Abbreviation: DCPO

A directed complete partial order is a poset $\mathbf{P}=\langle P,\leq \rangle $ such that every directed subset of $P$ has a least upper bound: $\forall D\subseteq P\ (D\ne\emptyset\mbox{and}\forall x,y\in D\ \exists z\in D (x,y\le z)\Longrightarrow \exists z\in P(z=\bigvee D))$.

Let $\mathbf{P}$ and $\mathbf{Q}$ be directed complete partial orders. A morphism from $\mathbf{P}$ to $\mathbf{Q}$ is a function $f:Parrow Q$ that is Scott-continuous, which means that $f$ preserves all directed joins:

$z=\bigvee D\Longrightarrow f(z)= \bigvee f[D]$

Example 1: $\langle \mathbb{R},\leq \rangle $, the real numbers with the standard order. Example 1: $\langle P(S),\subseteq \rangle $, the collection of subsets of a sets $S$, ordered by inclusion.

$\begin{array}{lr} f(1)= &1\\ f(2)= &\\ f(3)= &\\ f(4)= &\\ f(5)= &\\ f(6)= &\\ \end{array}$

Complete semilattices

Directed partial orders