., 2008. doi: 10.1155/2008/893409

2008 doi
[13]

Tyson.Sobolev Spaces on Metric Measure Spaces: An Approach Based on Upper Gradients

Juha Heinonen, Pekka Koskela, Nageswari Shanmugalingam, and Jeremy T. Tyson.Sobolev Spaces on Metric Measure Spaces: An Approach Based on Upper Gradients. New Mathematical Monographs. Cam- bridge University Press, Cambridge, 2015. doi: 10.1017/CBO9781316135914

2015 doi
[14]

Trace operator on von Koch’s snowflake.Potential Anal., 61:659–684, 2024

Krystian Kazaniecki and Micha l Wojciechowski. Trace operator on von Koch’s snowflake.Potential Anal., 61:659–684, 2024. doi: 10.1007/s11118-024-10124-w

2024 doi
[15]

Pointwise characterizations of Besov and Triebel–Lizorkin spaces and quasiconformal mappings.Adv

Pekka Koskela, Dachun Yang, and Yuan Zhou. Pointwise characterizations of Besov and Triebel–Lizorkin spaces and quasiconformal mappings.Adv. Math., 226(4):3579–3621, 2011. doi: 10.1016/j.aim.2010.10.020. 20 ALEKSEI Y. CHIKALOV

2011 doi
[16]

Springer-Verlag, Berlin–New York, 1977

Joram Lindenstrauss and Lior Tzafriri.Classical Banach Spaces I: Sequence Spaces, volume 92 ofErgeb- nisse der Mathematik und ihrer Grenzgebiete. Springer-Verlag, Berlin–New York, 1977

1977
[17]

A counter-example connected with Gagliardo’s trace theorem.Commentationes Mathemati- cae

Jaak Peetre. A counter-example connected with Gagliardo’s trace theorem.Commentationes Mathemati- cae. Special Issue, pages 277–282, 1979

1979
[18]

Besov spaces via hyperbolic fillings, 2016

Tom´ as Soto. Besov spaces via hyperbolic fillings, 2016. arXiv:1606.08082

2016 arXiv

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.

[1] [6]

Chikalov

Aleksei Y. Chikalov. Traces of Besov spaces to regular subsets of metric measure spaces: the limiting case,

[2] [19]

Tyulenev

Alexander I. Tyulenev. Traces of Sobolev spaces to irregular subsets of metric measure spaces.Sb. Math., 214(9):1241–1320, 2023. doi: 10.4213/sm9893e. Steklov Mathematical Institute of Russian Academy of Sciences Email address:chikalov.a@phystech.edu

doi:10.4213/sm9893e 2023

[3] [1]

O. V. Besov. Investigation of a family of function spaces in connection with theorems of imbedding and extension.Translations of the American Mathematical Society. Series 2, 40:85–126, 1964. doi: 10.1090/trans2/040/03

doi:10.1090/trans2/040/03 1964

[4] [2]

Sobolev spaces and hyperbolic fillings.Journal f¨ ur die Reine und Ange- wandte Mathematik, 737:161–187, 2018

Mario Bonk and Eero Saksman. Sobolev spaces and hyperbolic fillings.Journal f¨ ur die Reine und Ange- wandte Mathematik, 737:161–187, 2018. doi: 10.1515/crelle-2015-0036

doi:10.1515/crelle-2015-0036 2018

[5] [3]

Triebel–Lizorkin spaces on metric spaces via hyperbolic fillings.Indiana Univ

Mario Bonk, Eero Saksman, and Tom´ as Soto. Triebel–Lizorkin spaces on metric spaces via hyperbolic fillings.Indiana Univ. Math. J., 67(4):1625–1663, 2018. doi: 10.1512/iumj.2018.67.7282

doi:10.1512/iumj.2018.67.7282 2018

[6] [4]

V. I. Burenkov and M. L. Gol’dman. On the extension of functions fromL p.Trudy Matematicheskogo Instituta Imeni V. A. Steklova, 150:31–51, 1979

1979

[7] [5]

On the mutually non isomorphicℓ p(ℓq) spaces.Math

Pilar Cembranos and Jos´ e Mendoza. On the mutually non isomorphicℓ p(ℓq) spaces.Math. Nachr., 284(16):2013–2023, 2011. doi: 10.1002/mana.201010056

doi:10.1002/mana.201010056 2013

[8] [7]

American Mathematical Society, Providence, RI, 1977

Joseph Diestel and John Jerry Uhl, Jr.Vector Measures, volume 15 ofMathematical Surveys and Mono- graphs. American Mathematical Society, Providence, RI, 1977. AMS, Mathematical Surveys and Mono- graphs, Vol. 15

1977