3
$\begingroup$

Given a presentable Cartesian symmetric monoidal $\infty$-category $C$, every object is a cocommutative comonoid and for a fixed $Z\in C$ there is an equivalence $C_{/Z}\simeq LCoMod_{Z}(C)$ where the coaction associated to $X→Z$ is $X\to X\times X\to Z\times X$. If we further assume that $Z$ is an algebra object of $C$ then we can equip $LCoMod_Z(C)$ with a monoidal structure, the Day convolution structure, where, roughly, the tensor product of $f\colon X\to Z$ and $g\colon Y\to Z$ is $\mu_Z\circ(f\times g)\colon X\times Y\to Z\times Z\to Z$.

If we consider the trivial map $X\to \ast\to Z$, using the algebra structure of $Z$ to produce the unit $\ast\to Z$ it seems that this should define a comonoid in $C_{/Z}$ via the diagonal $X\to X\times X$ where the map to $Z$ on the left is $\ast$ and the map on the right is $\mu_Z\circ (\ast\times\ast)\circ\Delta\simeq\ast$. In the case of 1-categories this is pretty straightforward to prove.

Even more, given any other map $f\colon X\to Z$, there should be an $(X,\ast)$-comodule structure on $(X,f)$, again via the diagonal map, since $f\simeq \mu_Z\circ (f\times \ast)\circ\Delta$.

Does anyone know of a good categorical way to prove this, ideally $\infty$-categorical? Again, it's all relatively straightforward to check in a 1-category, but in an $\infty$-category it's not at all clear to me how to prove this.

$\endgroup$
3
  • $\begingroup$ I believe I know how to produce a morphism $(X,f)\to (X,f)\otimes (X,\ast)\simeq (X\times X, \mu_Z\circ (f\times \ast)\circ \Delta)$ in the $\infty$-category $C_{/Z}$, but it's not clear to me how to extend it to a fully structured comodule coaction. $\endgroup$ Commented Feb 25, 2021 at 21:31
  • 2
    $\begingroup$ If I understand correctly your question. If $A \to B$ is a map of monoids, then the forgetfull functor $C/A \to C/B$ is monoidal, so it sends comonoid object to comonoid objects. The unit map $ 1 \to Z$ is a monoid morphism, so it sends the comonoid object X to a comonoid object ? $\endgroup$ Commented Feb 25, 2021 at 21:36
  • $\begingroup$ @SimonHenry yeah that makes perfect sense for getting the comonoid structure on $(X,\ast)$, thanks! $\endgroup$ Commented Feb 25, 2021 at 21:37

1 Answer 1

2
$\begingroup$

It is shown in detail how to construct the above described comodule structure on $f\colon X\to Z$ in the thesis of Aras Ergus. Specifically, Construction 2.0.11 of "Hopf algebras and Hopf-Galois extensions in $\infty$-categories" which is available here: https://www.aergus.net/academic/documents/.

$\endgroup$

You must log in to answer this question.

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