Cellular-compact spaces and their applications (Q2338593)

In the following all spaces are Hausdorff. The authors define a space \(X\) to be \textit{cellular-compact} if for every family \(\mathcal{U}\) of disjoint nonempty open subsets of \(X\) there exists a compact subspace \(K\subset X\) such that \(K\cap U\neq\emptyset\) for every \(U\in\mathcal{U}\), and they recall that in [Monatsh. Math. 186, No. 2, 345--353 (2018; Zbl 1398.54008)], \textit{A. Bella} and \textit{S. Spadaro} defined a space \(X\) to be \textit{cellular-Lindelöf} if for every family \(\mathcal{U}\) of disjoint nonempty open subsets of \(X\) there exists a Lindelöf subspace \(L\subset X\) such that \(L\cap U\neq\emptyset\) for every \(U\in\mathcal{U}\). They note that every cellular-compact space is cellular-Lindelöf, and as was shown by others with the property cellular-Lindelöf, it is straightforward to verify that the property cellular-compact is preserved by continuous maps, inherited by regular-closed subspaces, and possessed by extension spaces, and furthermore, the closure of the set of isolated points of any cellular-compact space is compact. The authors next recall that a space \(X\) is said to be \textit{linearly \(H\)-closed} if every nested open cover of \(X\) has a dense member. This concept was studied in several articles, where it was shown that every linearly \(H\)-closed space \(X\) is \textit{feebly compact} (i.e., every locally finite family of open subsets of \(X\) is finite). They then prove that every cellular-compact space is linearly \(H\)-closed, and hence, or as can easily be shown directly, every cellular-compact space is feebly compact. Examples showing these concepts are distinct are given, including the Isbell-Mrówka space \(\Psi\), which was previously shown by others to be linearly \(H\)-closed, for by the authors' results concerning isolated points, that space cannot be cellular-compact, although it is cellular-Lindelöf. As a way of constructing non-compact, cellular-compact spaces, they prove that for any regular cellular-compact space \(X\) and non-isolated point \(p\) of \(X\), the space \(X\setminus\{p\}\) is cellular-compact if and only if there exists no disjoint local \(\pi\)-base at the point \(p\) in \(X\). A number of interesting corollaries of this theorem are obtained such as: if \(\kappa\) is an uncountable cardinal and \(X=[0,1]^\kappa\), then \(X\setminus\{p\}\) is cellular-compact for any point \(p\in X\), and if \(S\) is the Corson \(\Sigma\)-product in \(X\) based at the origin, then \(S\) is cellular-compact, as well as collectionwise normal, Fréchet-Urysohn, non-compact, and countably compact, and for each point \(p\in S\), the space \(S\setminus\{p\}\) is a cellular-compact space whose extent is \(\kappa\) and which has no dense countably compact subspace. The authors give an example which shows that under CH, there exists a Tychonoff first countable separable cellular-compact space which is not compact. Several results presented are of the form: If \(X\) is a cellular-compact space which has some particular property \(P\), then \(X\) is compact. Two especially nice theorems they obtain are: If \(X\) is a cellular-compact first countable regular space, then \(X\) is countably compact and \(\vert X\vert \leq\mathfrak{c}\). A regular sequential cellular-compact space is maximal cellular-compact if and only if it has a disjoint local \(\pi\)-base at every point, and thus every first countable compact space is maximal cellular-compact. They conclude by raising a number of open questions.