1
$\begingroup$

Suppose that $(X,\tau)$ is a $T_0$-topological space without isolated point. As it is well-known, all cofinite subsets together with the empty-set forms a topology on $X$ which is a $T_1$ topology on $X$ and we call it the cofinite topology and denote it by $(X,\tau_f)$. Now $(X,\tau\vee\tau_f)$ is another topological space. Can someone help me to prove our disprove the following statement?

If $U\in \tau\vee\tau_f$ but $U\notin \tau$, then $U$ has finite complement.

The topology $\tau\vee\tau_f$ is defined to be the smallest topology which contains both $\tau$ and $\tau_f$.

$\endgroup$
3
$\begingroup$

Here's a counter-example.
Define the topology $\tau$ on $\mathbb N$ with the basis: $$\{\{n,n+2,n+4,\ldots\}:n \in \mathbb N\}.$$ It is clear that this is a basis since the intersection of any two such sets is yet another one of these (or empty) and $\mathbb N = \{0,2,4,\ldots\}\cup\{1,3,5,\ldots\}$.
There are no isolated points in $\tau$: if $n \in U$ for some $U \in \tau$, then $n+2 \in U$.
The topology is $T_0$: if $n<m$, then $n \notin \{m,m+2,m+4,\ldots\}$.
The set $E$ of even numbers is open: $E=\{0,0+2,0+4,\ldots\}$.

In $\tau_f$, the set $\mathbb N\setminus\{10\}$ is open.
Let $E_{10}=(\mathbb N\setminus\{10\}) \cap E$. Then $E_{10} \in\tau\vee\tau_f$.
But $E_{10}$ is not co-finite, and so $E_{10} \notin \tau_f$.
And $E_{10} \notin \tau$ either, because $0 \in E_{10}$ and for $U \in \tau$, if $0 \in U$ then $10 \in U$.

$\endgroup$

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.