I know there is a duplicate answer here, but I'm having a lot of trouble following it. I apologize if I'm not supposed to set up a new question for this. Please flag me if this violates any of the rules. The proof from the link is as follows:

If there are disjoint sets A,B which intersect E and such that $A\cup B\subseteq E$, then $A\cap E$ and $B\cap E$ form a disconnection of E, so E is disconnected. Conversely, suppose that A' and B' are open sets such that $A'\cup B'\subseteq E$ and $A'\cap E\ne\emptyset$, $B'\cap E\ne\emptyset, A'\cap B'\cap E=\emptyset$ (by definition of disconnected). Then setting

$A=\{x:d(x,A'\cap E)<d(x,B'\cap E)\}$
$B=\{x:d(x,A'\cap E)>d(x,B'\cap E)\}$

We have that A and B are open sets (since $d(x,A'\cap E)-d(x,B'\cap E$) is a continuous function), disjoint, and for any $x\in A'\cap E$, letting r be such that $B(r,x)\subseteq A'$, we have $d(x,A'\cap E)=0$ (of course) and $d(x,B'\cap E)\ge r$ (since $A'\cap B'\cap E=\emptyset$), so $A\ne\emptyset$, and similarly $B\ne\emptyset$.

I understand the first proof (which I assume is the $\leftarrow$ direction). I don't understand the proof in the other direction. My questions are :

  1. Is it correct to say that we set $A=\{x:d(x,A'\cap E)<d(x,B'\cap E)\}$ and $B=\{x:d(x,A'\cap E)>d(x,B'\cap E)\}$ so that this makes A,B disjoint? At least that's the way I'm visualizing it.
  2. How did we get $d(x,A'\cap E)=0$ and $d(x,B'\cap E)\ge r$ and how does this give us that $A \neq \emptyset$ and $B \neq \emptyset$?
  • $\begingroup$ Where did you get that definition of disconnected? I'm having trouble with it,for instance take $E=[1,2],A=[1,\sqrt2),B=(\sqrt2,2]$ then it fits the definition but $E$ is not disconnected,maybe the inclusion shoulde be an equal instead or maybe E is contained in the union? I hope you can enlighten me,good work so far. $\endgroup$ – AHandsomeAlien Oct 21 '16 at 3:14
  • $\begingroup$ I'm assuming,at my comment,the topology induced by the usual metric over the real line,sorry for not mentioning it. $\endgroup$ – AHandsomeAlien Oct 21 '16 at 3:19

Ok I just realized how stupid this question is. It's really late and I apologize.

If it's helpful for anyone else / in case my train of thought is wrong:

If we form a metric ball around X with R that's inside A', naturally the distance between X and A' would be 0 (duh, because it lives in A). Naturally, the distance x from B would be greater than r because there is no intersection between A' and B'. From the definition give, this shows that both A and B are non-empty.


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.