partial model category

Partial model categories are one of the many intermediate notions between relative categories and model categories.

They axiomatize those properties of model categories that only involve weak equivalences.

A **partial model category** is a relative category $(C, W)$ that satisfies the 2-out-of-6 property (if $s r$ and $t s$ are weak equivalences, then so are $r$, $s$, $t$, $t s r$) and admits a 3-arrow calculus?, i.e., there are subcategories $U, V \subseteq W$ of the weak equivalences (which can be thought of as analogues of acyclic cofibrations and acyclic fibrations) such that $U$ is closed under cobase changes along arbitrary morphisms of $C$ (which are required to exist), $V$ is closed under base changes along arbitrary morphisms of $C$, and any weak equivalence can be functorially factored as the composition $v u$ for some $u\in U$ and $v\in V$.

If $(C,W)$ is a partial model category, then any Reedy fibrant replacement of the Rezk nerve? $N(C,W)$ is a complete Segal space.

- Clark Barwick, Daniel M. Kan,
*Partial model categories and their simplicial nerves*, arXiv.

Last revised on July 12, 2021 at 16:18:57. See the history of this page for a list of all contributions to it.