Retractions of quasi-homogeneous bicompacta (Q1072147)

The author introduces a nice construction which associates with every compact space X a compact space K(X) such that: (a) X is a retract of K(X), (b) K(X) is decomposable-homogeneous i.e. can be represented as a union of two homogeneous subspaces, (c) K(X) has a homogeneous and completely metrizable dense subspace, (d) K(X) preserves various properties of X like being sequential, Fréchet-Urysohn, Corson or Eberlein compact. As the author announces, D. Motorov has shown that the subset cl\(\{\) (x,sin\(\frac{1}{x}):\) \(0<x\leq 1\}\) of the plane is not a retract of any homogeneous compact space. The problem, raised by A. V. Arkhangel'skij, whether every compact space is a continuous image of some homogeneous compact space remains open.