The product1 of $\left\{ A_{i}\right\} _{i\in I}$ is the product of $\left\{ A_{i}\right\} _{i\in I}$ in $\mathsf{Sets}$ as in ,
.
- 1Further Terminology: Also called the Cartesian product of $\left\{ A_{i}\right\} _{i\in I}$.
Let $\left\{ A_{i}\right\} _{i\in I}$ be a family of sets.
The product1 of $\left\{ A_{i}\right\} _{i\in I}$ is the product of $\left\{ A_{i}\right\} _{i\in I}$ in $\mathsf{Sets}$ as in ,
.
Concretely, the product of $\left\{ A_{i}\right\} _{i\in I}$ is the pair $\webleft (\prod _{i\in I}A_{i},\left\{ \operatorname {\mathrm{\mathrm{pr}}}_{i}\right\} _{i\in I}\webright )$ consisting of:
The Limit. The set $\prod _{i\in I}A_{i}$ defined by
The Cone. The collection
of maps given by
for each $f\in \prod _{i\in I}A_{i}$ and each $i\in I$.
We claim that $\prod _{i\in I}A_{i}$ is the categorical product of $\left\{ A_{i}\right\} _{i\in I}$ in $\mathsf{Sets}$. Indeed, suppose we have, for each $i\in I$, a diagram of the form
for each $x\in P$.
Less formally, we may think of Cartesian products and projection maps as follows:
We think of $\prod _{i\in I}A_{i}$ as the set whose elements are $I$-indexed collections $\webleft (a_{i}\webright )_{i\in I}$ with $a_{i}\in A_{i}$ for each $i\in I$.
We view the projection maps
as being given by
for each $\webleft (a_{j}\webright )_{j\in I}\in \prod _{i\in I}A_{i}$ and each $i\in I$.
Let $\left\{ A_{i}\right\} _{i\in I}$ be a family of sets.
Functoriality. The assignment $\left\{ A_{i}\right\} _{i\in I}\mapsto \prod _{i\in I}A_{i}$ defines a functor
where
Action on Objects. For each $\webleft (A_{i}\webright )_{i\in I}\in \operatorname {\mathrm{Obj}}\webleft (\mathsf{Fun}\webleft (I_{\mathsf{disc}},\mathsf{Sets}\webright )\webright )$, we have
Action on Morphisms. For each $\webleft (A_{i}\webright )_{i\in I},\webleft (B_{i}\webright )_{i\in I}\in \operatorname {\mathrm{Obj}}\webleft (\mathsf{Fun}\webleft (I_{\mathsf{disc}},\mathsf{Sets}\webright )\webright )$, the action on $\operatorname {\mathrm{Hom}}$-sets
of $\prod _{i\in I}$ at $\webleft (\webleft (A_{i}\webright )_{i\in I},\webleft (B_{i}\webright )_{i\in I}\webright )$ is defined by sending a map
in $\operatorname {\mathrm{Nat}}\webleft (\webleft (A_{i}\webright )_{i\in I},\webleft (B_{i}\webright )_{i\in I}\webright )$ to the map of sets
defined by
for each $\webleft (a_{i}\webright )_{i\in I}\in \prod _{i\in I}A_{i}$.