Simple proof of a criterion for cones in \(\mathbb{R}^ 3\) (Q689812)

A subset \(S\) of \(\mathbb{R}^ d\) is a cone if and only if it is the union of closed halflines with a common endpoint. A point \(x\) in \(S\) is clearly \(R\)-visible via \(S\) from a point \(y\) if and only if there exists a neighbourhood \(N\) of \(x\) such that all closed halflines issuing from \(y\) through all points of \(N\cap S\) lie in \(S\). An alternative short proof of the following earlier criterion for cones is given: If every two local nonconvexity points of a closed connected and nonconvex set \(S\) in \(\mathbb{R}^ 3\) are clearly \(R\)-visible via \(S\) from a common point of \(S\), then \(S\) is a cone.