5
$\begingroup$

Let $C$ be a model category and $B$ a direct category. By Theorem 5.1.3 in Mark Hovey's book Model categories, there is a model category structure on the diagram category $C^B$ such that weak equivalences and fibrations are defined pointwise.

Assume that $C\leftrightarrows D$ is a Quillen equivalence. By pointwise application, we obtain a pair of functors $C^B\leftrightarrows D^B$. Is this also a Quillen equivalence?

$\endgroup$
7
  • $\begingroup$ (If the question is too basic, please feel free to migrate it over to MSE!) $\endgroup$
    – Rasmus
    Jul 12, 2013 at 19:46
  • $\begingroup$ I think it follows easily from the definition and from the fact that a cofibrant object in $C^B$ is pointwise cofibrant. $\endgroup$ Jul 12, 2013 at 20:27
  • $\begingroup$ @FernandoMuro Do we want that fibrant objects in $D^B$ are pointwise fibrant, since the OP write that "weak equivalences and fibrations are defined pointwise"? $\endgroup$ Jul 12, 2013 at 20:30
  • 1
    $\begingroup$ I'm not saying that pointwise cofibrant diagrams are cofibrant, but that cofibrant diagrams are pointwise cofibrant, see for instance Corollary 15.3.12 in Hirschhorn's book. That's enough, I think. $\endgroup$ Jul 12, 2013 at 21:08
  • 2
    $\begingroup$ Anyway, I've just found out that it is Proposition 15.4.1 in Hirschhorn's book. $\endgroup$ Jul 12, 2013 at 21:35

1 Answer 1

1
$\begingroup$

Edit: As Fernando pointed out, the claim can be found in Hirschhorn's book as Proposition 15.4.1.

Here is my own attempt. We write $(F,G)\colon C\leftrightarrows D$ for the given Quillen equivalence and $(F^X,G^X)\colon C^X\leftrightarrows D^X$ for the pointwise induced adjunction. First we check that $(F^X,G^X)$ is a Quillen adjunction. It suffices to show that $G^X$ preserves fibrations and trivial fibrations which is clear because of the pointwise definition of fibrations and weak equivalences. It remains to show that $(F^X,G^X)$ is a Quillen equivalence, that is, a map $F^X(c^X)\to d^X$ in $D^X$ with $c^X$ cofibrant and $d^X$ fibrant is a weak equivalence if and only if the adjoint map $c^X\to G^X(d^X)$ is a weak equivalence in $C^X$. This follows from the pointwise definition of weak equivalences and fibrations once we know that cofibrant objects in $C^X$ are pointwise cofibrant (which can be shown inductively similarly to the proof of Proposition 6.8 in DIAGRAM SPACES AND SYMMETRIC SPECTRA by STEFFEN SAGAVE AND CHRISTIAN SCHLICHTKRULL).

$\endgroup$
7
  • $\begingroup$ (If this is correct, then it is easier than I thought.) $\endgroup$
    – Rasmus
    Jul 12, 2013 at 21:52
  • $\begingroup$ $c^X$ must be cofibrant and $d^X$ fibrant. $\endgroup$ Jul 13, 2013 at 11:53
  • $\begingroup$ @FernandoMuro: Indeed. That's seems to be missing in the wikipedia article. $\endgroup$
    – Rasmus
    Jul 13, 2013 at 13:30
  • $\begingroup$ (I added it here and on wikipedia.) $\endgroup$
    – Rasmus
    Jul 13, 2013 at 13:37
  • $\begingroup$ This is why it is relevant that cofibrant diagrams are pointwise cofibrant $\endgroup$ Jul 13, 2013 at 14:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.