6
$\begingroup$

The category (that is, $(\infty, 1)$-category) of derived affine schemes is the opposite category of the localization of simplicial commutative rings in weak equivalences.

See extensive category. Does this result exist in the literature?

I have difficulty transferring the usual (obvious) proof from discrete rings to simplicial ones, because the product $A \times B$ is not cofibrant. And with cofibrant replacement, idemptotents cease to be strict.

It's surprising if some work is actually needed here, because conceptually the fact should be very simple. I'm probably just missing something.

UPD. I removed the erroneous comment about equivalence with the category of dg algebras, but in reality this result is enough for me for algebras over the field of characteristic 0 (where there is an equivalence with dg algebras), if this, of course, simplifies something.

$\endgroup$

1 Answer 1

5
$\begingroup$

(First, animated rings are not modeled by connective CDGA's beyond $\mathbb Q$-algebras.)

An $E_\infty$-version is in [Lurie: Spectral Algebraic Geometry, (b) in Proof of Prop B.6.1.6], which implies the animated version. The proof is the same as the classical version: given two commutative ring spectra $A,B$, the unit $1\in\pi_0(A\times B)\cong\pi_0A\times\pi_0B$ is the sum $e_1+e_2$ of two idempotents, where $e_1$ is the image of the unit $1\in\pi_0A$ under the map $\pi_0A\to\pi_0A\times\pi_0B$. The image of $e_1$ (resp. $e_2$) under the map $\pi_0A\times\pi_0B\to\pi_0B$ (resp. $\pi_0A\times\pi_0B\to\pi_0A$) is $0$, thus the image of $1$ under the map $\pi_0A\times\pi_0B\to\pi_0(A\otimes_{A\times B}B)$ is $0$ as well, which shows that the ring $\pi_0(A\otimes_{A\times B}B)$ is $0$, and thus so is the commutative ring spectrum $A\otimes_{A\times B}B$.

$\endgroup$
3
  • 2
    $\begingroup$ Oh, I was actually thinking about algebras over the field of characteristic 0, thanks. Thank you very much for the link and answer! Generally speaking, the disjointness of coproducts is significantly weaker than the extensiveness (for example, coproducts in vector spaces are disjoint, but not even universal, and certainly not van Kampen). But your answer inspired me to think again about how to show the required equivalence of slice categories, I’ll try and write if I succeed. $\endgroup$ Commented Jun 9 at 17:47
  • 1
    $\begingroup$ @ArshakAivazian The finite coproduct being universal is exactly the same as what is on nLab for usual rings, can be found in the same section, i.e. Prop B.1.6.3. $\endgroup$
    – Z. M
    Commented Jun 10 at 7:32
  • $\begingroup$ Indeed, but I found it in B.6.1.5(d). The question is resolved, thank you very much! $\endgroup$ Commented Jun 10 at 23:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .