Factorization of maps on limits of inverse systems in Top (Q2217233)

The paper investigates the factorization of maps on limits of inverse systems via limits of subsystems (which is more general than factorization of maps on products via subproducts). For some productive and closed hereditary (i.e., epireflexive) class \(\mathcal{C}\) of Hausdorff spaces a cardinal \(\kappa\) will be obtained such that every continuous map of an inverse system in \(\mathcal{C}\) into a given space can be factorized via a limit of a subsystem of cardinality less than \(\kappa\). For some results in the paper under review, large cardinals are needed. For instance, it is shown in the paper that, for a given class \(\mathcal{C}\) in our context, if we let \(\mathfrak{m}_\mathcal{C}\) denote the least cardinal \(\kappa\) such that a discrete space of cardinality \(\kappa\) does not belong to \(\mathcal{C}\), then \(\mathfrak{m}_\mathcal{C}\) is either measurable or \(\infty\) (where \(\infty\) denotes the result of the situation on which we are looking for the smallest cardinal of a certain class of cardinals and the class turns out to be empty). That is, given an epireflexive class of Hausdorff spaces, either for any infinite cardinal \(\kappa\) there is a discrete space of size \(\kappa\) in the class or the smallest possible cardinality of a discrete space which is not in the class is given by a measurable cardinal. Some factorizations in the paper use strongly compact cardinals, and at the end of the paper the author remarks that it is not clear whether such large cardinals are really needed in those particular results.