Suggest "presence" or "coverage" instead of "geometric_opacity" about openpbr HOT 2 OPEN

Hi Andre,

I'm not sure I see the issue with terming the parameter "opacity". As we state in the spec:

So we do explicitly point out that the opacity/alpha is functioning as a presence weight, according to what the statistical mix operator is supposed to mean.

It's true that the transparency generated is not due to varying volumetric absorption in a finite width slab of material, rather it is due to microscopic holes in the thin surface. I think both situations are correctly interpreted as opacity/transparency though. (You could perhaps think of it as corresponding to the thin-wall limit of an absorbing slab, with a constant optical depth tau such that transmittance = exp(-tau) = 1 - opacity).

from openpbr.

Also, in the thin-walled case, we do say that the geometry_opacity functions explicitly as a fractional value, i.e. specifying 1 minus transmittance (so not just a {1, 0} valued signal for cutouts):

from openpbr.