Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Find an uncountable number of subsets of $\ell_{n}^{p}(\mathbb{R})$ and $\ell_{n}^{p}(\mathbb{C})$ that are neither open nor closed.

Attempt: For $\ell_{n}^{p}(\mathbb{R})$ , take the collection of sets $A_n=(n,n+1$] with $\mathbb{R}$ as the index set.

For $\ell_{n}^{p}(\mathbb{C})$ , since $\mathbb{R\subset\mathbb{C}}$, wouldn't the same subsets I use for $\ell_{n}^{p}(\mathbb{R})$ work?

share|cite|improve this question
What's $l^p_n$? – Rudy the Reindeer Jan 31 '12 at 20:39
$\ell_{n}^{p}(\mathbb{R})$ is the space ($\mathbb{R^n},d_p$) where $d_p(x,y)= \max_{1\leq k\leq n}|x_j-y_j|,1\leq p \leq\infty$. – Emir Jan 31 '12 at 22:54
Usually, $\ell^p$ is used to denote that the metric is $d(x,y)=\bigl(\sum |x_i-y_i|^p\bigr)^{1/p}$. What you have above is denoted by $\ell^\infty$. All these metrics induce equivalent norms, however. – David Mitra Jan 31 '12 at 23:06
up vote 2 down vote accepted

Elements in $\ell_n^p$ are sequences with $n$-coordinates. I do not think your $A_n$ will work.

Hint: Think of closed balls with their centers deleted: $\overline B_\epsilon(x)/\{x\}$, where $x\in\ell_n^p$. None of these are open since no open ball containing a boundary point of $\overline B_\epsilon(x)/\{x\}$ is contained in $\overline B_\epsilon(x)/\{x\}$. And none are closed since $x$ is a limit point of $\overline B_\epsilon(x)/\{x\}$ not contained in $\overline B_\epsilon(x)/\{x\}$.

share|cite|improve this answer
I'm a little confused about the reasoning why $\overline B_\epsilon(x)/\{x\}$ is not open -- isn't a closed ball by definition a closed set? – Emir Jan 31 '12 at 22:49
@Emir. I was showing directly that these sets were not open. By "on the boundary of $\overline B_\epsilon(x)/\{x\}$", I meant any point $y$ with $d(x,y)=\epsilon$. No open ball containing $y$ is contained in $\overline B_\epsilon(x)/\{x\}$. So this set is not open. Closed balls are closed sets in these metric spaces, but I think a bit more argument is needed... – David Mitra Jan 31 '12 at 23:02
Ah, now I see why my original argument was insufficient. So if these balls are neither open nor closed in $\ell_{n}^{p}(\mathbb{R})$ they are also neither open nor closed in $\ell_{n}^{p}(\mathbb{C})$ since $\mathbb{R}\subset\mathbb{C}$, right? – Emir Feb 1 '12 at 2:35

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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