On extension functions for image space with different separation axioms (Q524326): Difference between revisions

A space \(X\) is called a regular-\(U(\alpha)\)-space provided that for each \(x\in X\) and each closed \(C\subset X\) with \(x\notin F\) there are \(\alpha\)-hulls \(U\) of \(x\) and \(V\) of \(C\) such that \(\overline{U}\cap\overline{V}=\varnothing\) (although not explicitly stated it is also assumed that \(X\) is \(T_1\)). For an ordinal \(\alpha\) an open neighbourhood \(U\) of a set \(A\) is an \(\alpha\)-hull provided there are neighbourhoods \(\langle U_\beta\rangle_{\beta\leq\alpha}\) of \(A\) such that \(\overline{U_\beta}\subset U_{\beta+1}\) whenever \(\beta+1\leq\alpha\) and \(U=\cup_{\beta\leq\alpha}U_\beta\). The main result states that if \(S\) is a dense subset of a space \(X\), \(Y\) is a regular-\(U(\alpha)\)-space and \(f:S\to Y\) is continuous then \(f\) has a continuous extension over \(X\) if and only if \(\bigcap_{i\in I}\overline{f^{-1}(C_i)}=\varnothing\) whenever \(\{C_i \mid i\in I\}\) is a family of closed sets in \(Y\) having empty intersection. While \(\alpha\) is unspecified, it seems that we may take \(\alpha=0\). A related result is given for similarly extending \(\theta\)-continuous functions. Editorial remark: The paper has quite a few, fairly minor but irritating, mistakes in it (in addition to grammatical errors).