4
$\begingroup$

I am reading the paper "Revisiting the de Rham-Witt complex" by Bhatt-Lurie-Mathew and I am a bit confused about the difference between $W\Omega_R^*$ and $\hat{\Omega}^*_{W(R)}$.

Let $\mathrm{CAlg}_{\mathrm{\mathbb{F}_p}}^{\mathrm{reg}}$ be the category of regular noetherian $\mathbb{F}_p$-algebras. In my understanding, the motivation to define the de Rham-Witt complex is to construct a functor $$ \mathrm{CAlg}_{\mathrm{\mathbb{F}_p}}^{\mathrm{reg}}\to \mathrm{Ch}(\mathrm{Ab}) $$ which can be described explicitly and computes the crystalline cohomology. To construct such a functor, they are introducing the notion of "strict Dieudonne complex" and defining $W\Omega_R^*$ to be the "strictification" of $\hat{\Omega}^*_{W(R)}$. However, it seems to me that the functor $\hat{\Omega}^*_{W(-)}$ already has the required property. Indeed, their proof of the comparison theorem (between de Rham-Witt and crystalline cohomology) relies only on the following theorem:

Theorem 10.1.2. Let $A$ be a p-complete commutative algebra object in $D(\mathrm{Fun}(\mathrm{CAlg}_{\mathrm{\mathbb{F}_p}}^{\mathrm{reg}},\mathrm{Ab}))$ equipped with an isomorphism $u_0\colon \Omega^*_{(-)}\xrightarrow{\sim} A\otimes^L\mathbb{F}_p$. Then $u_0$ lifts uniquely to an isomorphism $R\Gamma_{\mathrm{crys}}\xrightarrow{\sim} A$.

It seems to me that $A=\hat{\Omega}^*_{W(-)}$ already satisfies the assumption of the theorem. If so, what is the motivation for taking "strictification"? If not so, where am I misunderstanding?

$\endgroup$
1
  • 2
    $\begingroup$ I can't speak to the authors motivation, but maybe two things: part of their work is that you can use the completion of the saturation of the de Rham complex of any flat lift, which is by far easier to compute (I don't think computing the $W(R)$ or its dR-cplx is that feasalbe, even though I'd like to be proven wrong). Completing and saturation "washes away" the problem of a choice of a lift, which is nice. Second a main thing in this paper is the extension to the singular case, where saturation is again necesarry to erase pathologies. $\endgroup$ Commented Jun 21, 2022 at 11:53

0

You must log in to answer this question.