Hinich's model for Day convolution revisited

Christoph Winges

Hinich's model for Day convolution revisited

T0 review · 0 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Hinich's pullback model of Day convolution is the exponential of ∞-operads over a fixed base operad, with algebras described as a bivariant functor.

desk verdict A careful, honest proof note that gives a clean direct argument for Hinich's Day convolution universal property; the two flagged equivalences are indeed formal, so the paper is solid if low in novelty. read the letter →

arxiv 2506.06025 v1 pith:EE6ZM64D submitted 2025-06-06 math.CT math.AT

Christoph Winges This is my paper

classification math.CTmath.AT MSC 18N7018N60

keywords Dayconvolutionoplaxarrowcategoryorthofibration∞-operadsO-monoidal∞-categoriesbivariantfunctoralgebrasinexponential

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 note establishes that Hinich's pullback construction of Day convolution is the right universal object: for every map of $\infty$-operads $A^{\otimes} \to \mathcal{O}^{\otimes}$ and every pair of $\mathcal{O}$-monoidal $\infty$-categories $C$ and $D$, algebras for $A$ in the Day convolution of $C$ and $D$ are naturally equivalent to algebras for $A \times_{\mathcal{O}} C$ in $D$. In particular, Day convolution is an exponential in the $\infty$-category of $\infty$-operads over $\mathcal{O}$, meaning it is a right adjoint of the product with the first variable, and formation of algebras in it is a bivariant functor. The proof verifies the universal property directly rather than comparing models, by straightening categories of algebras over the oplax arrow category to the functor $(X,Y) \mapsto \mathrm{Alg}_{X/\mathcal{O}}(Y)$. The upshot is that Hinich's model is not merely one construction among many: it is the universal object that algebras over any $A \to \mathcal{O}$ see.

What carries the argument

The load-bearing device is the orthofibration, a functor $X \to Y \times Z$ that is simultaneously a cartesian fibration over $Y$ and a cocartesian fibration over $Z$, with cartesian lifts projecting to equivalences in $Z$ and cocartesian lifts to equivalences in $Y$. Orthofibrations straighten to functors $Y^{\mathrm{op}} \times Z \to \mathrm{Cat}$, and the paper applies this to the source and target maps $(s,t): \mathrm{Ar}^{\mathrm{opl}} \to \mathrm{Cat} \times \mathrm{Cat}$ on the oplax arrow category. Theorem 2.7 identifies the straightening of the induced functor on categories of algebras as $(X,Y) \mapsto \mathrm{Alg}_{X/\mathcal{O}}(Y)$. Free cartesian and cocartesian fibrations provide the adjunctions used to pin down that straightening.

What would settle it

Check the two unproved equivalences in the displayed chain of the proof of Theorem 2.1 for a concrete operad map, such as the map from the associative operad to the commutative operad: if $A \times_{\mathcal{O}} \mathrm{Day}_{C,D}$ is not equivalent to $\mathrm{Day}_{A \times_{\mathcal{O}} C, A \times_{\mathcal{O}} D}$, or if $\mathrm{Alg}_{A/\mathcal{O}}(P)$ is not equivalent to $\mathrm{Alg}_A(A \times_{\mathcal{O}} P)$, the claimed universal property does not follow from Theorem 2.7.

Watch

Extended reading notes

Core claim

The central claim is Theorem 2.1: for any operad map $\alpha: A^{\otimes} \to \mathcal{O}^{\otimes}$ and $\mathcal{O}$-monoidal $\infty$-categories $C$ and $D$, there is a natural equivalence $\mathrm{Alg}_{A/\mathcal{O}}(\mathrm{Day}_{C,D}) \simeq \mathrm{Alg}_{A \times_{\mathcal{O}} C/\mathcal{O}}(D)$. The note's main step is Theorem 2.7, which proves that the functor $(s_*, t_*) : \mathrm{Alg}_{\mathcal{O}}(\mathrm{Ar}^{\mathrm{opl}}) \to \mathrm{Alg}_{\mathcal{O}}(\mathrm{Cat}) \times \mathrm{Alg}_{\mathcal{O}}(\mathrm{Cat})$ is an orthofibration straightening to the functor $(X, Y) \mapsto \mathrm{Alg}_{X/\mathcal{O}}(Y)$. From this straightening the universal property of Day convolution is derived for an arbitrary source operad $A$, and applying it with $A = \mathrm{Day}_{C,D}$ gives the evaluation map witnessing the exponential. Section 4 records the cocompleteness assumptions under which $\mathrm{Day}^{\otimes}_{C,D} \to \mathcal{O}^{\otimes}$ is again an $\mathcal{O}$-monoidal $\infty$-category.

Load-bearing premise

The proof of Theorem 2.1 relies on the assertion, not separately justified, that Day convolution and algebra formation both commute with pullback along an operad map; the universal property for arbitrary $A$ follows only if those two equivalences hold.

Editorial extensions

If this is right

If the main theorem is correct, Hinich's pullback $\mathrm{Day}^{\otimes}_{C,D}$ is literally the exponential of $\mathcal{O}$-monoidal $\infty$-categories in the $\infty$-category of $\infty$-operads over $\mathcal{O}$, so no comparison with a different construction is needed for that universal property.
The category of algebras in the Day convolution operad is a bivariant functor: contravariant in $C$ and covariant in $D$, with functoriality induced by pre- and postcomposition with operad maps.
Specialising Theorem 2.7 to the trivial base operad recovers the fact that the oplax arrow category straightens to the functor category, and specialising it to a single variable recovers the cotensoring formula for algebra categories over $\mathcal{O}$.
Under the colimit assumptions of Section 4, $\mathrm{Day}^{\otimes}_{C,D}$ is an $\mathcal{O}$-monoidal $\infty$-category, so the exponential description comes with an explicit $\mathcal{O}$-monoidal structure.
The Section 3 variations give analogous algebra equivalences for suboperads of $\mathrm{Cat}^{\times}$, such as stable categories and exact functors.

Reading between the lines

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

An immediate editorial extension is that the two pullback-compatibility equivalences used inside the proof of Theorem 2.1 could be stated and proved as independent lemmas; this would make the derivation of the universal property self-contained rather than relying on the displayed chain.
The orthofibration straightening method should transfer to other non-oplax arrow categories and to two-variable fibrations, producing Day-convolution-type exponentials in settings such as spans or correspondences.
In concrete examples where $\mathrm{Alg}_{A \times_{\mathcal{O}} C/\mathcal{O}}(D)$ is easier to compute than $\mathrm{Alg}_{A/\mathcal{O}}(\mathrm{Day}_{C,D})$, the bivariant description gives a practical formula for algebras in the Day convolution.

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.

Desk Editor's Note

read the letter

The main thing to know: this is a proof note, not a new theorem. It proves Theorem 2.1 (the universal property for Hinich's Day convolution) by a direct orthofibration/straightening argument, and it says up front that this is implicit in CDH+23. That honesty is typical of the whole paper.

What is actually new is Theorem 2.7, which identifies the functor (s_*, t_*) on O-algebras in the oplax arrow category as an orthofibration straightening to (X,Y) -> Alg_{X/O}(Y). This bivariant statement is a genuinely useful addition and is proved in detail using standard machinery from GHN17 and HHLN23. I also find the variation in Section 3, handling suboperads like stable categories, a nice bonus, and Section 4's reproof of the cocompleteness criterion is clearly written even though it is known.

The reader flagged two unproved equivalences in the proof of Theorem 2.1: pullback compatibility of Day convolution and the identification Alg_{A/O}(P) ≃ Alg_A(A ×_O P). I checked both against the definitions. The first is pullback pasting; the second is the universal property of the pullback. They are formal and not load-bearing in any problematic sense, though the paper could have said a word. That is a minor presentational issue, not a real weakness.

The bigger soft spot is that the proof of Theorem 2.7 is heavy: it leans on the HHLN23 machinery for two-variable fibrations and the reader has to trust a fair amount of externa. But the chain is careful, I saw no circularity, and the citations are appropriate. The novelty is modest because the main theorem was already known, but the paper delivers a cleaner universal-property proof and a nice bivariant description.

Who gets value from this? Practitioners of ∞-category theory who want a concrete, provably correct description of Day convolution and algebra formation over O, especially if they plan to work with Day convolution in suboperads. It is not a breakthrough, but it is a useful reference.

Recommendation: this deserves a serious referee. It should not be desk rejected. I would send it to review expecting a minor revision, mainly asking the author to point out the two formal equivalences in the proof of Theorem 2.1 and to add a short remark on why they are immediate.

Referee Report

0 major / 5 minor

Summary. The paper gives an alternative proof of Hinich's theorem that the Day convolution operad of two O-monoidal infinity-categories is the exponential in the infinity-category of infinity-operads over O. The main result, Theorem 2.1, establishes a natural equivalence Alg_{A×_O C/O}(D) ≃ Alg_{A/O}(Day_{C,D}) for any operad map A⊗→O⊗. The proof is based on a new statement, Theorem 2.7, which identifies the straightening of the orthofibration (s_*,t_*): Alg_O(Ar_op^l) → Alg_O(Cat)×Alg_O(Cat) with the functor (X,Y) ↦ Alg_{X/O}(Y). The paper also derives consequences for suboperads of Cat^× and gives a proof that Day_{C,D} is an O-monoidal category under suitable cocompleteness assumptions.

Significance. If correct, the paper provides a clean and direct verification of the universal property of Hinich's Day convolution model, avoiding a comparison with Lurie's more general construction. The identification of algebra formation as a bivariant functor is a useful structural result. The arguments are detailed and rely on established machinery (Lurie, GHN17, HHLN23), with no apparent circularity. The note is written carefully and should be of interest to researchers working on infinity-operads and Day convolution.

minor comments (5)

[Section 2, proof of Theorem 2.1] The equivalences Alg_{A/O}(P) ≃ Alg_A(A×_O P) and A×_O Day_{C,D} ≃ Day_{A×_O C, A×_O D} are used without comment. They are direct consequences of pullback pasting and the universal property of pullback, but a sentence of justification would improve readability and address a potential concern for the reader.
[Section 2, proof of Theorem 2.7] The final step of the proof, ending with 'which is precisely what we needed to show', is very compressed. Since the equivalence obtained is for the composite with the core functor, it would be helpful to spell out the Yoneda argument that identifies the Cat-valued straightening of (s_*,t_*) with the functor (X,Y) ↦ Alg_{X/O}(Y).
[Section 3] The construction of the symmetric monoidal structure U^⊗ from the two assumptions on U is stated without proof. As this section is a variation on the main theme, a reference or a brief indication of the construction would help the reader.
[Introduction, conventions] The notation Alg_{A/O}(P) is used without an explicit definition; a preliminary definition in the conventions section would help readers, especially since the notation is central to Theorem 2.1.
[References and typos] There are a few minor typos: the URL for [Lur] appears twice in the reference list, and in Section 4 the notation Hom^φ_{Day⊗_{C,D}}(F,G) is introduced without a formal definition of the fiber it denotes.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Theorem 2.1 is derived from the independently proved Theorem 2.7, whose proof uses external orthofibration machinery and does not presuppose the Day convolution universal property.

full rationale

The central equivalence in Theorem 2.1 is proved by a chain whose two unstated identifications are formal: Alg_{A/O}(P) ≃ Alg_A(A ×_O P) is the pullback/base-change adjunction for algebras over operads, and A ×_O Day_{C,D} ≃ Day_{A×_O C, A×_O D} is pullback pasting applied to the defining pullback of Theorem 1.1. The substantive input is Theorem 2.7, whose proof in Section 2 establishes the straightening of (s_*, t_*) by independent orthofibration machinery (Proposition 2.14, Lemma 2.17, Proposition 2.18, Lemma 2.26) and does not use Theorem 2.1 or Hinich's theorem. All cited results are external ([Lur], [GHN17], [HHLN23a,b], [CDH+23]); there are no self-citations by the author. Section 4 reuses Theorem 2.7 to compute mapping anima, but that is an application rather than an input. No prediction is fitted, no uniqueness is imported from the authors' prior work, and no ansatz is smuggled in by citation. The argument is self-contained modulo standard, independently published results.

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

The paper introduces no new mathematical entities and has no free parameters. It depends on established theorems in ∞-category theory, mainly from Lurie's Higher Algebra, GHN17, and HHLN23, all cited explicitly.

assumptions (4)

standard math Unstraightening equivalences for curved orthofibrations: Fun(Y^op, Cocartlax(Z))^cart ≃ CurvOrtho(Y,Z) ≃ Fun(Z, Cartopl(Y))^cocart
Imported from [HHLN23a, Corollary 2.3.4]; used throughout Section 2 to model and straighten two-variable fibrations.
standard math Existence of free cartesian fibrations (left adjoint of Cart(I) → Cat/I)
From [GHN17, Theorem 4.5]; used to construct free orthofibrations in Corollary 2.19 and to prove Lemma 2.21.
standard math The functor (s,t) : Arop^l → Cat×Cat is an orthofibration
Proved in Proposition 2.14; this is the seed from which the algebra-level statement is bootstrapped.
standard math Characterization of cocartesian fibrations of operads via O-monoids
Used implicitly in Lemma 2.26 and Section 4, following Lurie's Higher Algebra.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Hinich's model for Day convolution revisited." pith.science (2026). https://pith.science/paper/EE6ZM64D

@misc{pith2026250606025,
  author       = {Pith},
  title        = {Pith review of: Hinich's model for Day convolution revisited},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EE6ZM64D}},
  note         = {Machine review of arXiv:2506.06025}
}

abstract

We prove that Hinich's construction of the Day convolution operad of two $\mathcal{O}$-monoidal $\infty$-categories is an exponential in the $\infty$-category of $\infty$-operads over $\mathcal{O}$, and use this to give an explicit description of the formation of algebras in the Day convolution operad as a bivariant functor.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

10 extracted references · 10 canonical work pages

[1]

Calm \`e s, E

B. Calm \`e s, E. Dotto, Y. Harpaz, F. Hebestreit, M. Land, K. Moi, D. Nardin, T. Nikolaus, and W. Steimle. Hermitian K -theory for stable  -categories. I : Foundations . Sel. Math., New Ser. , 29(1):269, 2023. Id/No 10

work page 2023
[2]

Gepner, R

D. Gepner, R. Haugseng, and T. Nikolaus. Lax colimits and free fibrations in  -categories. Doc. Math. , 22:1225--1266, 2017

work page 2017
[3]

S. Glasman. Day convolution for  -categories. Math. Res. Lett. , 23(5):1369--1385, 2016

work page 2016
[4]

Haugseng, F

R. Haugseng, F. Hebestreit, S. Linskens, and J. Nuiten. Lax monoidal adjunctions, two-variable fibrations and the calculus of mates. Proc. Lond. Math. Soc. (3) , 127(4):889--957, 2023

work page 2023
[5]

Haugseng, F

R. Haugseng, F. Hebestreit, S. Linskens, and J. Nuiten. Two-variable fibrations, factorisation systems and  -categories of spans. Forum Math. Sigma , 11:70, 2023. Id/No e111

work page 2023
[6]

V. Hinich. Rectification of algebras and modules. Doc. Math. , 20:879--926, 2015

work page 2015
[7]

V. Hinich. Yoneda lemma for enriched  -categories. Adv. Math. , 367:119, 2020. Id/No 107129

work page 2020
[8]

Haugseng, V

R. Haugseng, V. Melani, and P. Safronov. Shifted coisotropic correspondences. J. Inst. Math. Jussieu , 21(3):785--849, 2022

work page 2022

Show all 10 references

[9]

J. Lurie. Higher A lgebra. Available on the author's homepage: https://www.math.ias.edu/ lurie/papers/HA.pdf https://www.math.ias.edu/ lurie/papers/HA.pdf
[10]

J. Lurie. Higher topos theory , volume 170. Princeton, NJ: Princeton University Press, 2009

2009

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Calm \`e s, E

B. Calm \`e s, E. Dotto, Y. Harpaz, F. Hebestreit, M. Land, K. Moi, D. Nardin, T. Nikolaus, and W. Steimle. Hermitian K -theory for stable \( \) -categories. I : Foundations . Sel. Math., New Ser. , 29(1):269, 2023. Id/No 10

work page 2023

[2] [2]

Gepner, R

D. Gepner, R. Haugseng, and T. Nikolaus. Lax colimits and free fibrations in \( \) -categories. Doc. Math. , 22:1225--1266, 2017

work page 2017

[3] [3]

S. Glasman. Day convolution for \( \) -categories. Math. Res. Lett. , 23(5):1369--1385, 2016

work page 2016

[4] [4]

Haugseng, F

R. Haugseng, F. Hebestreit, S. Linskens, and J. Nuiten. Lax monoidal adjunctions, two-variable fibrations and the calculus of mates. Proc. Lond. Math. Soc. (3) , 127(4):889--957, 2023

work page 2023

[5] [5]

Haugseng, F

R. Haugseng, F. Hebestreit, S. Linskens, and J. Nuiten. Two-variable fibrations, factorisation systems and \( \) -categories of spans. Forum Math. Sigma , 11:70, 2023. Id/No e111

work page 2023

[6] [6]

V. Hinich. Rectification of algebras and modules. Doc. Math. , 20:879--926, 2015

work page 2015

[7] [7]

V. Hinich. Yoneda lemma for enriched \( \) -categories. Adv. Math. , 367:119, 2020. Id/No 107129

work page 2020

[8] [8]

Haugseng, V

R. Haugseng, V. Melani, and P. Safronov. Shifted coisotropic correspondences. J. Inst. Math. Jussieu , 21(3):785--849, 2022

work page 2022