Lemma 20.6.4. The functor \(\underset{[m] : \Delta\catop}{\colim}\, \underset{[n] : \Delta\catop}{\colim}\, w_m\Scons_n(-)^\simeq\colon\WaldCat\to\An_{\bullet}\) is additive.

Proof

We observe that \[ w_m(\ExactSeq(A))=\ExactSeq(w_m(A)) \] functorially in \([m]\) and \(A\). Therefore, it follows from Lemma 20.6.3, it suffices to prove that the functor \[ \Phi\colon A\mapsto \underset{[m] : \Delta\catop}{\colim}\, w_m(A)^\simeq \] sends the functor \[ i\colon A\to w_1(A)\ , \quad x\mapsto \id_x \] to an equivalence. Let us consider the functor \[ \ev_0\colon w_1(A)\to A\, . \] It suffices to check that \(i\circ\ev_0\) becomes simplicially homotopic to the identity after applying the functor \[ w(-)^\simeq \colon\WaldCat\to\PSh(\Delta)\, . \] The map \(\min\colon [1]\times [1]\to [1]\) induces a functor \[ h\colon[1]\times\Fun([1],A)\to\Fun([1],A) \] corresponding by transposition to the functor \[ \ev^*_0\colon\Fun([1],A)\to\Fun([1]\times [1],A)\, . \] Taking the Rezk nerve yields a map \[ N(h)\colon\Delta^1\times N(\Fun([1],A))\to N(\Fun([1],A))\, . \] We observe that there is a canonical embedding \[ w(A)^\simeq\hookrightarrow N(\Fun([1],A)) \] so that the map \(N(h)\) restricts to a map \[ H\colon \Delta^1\times w(A)^\simeq\to w(A)^\simeq \] that is the homotopy we were looking for.

From Section 20.6, The additivity theorem.