ScalingStacks

[0MK2]

Proof. Both statements follow by induction. First note that Θn\Theta_{n} itself is closed under retracts [9, Proposition 3.14]. In the base case, the category Θ1=Δ\Theta_{1}=\Delta consists of precisely the grids, and the sets of morphisms agree SegalΘ1=SegalΔ\mathrm{Segal}_{\Theta_{1}}=\mathrm{Segal}_{\Delta}. Now assume, by induction, that every object of o∈Θn−1o\in\Theta_{n-1} is the retract of a grid δn​(𝐦o)\delta_{n}(\mathbf{m}^{o}), for some object 𝐦o=[m1o]×⋯×[mn−1o]∈Δ×n\mathbf{m}^{o}=[m_{1}^{o}]\times\cdots\times[m_{n-1}^{o}]\in\Delta^{\times n}. In fact, given any finite collection of objects {oi∈Θn−1}\{o_{i}\in\Theta_{n-1}\} they may be obtained as the retract of a single grid. This grid may be obtained as the image of 𝐤=[k1]×⋯×[kn−1]\mathbf{k}=[k_{1}]\times\cdots\times[k_{n-1}], where kjk_{j} is the maximum of the collection {mjoi}\{m_{j}^{o_{i}}\}. It now follows easily that the object ([n],o1,…,oi)∈Θn([n];o_{1},\dots,o_{i})\in\Theta_{n} is a retract of the grid coming from the object [n]×𝐤[n]\times\mathbf{k}.

To prove the second statement we note that there are two types of maps in SegalΘn\mathrm{Segal}_{\Theta_{n}}, those in σ!(SegalΘn−1)\sigma_{!}(\mathrm{Segal}_{\Theta_{n-1}}) and the maps

j({0,…,k};o1,…,ok)∪j⁡({k})j({k,…,m};ok+1,…,om)→j([m];o1,…,om)j({\{0,\dots,k\}};o_{1},\dots,o_{k})\cup^{j({\{k\}})}j({\{k,\dots,m\}};o_{k+1},\dots,o_{m})\to j({[m]};o_{1},\dots,o_{m})

for 0≤k≤m0\leq k\leq m and oi∈Θn−1o_{i}\in\Theta_{n-1}. This later map is a retract of the image under (δn)!(\delta_{n})_{!} of the map

(j{0,1,…,k}∪j​{k}j{k,…,m}→j[m])⊠j(𝐦)\left(j{\{0,1,\dots,k\}}\cup^{j{\{k\}}}j{\{k,\dots,m\}}\to j[m]\right)\boxtimes j(\mathbf{m})

which is a map in SegalΔ×n\mathrm{Segal}_{\Delta^{\times n}}. Here 𝐦\mathbf{m} is such that ({0,…,k},o1,…,ok)({\{0,\dots,k\}};o_{1},\dots,o_{k}) is the retract of the grid corresponding to [k]×𝐦[k]\times\mathbf{m}.

The former class of morphisms in SegalΘn\mathrm{Segal}_{\Theta_{n}}, those in σ!(SegalΘn−1)\sigma_{!}(\mathrm{Segal}_{\Theta_{n-1}}), are also retracts on elements in SegalΔ×n\mathrm{Segal}_{\Delta^{\times n}}. Specifically, if σ!(f)∈σ!(SegalΘn−1)\sigma_{!}(f)\in\sigma_{!}(\mathrm{Segal}_{\Theta_{n-1}}), then by induction ff is the retract of (δn−1)!(g)(\delta_{n-1})_{!}(g) for some g∈SegalΔ×n−1g\in\mathrm{Segal}_{\Delta^{\times n-1}}. One may then readily check that σ⁡(f)\sigma(f) is the retract of j⁡[1]⊠g∈SegalΔ×nj[1]\boxtimes g\in\mathrm{Segal}_{\Delta^{\times n}}. ∎

Original mathematics by the credited authors. Source collection and HTML conversion remain in progress.

Clark Barwick, Christopher Schommer-Pries

Original source: arXiv:1112.0040v6

    Original source · 1112.0040v6