1
$\begingroup$

Let $(X,\tau)$ be a topological space. $A\subseteq X$ is said to be regular open if $A = \text{int}(\text{cl}(A))$ and let $\text{RO}(X,\tau)$ denote the collection of regular open sets of $X$. A standard exercise exercise shows that $(\text{RO}(X,\tau),\subseteq)$ is not only a lattice, but even a Boolean algebra.

Question. Is there a topological space $(X,\tau)$ such that $$\text{RO}(X,\tau) \cong {\cal P}(\omega)/(\text{fin})?$$

(The Boolean algebra ${\cal P}(\omega)/(\text{fin})$ is defined here.)

Improve this question
$\endgroup$
1
  • 6
    $\begingroup$ No, $\textrm{RO}(X,\tau)$ is complete, while ${\mathcal P}(\omega)/\textrm{fin}$ is not. $\endgroup$
    – Keith Kearnes
    Commented Sep 20, 2018 at 12:36

2 Answers 2

Reset to default
6
$\begingroup$

No, $\mathrm{RO}(X)$ is complete; $\mathcal{P}(\omega)/\mathit{fin}$ is not (no strictly increasing sequence has a supremum).

Improve this answer
$\endgroup$
1
  • 2
    $\begingroup$ And conversely, every complete Boolean algebra is isomorphic to $\mathrm{RO}(X)$ for a suitable $X$. $\endgroup$
    – Emil Jeřábek
    Commented Sep 20, 2018 at 14:57
4
$\begingroup$

The algebra of regular open sets in a completely regular space is complete. As your algebra is not complete, it cannot be isomorphic to such an algebra. So the answer is no, at least on the class of completely regular spaces.

Improve this answer
$\endgroup$
1
  • 5
    $\begingroup$ no separation axioms needed; that algebra is always complete: $\sup\mathcal{O}=\operatorname{int}\operatorname{cl}(\bigcup\mathcal{O})$. $\endgroup$
    – KP Hart
    Commented Sep 20, 2018 at 14:45

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .