Lightness of induced mappings (Q1818030)

If \(X\) is a metrizable continuum then \(2^X\) and \(C(X)\) denote its hyperspace of nonempty compact subsets and of nonempty subcontinua, respectively, endowed with the Vietoris topology. The identification \(x\mapsto \{x\}\) embeds \(X\) into \(2^X\) (and also in \(C(X)\)). It is therefore possible to think of \(X\) as a subspace of \(2^X\) (and hence also of \(C(X)\)). If \(f: X\to Y\) then the function \(A\mapsto f[A]\) is a continuous extension of \(f\) over both \(2^X\) and \(C(X)\). So, given a map \(f\) there are two other related hyperspace maps, denoted by \(2^f\) and \(C(f)\). It is natural to ask how they are related. In the present paper the authors discuss among other things the concept of lightness. They characterize lightness of \(2^f\) and derive some corollaries of this result. It is clear that if \(2^f\) or \(C(f)\) is light then so is \(f\). The converse need not be true. There is a light map \(f\) such that neither \(2^f\) nor \(C(f)\) are light. There is also a light map \(f\) such that \(C(f)\) is light, but \(2^f\) is not. Etc. There are also results for local homeomorphisms in this paper.