Problem: Let $b\in E$(normed vector space), show that $d(b,B(a,r))=d(b,\overline{B}(a,r))$.
Notation: $d(a,C)=inf\{d(a,x):x\in C\}$
My attempt: Well, since $B(a,r)\subset \overline(B)(a,r)$, then $d(b,\overline{B}(a,r)) \leq d(b,B(a,r))$. So, I suppose $d(b,\overline{B}(a,r)) < d(b,B(a,r))$ to find a contradiction.
The real problem is that I have no idea how to continue nor even if im in the right path.
PD: Please do not use convergence :)
Thanks for the help!
Let $x \in \overset {-} B (a,r)$. Then there exists $\{x_n\} \subset B (a,r)$ such that $x_n \to x$. We have $d(b,B (a,r)) \leq d(b,x_n)$ for all $n$. If you let $n \to \infty$ you get $d(b,B (a,r)) \leq d(b,x)$. Taking inf over all $x$ we get $d(b,B (a,r)) \leq d(\overset {-} B (a,r))$. For a different proof not involving sequences let $x \in \overset {-} B (a,r)$ and $\epsilon >0$. There exists $y \in B(b,r)$ such that $d(x,y) <\epsilon$. We have $d(b,B(a,r))\leq d(b,y)<d(b,x)+\epsilon$. Take infimum over $x$ to get $d(b,B(a,r))\leq d(b,\overset {-} B (a,r))+\epsilon$. Since $\epsilon$ is arbitrary this gives $d(b,B(a,r))\leq d(b,\overset {-} B (a,r))$