I'm trying to prove the following: Let $X$ a topological space and $Z\subseteq X$ $\text{For every topological space } Y, \text{ and for every continuous function } f:Z\rightarrow Y \text{ exists } g:X\rightarrow Y$ continuous such that $g|_Z:Z\rightarrow Y$ is equal to $f$ if and only if exists $r:X\rightarrow Z$ continuous such that $r|_Z(x)=Id_Z(x)$ i.e., $\forall x\in Z: r(x)=x$.

My question is about my proof. Notice that the implication $\Leftarrow$] is easy because $g(x)=f(r(x))$ is the function we want.

The implication [$\Rightarrow$ is my problem. I have two ideas:

1.- If $\delta(Z)=\emptyset$, the proof is finished. Otherwise, We have that $\overline{Z}\cup \overline{X\setminus Z}$ is a closed finite cover of $X$. If I can make $r_1:\overline{Z}\rightarrow Z$ and $r_2: \overline{X\setminus Z}\rightarrow Z$ continuous functions such that are equal to the identity in $Z$, then, the proof is complete.

2.-Since $Z$ is a topological space itself (the topology of $X$ restricted to $z$), we define $r:Z\rightarrow Z$ as $r(x)=x$ for all $x\in Z$. By hypothesis, there exists a continuous extension $g:X\rightarrow Z$ such that $g|_Z(z)=r(z)=z$. So, the proof is complete.

I have some problems accepting my second proof. Is right?. The first idea I can't finish it.

NOTATION: $\delta(A)$ denotes the boundary of a set and $\overline{A}$ the closure of a set.


Hint: Apply the hypothesis to $Y=Z$, $f=\mathrm{id}$. You got it from here?

  • $\begingroup$ Yes, I follow your argument. The idea is extend to f. for a second, I was thinking that argument is wrong; but now, I think is right. $\endgroup$ – Uriel Herrera Mar 30 '18 at 18:57

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.