Given $E \subset \mathbb{R}$, show that there exists $x$ so that both $E \cap (x, \infty)$ and $E \cap (-\infty,x)$ are uncountable.
A very simple statement that seems to be eluding me..
Here are my thoughts so far:
Let $$\mathbb{R} = \bigcup_{n \in \mathbb{N}} (-n,n).$$
Then there exists $N \in \mathbb{N}$ so that $E \cap (-N,N)$ is uncountable. Now, as an open interval, $$(-N,N) = \bigcup_{j \in \mathbb{N}} I_j$$ where $I_j \cap I_k = \emptyset$ for $j \neq k$.
Suppose no such $x$ exists.
Notice that one and only one of the sets $E \cap I_j$ is uncountable then, for if not, we would be able to find such an $x$. Call this set $I_{m_0}$.
As an open interval, $$I_{m_0} = \bigcup_{j \in \mathbb{N}} I_{j_0}.$$ Again, only one such $E \cap I_{j_0}$ can be uncountable call it $I_{m_1}$. Repeating this process indefinitely, we may write
$$E \cap [-N,N] = (\bigcup_{j \in \mathbb{N}} E \cap I_j \setminus E \cap I_{m_0}) \cup (\bigcup_{j \in \mathbb{N}} E \cap I_{j_0} \setminus E \cap I_{m_1}) \cup \ldots \cup (\bigcup_{j \in \mathbb{N}} E \cap I_{j_{k-1}} \setminus E \cap I_{m_k}) \cup \ldots $$
But notice that this is a countable union of countable sets. Ergo, a contradiction, so such an $x$ must exist.
Is this proof correct? Something seems awry.
I'd like to know
i) if this proof is valid.
ii) A better proof. Surely one exists.