Homeomorphisms of Hashimoto topologies (Q1790226)

Let \((X,\tau)\) be a \(T_{1}\) topological space and \(\mathcal{I}\) be an admissible ideal. Then the family \(\{ U \setminus P : U \in \tau, P\in \mathcal{I}\}\) is a base of a topology. If \((X,\tau)\) is a second-countable topological space and \(\mathcal{I}\) is a \(\sigma\)-ideal then this kind of topology is called a ``Hashimoto topology'' and is denoted by \((X,\tau_{\mathcal{I}})\). The most common Hashimoto topology is the one on the real line consisting of sets \(U\setminus P\), where \(U\) is open in the Euclidean topology and \(P\) belongs to the family of all Lebesgue null sets, which is denoted by \(\mathcal{N}\). In this paper, the authors study the existence of homeomorphisms between Hashimoto topologies \((X,\tau_{\mathcal{I}})\), where \(\tau\) is the Euclidean topology on \(\mathbb{R}\) (or \( [0,1]\)) and \(\mathcal{I}\) is a certain admissible \(\sigma\)-ideal of subsets of \(\mathbb{R}\) (or \( [0,1]\)). Firstly, the authors prove that ``for any admissible \(\sigma\)-ideal \(\mathcal{I}\), every continous function \(f:(\mathbb{R},\tau) \rightarrow (\mathbb{R},\tau_{\mathcal{I}}) \) is constant.'' Then, let \(\mathcal{I}\) be an admissible shift invariant \(\sigma\)-ideal contained in \(\mathcal{N}\) closed under multiplication by positive constants. The authors show that the topologies \((\mathbb{R},\tau_{\mathcal{I}})\) and \((\mathbb{R},\tau_{\mathcal{N}})\) are not homeomorphic. Let \(\mathcal{I}_{w}\) be the \(\sigma\)-ideal of countable subsets. Then, the authors also prove that ``every continuous function \(f:(\mathbb{R},\tau_{\mathcal{I}_{w}})\rightarrow (\mathbb{R},\tau_{\mathcal{I}})\) is constant.'' They also show that if \(\mathcal{I}\) is {\parindent=6mm \begin{itemize}\item[--] the \(\sigma\)-ideal of Lebesgue null sets, \item[--] the \(\sigma\)-ideals of \(\alpha\)-dimensional Hausdorff measure, \item[--] the \(\sigma\)-ideals of sets of \(\sigma\)-finite \(\alpha\)-dimensional Hausdorff measure, \item[--] \(\sigma\)-ideals of sets with Hausdorff dimension not greater than \( \alpha\) for \(\alpha \in (0,1)\), \item[--] the \(\sigma\)-ideal of microscopic sets, \item[--] the \(\sigma\)-ideal of strong measure zero sets (under CH), \item[--] the \(\sigma\)-ideal of meager sets, \end{itemize}} then the Hashimoto topology \((\mathbb{R},\tau_{\mathcal{I}_{w}})\) is not homeomorphic to \( (\mathbb{R},\tau_{\mathcal{I}})\). Let \(\mathcal{K}\) denote the \(\sigma\)-ideals orthogonal to the \(\sigma\)-ideal of merger sets that means \(\sigma\)-ideals \(\mathcal{I}\) for which there exists \(I\in\mathcal{I}\) and \(J\in\mathcal{K}\) such that \(\mathbb{R}=I \cup J \). The authors prove that ``if \( J \) is an admissible \(\sigma\)-ideal orthogonal to \(\mathcal{K}\), then the topological spaces \((\mathbb{R},\tau_{\mathcal{J}})\) and \((\mathbb{R},\tau_{\mathcal{K}})\) are not homeomorphic.'' Finally, they show that \((\mathbb{R},\tau_{\mathcal{I}_{w}})\) and \((\mathbb{R},\tau_{\mathcal{K}})\) are not homeomorphic.