On a question by Alexey Ostrovsky concerning preservation of completeness (Q2268646): Difference between revisions

The authors have established the following results: Let \(X\) be a complete metric space, and let \(Y\) be a metrizable space. Let \(f:X\to Y\) be a surjective continuous map such that for each countable metrically discrete \(C\) and a neighbourhood \(V\), there is \(L\) such that \(C\subset L\subset V\) and \(f(L)\) is resolvable. Then \(Y\) is completely metrizable. Corollary: Let \(f:X\to Y\) be a continuous map from a completely metrizable zero-dimensional space \(X\) onto a metrizable space \(Y\). If \(f\) maps clopen sets in \(X\) to resolvable sets in \(Y\), then \(Y\) is completely metrizable. Corollary: Let \(X\) be a zero-dimensional hereditarily Baire metrizable space and let \(Y\) be a metrizable space. Let \(f:X\to Y\) be a continuous map such that \(f\) takes clopen sets in \(X\) to resolvable sets in \(Y\). Then \(Y\) is hereditarily Baire. In this connection, we recall the Ostrovsky theorem: Let \(X\) be a \(G_\delta\)-subspace of the irrationals and \(Y\) a metrizable space. Let \(f:X\to Y\) be a surjective continuous function such that \(f\) takes any clopen \(W\) in \(X\) to a set \(f(W)=U\cup V\) where \(U\) is open and \(V\) is closed in \(Y\). Then \(Y\) is completely metrizable. Ostrovsky's question: whether the union in the condition imposed upon \(f(W)\) can be changed to the intersection, or another combination of closed and open sets in \(Y\). The authors give a positive answer to this question.