Investigation: Does the 1st De Villiers point exist in elliptic (spherical) geometry?

Consider the sketch below where G is the incentre of the spherical triangle ABC, and D, E and F are the respective incentres of ABG, BCG and CAG. Drag any of the vertices to investigate whether AE, BF and CD are always concurrent. What do you notice? Can you prove (or refute) your observation?

Please enable Java for an interactive construction (with Cinderella). Below is a static sketch if the Java applet doesn't work.

Elliptic De Villiers1 Test

Created by Michael de Villiers, 29 December 2010 with Cinderella. A free version of Cinderella 1.4 - the only dynamic geometry software that can also do elliptic (spherical) geometry and hyperbolic geometry - can be downloaded from Download Cinderella 1.4.