Remarks on metrizability and generalized metric spaces (Q1295170)

Professor Nagata's remarks and reviews on metrization are always worth close scrutiny by anyone interested in topology. The two main results of the paper under review are: Theorem 1. A regular space \(X\) is metrizable iff it has a \(\sigma\)-closure-preserving \(k\)-network \(\bigcup{\mathcal F}_n\) such that each \({\mathcal F}_n\) is pseudo-interior-preserving; Theorem 2. A \(T_1\)-space \(X\) is metrizable iff there is a decreasing \(g\)-function \(g(n,x)\), \(n\in \mathbb{N}\), \(x\in X\), where each \(g(n,x)\) is a (not necessarily open) neighbourhood of \(x\), which satisfies \(\overline{A}= \bigcap_{n\in\mathbb{N}} \bigcup_{x\in A}g(n,x)\) for every \(A\subseteq X\) and \(y\in g(n,x)\) implies \(g(n,y)\subseteq g(n-1,x)\) \((n\geq 2)\). Five problems are posed.