When does \(C(K,X)\) contain a complemented copy of \(c_0(\Gamma )\) iff \(X\) does? (Q2305864): Difference between revisions

It was proved by \textit{E. Saab} and \textit{P. Saab} [Proc. Am. Math. Soc. 84, 44--46 (1982; Zbl 0484.46022)] that, if \(K\) is a compact Hausdorff space and \(X\) is any Banach space, then \(\ell_1\) is complemented in \(C(K, X)\) if and only if \(\ell_1\) is complemented in \(X\). It is natural to ask if the same happens for \(c_0\) instead of \(\ell_1\). In general, this is false for an infinite compact Hausdorff space \(K\), and it holds when \(K\) is finite. Following this question, the authors prove that, if \(\tau\) is an infinite cardinal such that \(\mathrm{cf}(\tau)> \mathrm{w}(K)\), then \(c_0(\tau)\) is complemented in \(C(K, X)\) if and only if \(c_0(\tau)\) is complemented in \(X\), where \(\mathrm{cf}(\tau)\) denotes the cofinality of \(\tau\) and \(\mathrm{w}(K)\) the weight of \(K\). Moreover, it is shown that the hypothesis \(\mathrm{cf}(\tau)> \mathrm{w}(K)\) is sharp.