Consider the metric space $Q$ of rational numbers with the Euclidean metric of $R$. Let $S$ consist of all rational numbers in the open interval ($a, b$), where $a$ and $b$ are irrational. Then $S$ is a closed and bounded subset of $Q$ which is not compact.

I am trying to prove this statement. It's easy to show that $S$ is closed and bounded but I'm having trouble showing that it is not compact. I think I need to find an example of an open covering of $S$ which doesn't have any finite subcover, but I can't think of such an example. Any help? Thanks.

  • 2
    $\begingroup$ You can use the fact that for metric spaces, compact is equivalent to sequentially compact. Finding sequences in $S$ without accumulation point (in $S$) is not difficult. But having found such a sequence, you can probably easily construct an open cover without finite subcover. $\endgroup$ – Daniel Fischer Apr 27 '14 at 17:53

HINT: Pick a sequence $a_n$ converging to $a$ from above, and consider $(a_n,b)$ as the open cover.

Another approach would be to prove that $(a,b)\cap\Bbb Q$ is not complete and that a compact metric space is always complete.

  • $\begingroup$ I guess smth like (a+(b-a)/n,b) would work? $\endgroup$ – takecare Apr 27 '14 at 18:07
  • $\begingroup$ For example. You can also insist on the intervals not to be clopen, by picking $a_n\to a$ and $b_n\to b$ of rational numbers (assuming $a_i<b_j$ for all $i,j$ and consider $(a_n,b_n)$. But the principle's the same. In fact, you can even take $(a+\frac1n,b)$. $\endgroup$ – Asaf Karagila Apr 27 '14 at 18:09

Consider $S=(a,b) \cap \mathbb Q$ as a subset of $\mathbb R$. Then $S$ is compact relatively to $\mathbb Q$ if and only if $S$ is compact relatively to $\mathbb R$. Clearly $S$ is not compact in $\mathbb R$, then $S$ is not compact in $\mathbb Q$.

  • $\begingroup$ I have edited your post - added MathJax - for better readability. (Please, check whether I did not unintentionally change something.) For some basic information about writing math at this site see e.g. here, here, here and here. $\endgroup$ – Martin Sleziak Jan 7 '17 at 16:38

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.