scientific article; zbMATH DE number 3969661

From MaRDI portal
Publication:3736558





The authors discuss the topics around the important example of \textit{R. Pol} in Proc. Am. Math. Soc. 82, 634-636 (1981; Zbl 0469.54014). First they discuss the historical background to show that the example connects with the famous cell-like (dimension raising) mapping problem. In the next section, they give a detailed construction of the example, and show that the example motivates their following new definition for infinite dimensionality, since Pol's example shows that weak infinite dimensionality is not hereditary. A collection \(\{(A_ i,B_ i):\) \(i\in J\}\) is a dimension family for X (d-family) if each pair \((A_ i,B_ i)\) of (not necessarily closed) subsets of X is separated in X and if \((A_ i\cup B_ i)\cap (\cup_{j\in J}(\bar A_ j\cap \bar B_ j))=\emptyset\). Such a d- family is essential if whenever \(\{S_ i: i\in J\}\) is a collection of closed subsets of X separating \(A_ i\) and \(B_ i\) in X, then \(\cap_{i\in J}S\not\subset \cup_{i\in J}(\bar A_ i\cap \bar B_ i)\). A space X has dimension \(\geq n\) with respect to dimension families (d-dim(X)\(\geq n)\) if X has an essential d-family of cardinality \(\geq n\). A space X is strongly infinite-dimensional with respect to dimension families (SID-d) if X has a denumerable essential family. Otherwise, X is weakly infinite-dimensional with respect to dimension families (WID-d). They show that X is WID-d if and only if X is hereditarily weak infinite dimensional. They also discuss the relations between their definition and that of Vajnshtejn. But, it is still unknown whether or not they are equivalent. They conclude the paper with a detailed proof of the well- known theorem: there exists a \(G_{\delta}\)-selection for every map between compact metric spaces. Let us remark that another easy proof is known [e.g. the reviewer, Math. Jap. 27, 177-195 (1982; Zbl 0475.54024)], which goes back to the idea of \textit{S. Mazurkiewicz} [Fundam. Math. 10, 311-319 (1927)]. We also note that besides Pol's example there are many applications which use higher dimensional totally disconnected complete metric spaces (e.g. the above paper by the reviewer or \textit{R. Engelking} and \textit{E. Pol} [Diss. Math. 216, 1-45 (1983; Zbl 0541.54042)]).











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3736558)