Lets look at the following question. Why would the proof be incorrect?
Let U be any set. Prove that there exists an $A \in \mathcal{P}(U)$ such that for every $B \in \mathcal{P}(U)$, $A \cap B = B$.
Proof. Our goal is $\exists A \in \mathcal{P}(U) \forall B \in \mathcal{P}(U)(A \cap B = B)$. We choose $A = B$. Clearly $B \cap B = B$.
I'm quite certain that the correct answer is Let A = U. What I'm struggling with is explaining why we can't Let A = B. Or is it fine to choose A = B?
Prove that there exist a real number x such that for every real number y, $y^2x = y-x$.
Our goal would be $\exists x \in \mathbb{R} \forall y \in \mathbb{R}(y^2x = y-x)$. The answer would be to choose $x = \frac{y}{y^2+1}$
In this case x is also changing with every chosen y. Is it because x is more specific in this case?
– fesodes Jan 14 '20 at 03:09