If $T$ be a linear operator between from one Banach space $X$ to another $Y$, then $T$ is compact iff $T(B)$ is relatively compact in $Y$ for every bounded subset $B$ of $X$.
Then, the book that I am reading (Applied Analysis by Hunter) says that:
We do not require the range of $T$ to be closed, so $T(B)$ need not be compact even if $B$ is a closed bounded set.
But, I am not quite understanding this. How does the closedness of $T(X)$ relate to the closedness of $T(B)$? Also, how is the closedness of $B$ relevant at all? I mean, regardless of whether $T(X)$ is closed or not, we do not expect $T(B)$ to be closed when $B$ is closed. Do we?
Continuous linear operators are generally not closed mappings. Even from $\mathbb R^2$ to $\mathbb R,$ the function $f((x,y))=x$ maps the closed subset $\{(x,1/x):x\ne 0\}$ of $\mathbb R^2$ onto a non-closed subset of $\mathbb R.$ The author is emphasizing this. A careless reading of the def'n might give the impression that a compact operator is a closed operator, which is generally false. My example $f$ is a compact operator...... A compact operator maps a bounded set onto a set whose closure is compact.
Some books use "completely continuous operator" to mean "compact operator".