The Glicksberg theorem for Tychonoff extension properties (Q266336)

All topological spaces under consideration are Tychonoff and infinite. A celebrated theorem by \textit{I. Glicksberg} [Trans. Am. Math. Soc. 90, 369--382 (1959; Zbl 0089.38702)] asserts that the Čech-Stone compactification of a product is homeomorphic to the product of the Čech-Stone compactifications of the factors exactly in the case when the product is pseudocompact. The Čech-Stone compactification of a space \(X\) is the largest extension of \(X\) which is compact and in which \(X\) can be embedded as a dense subspace. Largest extensions exist relative to other properties besides compactness: a property \(P\) is a Tychonoff extension property if \(P\) is closed-hereditary, invariant under arbitrary products and shared by every compact space; see [\textit{J. R. Porter} and \textit{R. G. Woods}, Extensions and absolutes of Hausdorff spaces. New York etc.: Springer-Verlag (1988; Zbl 0652.54016)]. If \(P\) is a Tychonoff extension property, then every space \(X\) has a maximal \(P\)-extension \(\beta_P(X)\), which is unique modulo homeomorphism. Thus it makes sense to ask for which Tychonoff extension properties an analogue of Glicksberg's theorem holds. Some results in this direction were known, both in the negative and in the positive side, see [\textit{S. Broverman}, Can. Math. Bull. 19, 13--19 (1976; Zbl 0334.54004)]. In the paper under review the author initiates an abstract and detailed study of Glicksberg-like theorems. Among the many results presented by the author, it is proved that Glicksberg's theorem holds for every Tychonoff extension property \(P\) implying \( \omega\)-hyperboundedness, where a topological space \(X\) is \( \omega\)-hyperbounded if every union of a countable family of compact subsets of \(X\) is contained in a compact subset of \(X\). Moreover, for every infinite cardinal \(\alpha\), if \(P\) implies \( \alpha \)-boundedness, then \(P\) satisfies Glicksberg's theorem up to products with at most \(\alpha\) factors. Sections 1 and 2 in the paper contain technical results needed in the subsequent sections; however, some notions introduced there have independent interest and are likely to find further applications. In particular, in Section 2 the author introduces an interesting notion of countable iterations of an ultrafilter; compare to [\textit{D. Booth}, Ann. Math. Logic 2, 1--24 (1970; Zbl 0231.02067)]. The paper contains also many interesting questions and suggestions for further research.