\(\mathcal D\)-forced spaces: A new approach to resolvability (Q2493891): Difference between revisions

The authors introduce a ZFC method to build special dense subspaces of Cantor cubes in which dense subsets are fully controlled. To be precise, if \(X\) is a space and \({\mathcal D}\) is a family of dense subsets of \(X\), the authors define \(X\) to be \({\mathcal D}\)-forced if every dense subset of \(X\) is of the form \[ \bigcup\{V\cap D_V:V\in {\mathcal V}\} \] where \({\mathcal V}\) is a maximal family of disjoint open sets in \(X\) and \(D_V\in {\mathcal D}\) for every \(V\in {\mathcal V}\). The main theorem proven states that for any \(\kappa\), any \(Y=\{y_\xi;\xi<\kappa\}\) dense in \(2^{2^\kappa}\) and any family \({\mathcal D}\) of subsets of \(\kappa\) each enumerating a dense subset of \(Y\), there is another dense \(Z=\{z_\xi:\xi\in \kappa\}\subseteq 2^{2^\kappa}\) such that (1) each \(D\in {\mathcal D}\) enumerates a dense subset of \(Z\). (2) \(Z\) is \({\mathcal D}\)-forced (3) every subset of \(Z\) of size \(<\kappa\) is nowhere dense. (4) \(Z\) is nodec (every nowhere dense subset is closed). From this main theorem numerous examples are deduced settling open questions posed in the literature. For example, (1) Taking \({\mathcal D}=\{Y\}\), the resulting \(Z\) is \(\{Z\}\)-forced, and so is a dense submaximal subspace of the Cantor cube \(2^{2^\kappa}\), answering a question from \textit{O. T. Alas, M. Sanchis, M. G. Tkačenko, V. V. Tkachuk} and \textit{R. G. Wilson} [Topology Appl. 107, No. 3, 259--273 (2000; Zbl 0984.54002)]. (2) If \({\mathcal D}=\{D_n:n\in \omega\}\) is a disjoint family of dense subsets of \(Y\subseteq 2^{2^{\omega_1}}\), then the resulting \(Z\) is an \(\omega\)-resolvable space that is not \(\omega_1\)-resolvable (hence not maximally resolvable). This settles an old problem of \textit{J. Ceder} and \textit{T. Pearson} [Pac. J. Math. 22, 31--45 (1967; Zbl 0153.24201)]. Other notable examples constructed include, (3) for each \(\kappa\) a \(\kappa^+\)-resolvable space that can be partitioned into \(\kappa\) many hereditarily irresolvable subspaces, answering a problem of \textit{F. W. Eckertson} [Topology Appl. 79, No. 1, 1--11 (1997; Zbl 0918.54035)]. (4) a maximally resolvable space with \(nwd(X)=| X| \), that is not extra-resolvable, answering a question of \textit{W. W. Comfort} and \textit{W. Hu} [Topol. Proc. 28, No. 1, 81--98 (2004; Zbl 1086.54003)]. Many other results are proven and open problems formulated.