Weak compactness is not equivalent to the fixed point property in \(c\) (Q2352196)

A classical result of \textit{B. Maurey} [Semin. Anal. Fonct. 1980--1981, Exposé No. 8, 18 p. (1981; Zbl 0476.46023)] asserts that every nonexpansive (i.e., \(1\) -Lipschitz) mapping \(T:C\rightarrow C\) defined on a weakly compact, convex subset \(C\) of the Banach space \((c_{0},\left\| \cdot \right\| _{\infty })\) or \((c,\left\| \cdot \right\| _{\infty })\) has a fixed point. \textit{P. N. Dowling} et al. [Proc. Am. Math. Soc. 132, No. 6, 1659--1666 (2004; Zbl 1057.47064)] solved a longstanding open problem in metric fixed point theory by showing that weak compactness is also a necessary condition for a closed, bounded and convex (c.b.c., for short) subset of \(c_{0}\) to have the fixed point property for nonexpansive mappings. As for the space \(c\), it was proved in [\textit{P. N. Dowling} et al., Stud. Math. 154, No. 3, 277--293 (2003; Zbl 1037.47039)] a little less, namely, that for every non-weakly compact c.b.c.\ set \(C\), there exists a further c.b.c.\ subset \(K\) and a nonexpansive mapping \( T:K\rightarrow K\) without fixed points. In this interesting paper, the authors give examples of c.b.c.\ subsets of \(c\) with the fixed point property for nonexpansive mappings. Specifically, they prove that a large family of sets, including \[ W=\left\{ (y_{n})\in c:1\geq y_{1}\geq y_{2}\geq y_{3}\geq\dots\geq 0\right\} \] and \[ Z=\left\{ (y_{n})\in c:1=y_{1}\geq y_{2}\geq y_{3}\geq\dots\geq 0\right\}, \] consists of hyperconvex sets and, consequently, every such set has the fixed point property. These are the first examples of non-weakly compact c.b.c.\ subsets of a space isomorphic to \(c_{0}\) with this property. A~related open question is whether there exists an isomorphic copy of \(c_{0}\) such that its every closed, bounded and convex subset has the fixed point property for nonexpansive mappings.