ScalingStacks

10. Segal space model category structure[0MTN]

In this section we prove (7.1).

The Segal space closed model category structure on s​𝒮s{\operatorname{\mathcal{S}}} is defined using (9.1) to be the localization of simplicial spaces with respect to the map φ=∐i≥0φi\varphi=\coprod_{i\geq 0}\varphi^{i}, where φn:G⁡(n)→F⁡(n)\varphi^{n}\colon G(n)\rightarrow F(n) is the map defined in (4.1). Parts (1)-(4) of (7.1) follow immediately from (9.1). The only thing left to prove is the compatibility of this model category structure with the cartesian closure.

To prove this, we need the notion of a cover of F⁡(n)F(n). Let αi:[k]→[n]\alpha^{i}\colon[k]\rightarrow[n] for i=0,…,n−ki=0,\dots,n-k denote the maps defined by αi​(j)=i+j\alpha^{i}(j)=i+j; we also write αi:F⁡(k)→F⁡(n)\alpha^{i}\colon F(k)\rightarrow F(n) for the corresponding map of simplicial spaces. We say that a subobject G⊆F⁡(n)G\subseteq F(n) is a cover of F⁡(n)F(n) if

  1. (1)

    GG and F⁡(n)F(n) have the same 00-space, i.e., G0=F​(n)0G_{0}=F(n)_{0}, and

  2. (2)

    GG has the form

    G=⋃λαiλ​F​(kλ)G=\bigcup_{\lambda}\alpha^{i_{\lambda}}F(k_{\lambda})

    where kλ≥1k_{\lambda}\geq 1 and iλ=0,…,kλ−1i_{\lambda}=0,\dots,k_{\lambda}-1.

In particular, F⁡(n)F(n) covers itself, and G⁡(n)⊂F⁡(n)G(n)\subset F(n) is the smallest cover of F⁡(n)F(n).

[05VH]

Lemma 10.1. Let G⊂F⁡(n)G\subset F(n) be a cover. Then the inclusion maps G⁡(n)→𝑖G→𝑗F⁡(n)G(n)\xrightarrow{i}G\xrightarrow{j}F(n) are weak equivalences in the Segal space model category structure.

[05VI]

Proof. In this proof, weak equivalence will mean weak equivalence in the Segal space model category structure. The composite map j​iji is a weak equivalence by construction, so it suffices to show that ii is also a weak equivalence. Given any αi1​F​(k1),αi2​F​(k2)⊂F⁡(n)\alpha^{i_{1}}F(k_{1}),\alpha^{i_{2}}F(k_{2})\subset F(n), we see that the intersection αi1​F​(k1)∩αi2​F​(k2)\alpha^{i_{1}}F(k_{1})\cap\alpha^{i_{2}}F(k_{2}) is either empty, or is equal to αi3​F​(k3)\alpha^{i_{3}}F(k_{3}) for some i3i_{3} and k3k_{3}. Thus GG can be written as a colimit over a partially ordered set of subcomplexes of the form αi​F​(k)\alpha^{i}F(k). Since G⁡(n)∩αi​F​(k)=αi​G​(k)G(n)\cap\alpha^{i}F(k)=\alpha^{i}G(k), we see that G⁡(n)G(n) is obtained as a colimit over the same indexing category of subobjects of the form αi​G​(k)\alpha^{i}G(k). Since by hypothesis the map Maps​𝒮⁡(αi​F​(k),W)→Maps​𝒮⁡(αi​G​(k),W)\Map_{s{\operatorname{\mathcal{S}}}}(\alpha^{i}F(k),W)\rightarrow\Map_{s{\operatorname{\mathcal{S}}}}(\alpha^{i}G(k),W) is a weak equivalence for any Segal space WW, we conclude that Maps​𝒮⁡(G,W)→Maps​𝒮⁡(G⁡(n),W)\Map_{s{\operatorname{\mathcal{S}}}}(G,W)\rightarrow\Map_{s{\operatorname{\mathcal{S}}}}(G(n),W) is also a weak equivalence for any Segal space WW, and hence ii is a weak equivalence in the Segal space model category, as desired. ∎

[05VJ]

Remark 10.2. The class of subobjects which are weakly equivalent to F⁡(n)F(n) is not exhausted by the coverings. For example, one can show that for 0<i<n0<i<n the subobject F˙​(n)∖di​F​(n−1)\dot{F}(n)\setminus d^{i}F(n-1) (the “boundary” of F⁡(n)F(n) with a “face” removed which is neither the first nor the last face) is weakly equivalent to F⁡(n)F(n) in the Segal space model category structure, but is not a cover.

To finish the proof of (7.1), we note that by (9.2) it suffices to show that for a Segal space WW, the simplicial space WF⁡(1)W^{F(1)} is also a Segal space; i.e., that the induced maps φk:(WF⁡(1))k≈Maps​𝒮⁡(F⁡(k),WF⁡(1))→Maps​𝒮⁡(G⁡(k),WF⁡(1))\varphi_{k}\colon(W^{F(1)})_{k}\approx\Map_{s{\operatorname{\mathcal{S}}}}(F(k),W^{F(1)})\rightarrow\Map_{s{\operatorname{\mathcal{S}}}}(G(k),W^{F(1)}) are weak equivalences. This follows immediately from (10.3) below.

[05VK]

Lemma 10.3. The inclusion F⁡(1)×G⁡(n)→F⁡(1)×F⁡(n)F(1)\times G(n)\rightarrow F(1)\times F(n) is a weak equivalence in the Segal space model category structure.

[05VL]

Proof. Let γi:[n+1]→[1]×[n]\gamma^{i}\colon[n+1]\rightarrow[1]\times[n] denote the map defined by

γi​(j)={(0,j)if j≤i,(1,j−1)if j>i.\gamma^{i}(j)=\begin{cases}(0,j)&\text{if $j\leq i$,}\\ (1,j-1)&\text{if $j>i$.}\end{cases}

Likewise, let δi:[n]→[1]×[n]\delta^{i}\colon[n]\rightarrow[1]\times[n] denote the map defined by

δi​(j)={(0,j)if j≤i,(1,j)if j>i.\delta^{i}(j)=\begin{cases}(0,j)&\text{if $j\leq i$,}\\ (1,j)&\text{if $j>i$.}\end{cases}

Then one can write F⁡(1)×F⁡(n)F(1)\times F(n) as a colimit of the diagram

γ0​F​(n+1)←δ0​F​(n)→γ1​F​(n+1)←δ1​F​(n)→…→γn​F​(n+1)\gamma^{0}F(n+1)\leftarrow\delta^{0}F(n)\rightarrow\gamma^{1}F(n+1)\leftarrow\delta^{1}F(n)\rightarrow\dots\rightarrow\gamma^{n}F(n+1) (10.4)

of subobjects. (This is analogous to the decomposition of the simplicial set Δ⁡[1]×Δ⁡[n]\Delta[1]\times\Delta[n] into a union of (n+1)(n+1) copies of Δ⁡[n+1]\Delta[n+1], attached along faces.) A straightforward computation shows that the maps γi​F​(n+1)∩(F⁡(1)×G⁡(n))→γi​F​(n+1)\gamma^{i}F(n+1)\cap(F(1)\times G(n))\rightarrow\gamma^{i}F(n+1) and δi​F​(n)∩(F⁡(1)×G⁡(n))→δi​F​(n)\delta^{i}F(n)\cap(F(1)\times G(n))\rightarrow\delta^{i}F(n) are covers, and hence by (10.1) are weak equivalences. Thus the result follows by comparing diagram (10.4) with the diagram obtained by intersecting each object of (10.4) with F⁡(1)×G⁡(n)F(1)\times G(n). ∎

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

Charles Rezk

Original source: arXiv:math/9811037v3

Original source · math/9811037v3