Balls separated from their translates outside the stabilizer #
For a properly discontinuous action on a locally compact metric space, every point x lies in a
ball which meets none of its translates by group elements outside the stabilizer of x: a group
element moving that ball to meet itself already fixes x.
This separation is one half of the localization of an orbit space near a point with nontrivial
stabilizer. The other half, that the ball is itself invariant under the stabilizer, does not
follow from the hypotheses here: under ContinuousConstSMul alone an element fixing x need not
preserve a ball about x. Once that invariance is supplied — for an isometric action, say, whose
elements fixing x preserve every ball about x — the orbit space of the whole group near x
agrees with the orbit space of the single stabilizer, which for a properly discontinuous action
is a finite group.
Main results #
TauCeti.eventually_smul_eq_self_of_image_smul_ball_inter_nonempty: for every small enough ball about a point, a scalar moving a point of the ball into the ball fixes the centre.TauCeti.exists_ball_disjoint_smul_of_notMem_stabilizer: a small enough ball about a point is disjoint from each of its translates by a group element not fixing that point.
Small balls are separated from translates that move the centre. For a properly
discontinuous scalar action on a locally compact Hausdorff pseudo-metric space, for every small
enough r > 0, a scalar moving some point of the ball of radius r about x into that ball
fixes x.
A small enough ball meets no translate of itself by an element outside the stabilizer.
For a properly discontinuous action on a locally compact Hausdorff pseudo-metric space, some ball
about x is moved to meet itself only by the elements fixing x.