Modifications of sequence selection principles (Q386851): Difference between revisions

A sequence \(\{f_n: n\in\omega\}\) of real-valued functions on a space \(X\) is said to quasi-normally converge to a function \(f:X\to {\mathbb R}\) if there exists a sequence \(\{\varepsilon_n: n\in\omega\}\) of non-negative reals converging to zero such that for any \(x\in X\) there exists \(m\in\omega\) such that \(|f_n(x)-f(x)|\leq \varepsilon_n\) for all \(n\geq m\). Suppose that \(\mathcal F\subset \mathbb R^X \) is a class of (not necessarily continuous) functions. Say that \(X\) is a \(wQN_{\mathcal F}\)-space (or has the property \(wQN_{\mathcal F}\)) if every sequence \(\{f_n: n\in \omega\}\subset \mathcal F\) converging to zero, has a subsequence \(\{f_{n_k}: k\in \omega\}\) that converges to zero quasi-normally. A space \(X\) has the sequence selection property with respect to \(\mathcal F\) or simply \(SSP_{\mathcal F}\) if for any sequence \(\{f_{mn}: n,m\in \omega\}\subset \mathcal F\) of sequences from \(\mathcal F\) such that \(\{f_{mn}: m\in \omega\}\) converges to zero for any \(n\in \omega\), there exist increasing sequences \(\{n_k: k\in \omega\}\) and \(\{m_k: k\in \omega\}\) such that the sequence \(\{f_{n_km_k}: k\in \omega\}\) converges to zero. The authors prove, among other things, that if \(X\) is a space and a family \(\mathcal F\subset \mathbb R^X\) is closed under multiplicaton by a positive real, then the property \(SSP_{\mathcal F}\) in \(X\) implies that \(wQN_{\mathcal F}\) holds in \(X\). If the family \(\mathcal F\) is closed under addition, taking the absolute value and \(\min\{c,f\}\in \mathcal F\) whenever \(f\in \mathcal F\) and \(c\in \mathbb R,\;c>0\), then the \(wQN_{\mathcal F}\) property of \(X\) implies that \(X\) has \(SSP_{\mathcal F}\).