I read about it on one of the European sites. It was also done on at least one model with good reports. As I sort of said the idea is to achieve a pitch and flying speed trim function without the need to alter the camber of the airfoil. Especially since plank wing airfoils don't have a lot of camber to begin with thanks to the need for a forward high point and reflex. Combine that with a further reduction due to the up trim required for slow speed flight where the lift coefficient needs to be running in the higher range and you have a good recipe for high drag and doggy performance. The variable CG apparently gets around that.

Obviously the range of travel for the CG needs to be small. Something like 3 to 5% of the wing chord at the most. The last thing you want to do is run it back behind the 25% MAC point or even get too close to that point. Perhaps if it can move from about 18% to 24% at the extremes that would be a good point of experimenting and then limit it to a range that doesn't produce any flight issues? So a fairly small weight on a movable setup over a fairly small range of travel is all that is needed.

I should also note that this was all being done for a thermal flying plank design. Slopers just don't need to worry about that last couple of % efficiency unless they are racing.