0
$\begingroup$

Let $X$ be a completely regular space (that is, any closed subset and point not in the closed set can be separated by continuous function).

Let $C = C(X,[0,1])$ be the collection of continuous functions $f$ defined from $X$ to $[0,1].$

Recall the construction of Stone-Cech Compactification of $X$ via embedding in a cube.

For any $x\in X,$ defined a map $\phi:X\to [0,1]^C$ by $\phi(x) = (f(x))_{f\in C}.$ By Tychonoff's theorem, the space $[0,1]^C$ is compact in Tychonoff product topology. Note that $\overline{\phi(X)},$ which is a closed subset of a compact set, is compact in $[0,1]^C$. Then we let $\overline{\phi(X)}$ to be the Stone-Cech compactification of $X,$ denoted as $\beta X.$

Question: Let $V$ be an open subset in $\beta X.$ Is it true that $U = V\cap X$ open in $\beta X?$

$\endgroup$
  • $\begingroup$ math.stackexchange.com/questions/353413/… $\endgroup$ – Eric Wofsey Mar 4 '18 at 8:44
  • $\begingroup$ @EricWofsey: How does the link answer my question? $\endgroup$ – Idonknow Mar 4 '18 at 8:49
  • $\begingroup$ The answer to your question is yes iff $X$ is open in $\beta X$. $\endgroup$ – Eric Wofsey Mar 4 '18 at 8:50
  • $\begingroup$ @EricWofsey: I see. By your link, $X$ is open in $\beta X$ iff $X$ is locally compact. $\endgroup$ – Idonknow Mar 4 '18 at 8:53
3
$\begingroup$

Firstly, the intersection is not well-defined as it stands. $X$ is not a subset of $\beta X$ in this setup. What does hold is that $\phi[X]$ is a subset of $\beta X \subseteq [0,1]^C$ which is homeomorphic to $X$ (this is what $\phi$ being an embedding means). Often $X$ is identified with this image, also in notation. So $V$ open in $\beta X$ (so relatively open in the product) means that $V \cap \phi[X]$ is open in $\phi[X]$ (or $X$), not in $\beta X$, unless $X$ or $\phi[X]$ is open in $\beta X$ (which happens iff $X$ is locally compact Hausdorff).

So making the identification $X \sim \phi[X]$ : for all open $V \subseteq \beta X$: $V \cap X$ open in $\beta X$ iff $X$ is locally compact Hausdorff.

$\endgroup$
  • $\begingroup$ To the proposer: Consider the case $V=\beta X, $ so $V\cap \phi[X]=\phi[X].$... A completely regular space $X$ is locally compact iff $c[X] $ is open in $cX$ for every compactification $c:X\to cX$ iff $c[X]$ is open in $cX$ for some compactification $c:X\to cX.$............+1 $\endgroup$ – DanielWainfleet Mar 5 '18 at 10:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.