10
$\begingroup$

Using the definitions from Peter May's A Concise Course in Algebraic Topology, a topological space $X$ is weak Hausdorff if for every compact Hausdorff space $K$ and continuous function $f:K\to X$, $f(K)$ is closed in $X$. A subset $A$ of $X$ is said to be compactly closed if for any compact Hausdorff space $K$ and continuous map $f:K\to X$, $f^{-1}(A)$ is closed in $K$. Finally, $X$ is said to be a $k$-space if every compactly closed subset is closed.

In the lemma at the top of page 40, May gives a characterization of weak Hausdorff spaces which can be paraphrased as such:

A $k$-space $X$ is weak Hausdorff if and only if the diagonal $\Delta_X$ is compactly closed in $X\times X$ under the product topology.

I was searching for a counterexample to this lemma when the $k$-space assumption is omitted. I came up with a proof of this lemma, and the forward direction didn't require the $k$-space assumption. So in other words, I am asking

What is an example of a topological space $X$ such that $\Delta_X$ is compactly closed in $X\times X$, but $X$ is not weak Hausdorff?

By the lemma, such an example must not be a $k$-space.

$\endgroup$
2
  • 1
    $\begingroup$ I have updated the preprint on arXiv and now it contains the answer to your question, see my answer below. $\endgroup$ Commented Apr 18, 2023 at 18:02
  • $\begingroup$ @TarasBanakh Thank you! I appreciate you updating the preprint to answer my question. $\endgroup$
    – Anon
    Commented Apr 20, 2023 at 20:13

1 Answer 1

11
$\begingroup$

The space $X$ constructed in Theorem 1.5 of this preprint has the required properties. This space contains a non-closed compact metrizable subspace $K$, so is not weakly Hausdorff.

On the other hand, the space $X$ is $k_2$-metrizable, which means that $X$ is the image of a metrizable space $M$ under a continuous bijective proper map $f:M\to X$ such that for every continuous map $g:H\to X$ from a compact Hausdorff space $H$, the map $f^{-1}\circ g:H\to M$ is continuous. This property implies that the diagonal $\Delta_X$ is compactly closed in $X\times X$.

$\endgroup$
2
  • 5
    $\begingroup$ You might enjoy knowing that the question dates back at least to Hoffmann's 1979 paper On Weak Hausdorff Spaces (see 4.2), and is repeated in Hoffmann's 1982 paper Factorization of cones II, with applications to weak Hausdorff spaces (see Remark 3.5). Until your post, I don't think any counterexample had appeared in the literature since Hoffmann's original question. $\endgroup$
    – Tyrone
    Commented Apr 19, 2023 at 19:06
  • 2
    $\begingroup$ @Tyrone Thank you for this historical information. Very interesting! $\endgroup$ Commented Apr 19, 2023 at 20:44

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .