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

I have two problems:

1.- Let $X$ be a compact Hausdorff space, then $X$ has a basis with cardinality less than or equal to $|X|$.

2.- Let $X$ be a Hausdorff space and $D$ a dense subset in $X$, then $|X|\leq|P(P(D))|$, where $P(D)$ is the power set of $D$.

If somebody know where I can find the proofs, tell me please.

share|cite|improve this question
All-caps in questions might be considered a bit rude. Also, is this homework? EDIT: Also also, I'm not sure I see two problems. – Jack Henahan Jun 15 '11 at 1:45
I only see one question here. What is the other problem? – Zev Chonoles Jun 15 '11 at 1:45
Thanks, @Zev, for editing in the questions. – Jack Henahan Jun 15 '11 at 1:50
Should the first problem say "...less than or equal to $|X|$" or perhaps "... less than or equal to that of $X$"? – Arturo Magidin Jun 15 '11 at 4:19

Problem 1. follows from the fact that the network weight of a compact Hausdorff space equals the weight of it, but this might be unfamiliar. A dedicated proof of this fact:

Let $\mathcal{T}$ be the original topology on $X$, the compact Hausdorff one. We can assume that $X$ is infinite, otherwise $X$ is discrete and finite, and its set of singletons is the required base.

For every pair $\{x,y\}$ with $x \neq y$ from $X$, we pick $U(x,y)$ and $V(x,y)$ from $\mathcal{T}$ such that $x \in U(x,y), y \in V(x,y), U(x,y) \cap V(x,y) = \emptyset$. Let $\mathcal{T}'$ be the topology generated by all sets $U(x,y)$ and $V(x,y)$, so that $\mathcal{T}'$ has a base of weight at most $X$ (as the cardinality of $X$ and the set of pairs from $X$ are equal, $X$ being infinite).

The identity map $f(x) = x$ from $(X,\mathcal{T})$ to $(X,\mathcal{T}')$ is a bijection that is continuous (as $\mathcal{T}' \subset \mathcal{T}$) and goes from a compact space to a Hausdorff one (we basically constructed the latter space to be Hausdorff), and so is a closed map as well, and hence a homeomorphism, from which we conclude that $\mathcal{T} = \mathcal{T}'$ and we are done.

share|cite|improve this answer

Let $A_q$ be the element in $2^{P(D)}$ which associates with each powerset $D$ a $1$ if the closure contains $q$, and $0$ otherwise. Then I claim that $A_q$ is unique for each point $q$ in the Hausdorff space, which proves the claim since $$|2^{P(D)}|=|P(P(D))|$$

If $A_p=A_q$, then consider disjoint neighborhoods $V_p$ and $V_q$ of $p$ and $q$ respectively. The closure of the elements of $D$ inside $V_p$ (which contains $p$ since every open set containing p must contain one of these points), cannot contain $q$ since it is contained in the closed set $\bar{V_q}$, so we have a contradiction!



Edit: Sorry, I only saw your second question when you originally posted.

Edit2: Fixed legitimate concerns in my proof.

share|cite|improve this answer
Rofler: The interval $(0,1)$ and the open set $(0,1/\pi)\cup(1/\pi,1)$ contain the same rationals, so I am not sure I understand your first sentence. – Andrés E. Caicedo Jun 15 '11 at 2:10
I have edited my answer to reflect your legitimate concern that my original proof was total nonsense :D – Rofler Jun 15 '11 at 6:37
More conventional rephrasing: every element of the closure of $D$ is the limit of a filter on $D$. How many filters are there on $D$? At most $2^{2^{|D|}}$. – GEdgar Jun 15 '11 at 13:04
Sorry, I never learned about filters. I just blindly applied the definition of Hausdorff space :P – Rofler Jun 15 '11 at 13:28

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.