Globular sets are to simplicial sets as globes are to simplices.
They are one of the major geometric shapes for higher structures: if they satisfy a globular Segal condition then they are equivalent to strict ∞-categories.
The globe category $\mathbb{G}$ is the category whose objects are the natural numbers, denoted here $[n] \in \mathbb{N}$ (N.B. not to be confused with ordinals in any structural sense) and whose morphisms are generated from
for all $n \in \mathbb{N}$ subject to the relations (dropping obvious subscripts)
A globular set, also called an $\omega$-graph, is a presheaf on $\mathbb{G}$. The category of globular sets is the category of presheaves
This means that a globular set $X \in gSet$ is given by a collection of sets $\{X_n\}_{n \in \mathbb{N}}$ (the set of $n$-globes) equipped with functions
called the $n$-source and $n$-target maps (or similar), such that the globular identities hold: for all $n \in \mathbb{N}$
$s_n \circ s_{n+1} = s_n \circ t_{n+1}$
$t_n \circ s_{n+1} = t_n \circ t_{n+1}$.
The globular identities ensure that two sequences of boundary maps
with $n,m \in \mathbb{N}$ and for $f_k, \in \{s_k, t_k\}$ are equal if and only if their last term $f_n$ coincides; for all $n,m \in \mathbb{N}$ we have
For $S$ a globular set we may therefore write unambiguously
with $i_n, s_n, t_m$ the sequence of $m$ consecutive identity-assigning, source or target maps, respectively.
The presheaf definition can understood from the point of view of space and quantity: a globular set is a space characterized by the fact that and how it may be probed by mapping standard globes into it: the set $S_n$ assigned by a globular set to the standard $n$-globe $[n]$ is the set of $n$-globes in this space, hence the way of mapping a standard $n$-globe into this spaces.
More generally:
A globular object $X$ in a category $\mathcal{C}$ is a functor $X : \mathbb{G}^{\mathrm{op}} \to \mathcal{C}$.
If to the globe category we add additional generating morphisms
satisfying the relations
we obtain the reflexive globe category, a presheaf on which is a reflexive globular set. In this case the morphism
is called the $n$th identity assigning map; it satisfies the globular identities:
A presheaf on the full subcategory of the globe category containing only the integers $[0]$ through $[n]$ is called an $n$-globular set or an $n$-graph. An $n$-globular set may be identified with an $\infty$-globular set which is empty above dimension $n$.
Note that a $1$-globular set is just a directed graph, and a $0$-globular set is just a set.
Any strict 2-category or bicategory has an underlying 2-globular set. Likewise, any tricategory has an underlying 3-globular set. Globular sets can be used as underlying data for n-categories as well; see for instance Batanin ∞-category.
A strict omega-category is a globular set $C$ equipped in each degree with the structure of a category such that for every pair $k_1 \lt k_2 \in \mathbb{N}$ the induced structure on the 2-graph $C_{k_2} \stackrel{\to}{\to} C_{k_1} \stackrel{\to}{\to} C_0$ is that of a strict 2-category.
The globular $n$-globe $G_n$ is the globular set represented by $n$, i.e. $G_n(-) := Hom_G(-,n)$.
The category of globes is not a weak test category according to Scholium 8.4.14 in Cisinski 06.
However, if we construct the free strict monoidal category on the category of globes, while ensuring that the terminal object becomes the monoidal unit, then the resulting category of polyglobes is a test category.
directed $n$-graph
Also related is the notion of computad, which is similar to a globular set in some ways, but allows “formal composites” of $n$-cells to appear in the sources and targets of $(n+1)$-cells.
The definition is reviewed around def. 1.4.5, p. 49 of
See also
Sjoerd Crans, On combinatorial models for higher dimensional homotopies (web)
R. Street, The petit topos of globular sets , JPAA 154 (2000) pp.299-315.
Denis-Charles Cisinski, Les préfaisceaux comme types d’homotopie, Astérisque 308 Soc. Math. France (2006), 392 pages [numdam:AST_2006__308__R1_0 pdf]
The definition of globular set, without using that term, is in 2.2 and 2.3 of the following paper:
and crossed complexes“, Cah. Top. G'eom. Diff. 22 (1981) 371-386.
The following paper constructs from the cubical case a strict globular $\omega$-groupoid of a filtered space:
$\omega$-groupoid of a filtered space“, Homotopy, Homology and Applications, 10 (2008), No. 1, pp.327-343.
Last revised on October 12, 2022 at 08:17:24. See the history of this page for a list of all contributions to it.