This is my understanding. We use a surfactant to make the glass hydrophilic. We use the cleaning methods to make the surfactant effective for a longer period of time.
I have found research articles that claim that a clean glass surface can be either hydrophobic or hydrophilic depending on the chemical nature of the silicon oxide surface.
The glass surface's water wettability depends on the degree of hydroxylation of the surface silicon (hydroxylation of the silicon in the glass). Strong acids or strong bases (alkali) hydroxylate the surface. So it seems to be that one wants Si-OH groups on the surface of the glass for wettability. You can react Si-OH with TMS (trimethylsilyl reagent) to get Si-O-Si(CH3)3 groups and now you truly have a mask from Hell.
I no longer think that the problem is coming from the silicone skirt. I believe that it is simply the problem one has with keeping a clean surface clean. Clean isn't the full answer, though. That clean surface can be water wettable or water beading.
A clean but water-beading surface can be overcome by coating it with defog. The defog is physically covering the water-repellent (and defog-repellent) surface and the net result is that the mask doesn't fog. A better approach is to convert the glass surface to its hydrophilic (water-loving) state and then apply defog. The defog now is attracted to the glass as both the glass and the defog are hydrophilic.
Either way, you need some sort of defog to protect the clean glass surface. My preference is spit on a clean hydrophilic glass surface. It works just fine. In addition to spit, household ammonia is a great cleaning agent as it both cleans crud really well and it converts the glass surface to its hydrophilic state. Cleaning the glass, applying a thin coat of straight baby shampoo and then storing is the next thing that I will examine before I wrap this up. I believe that I got to where I wanted to be with respect to mask fogging.
I totally agree with the nanotechnology approach. -doing everything right. But that approach is outside the scope of this investigation...