First off I found the sound track more of a distraction than anything said. Almost quit several times, and wish I had as I did not see anything really substantive.
The cute argument about the nose ignores cultural standards of beauty as a measure of reproductive desire ... with different standards in different cultures showing different preferences for the shape and size of noses.
The argument that female orgasms have no selective advantage ignores the willingness of participants to repeat experiences that are pleasant and avoid experiences that are not. It may be that it "came along for the ride" with selection for male orgasm, but once in the mix it does have sexual selection advantages: more breeding → more offspring.
And we know from observing Bonobos behavior that this isn't just a human reaction.
To my mind selection operates to trend towards stasis in any stable ecology for species, and it is only when you have changing ecologies, or emigration to new ecologies, that things change.
In comparison, drift operates to trend towards random distribution of traits and potential loss of traits that are/were adaptive, so it trends away from stasis. This becomes more apparent in separated populations, especially when one population is small (founder effect).
In this way drift would be more important to the development of (non-deleterious) variations within a population, and hence maybe more important to speciation than selection.
If a trait persists in a population from generation to generation it is in effect selected, regardless of relative adaptive benefit/cost.
Non-beneficial traits/genes get selected when they are bundled with ones that have high beneficial/selective value: is this selection or drift or a combination?
Or is it just perspective? Species and individuals survive and breed because they are capable of surviving and breeding, not because they are becoming better "adapted" or because they are becoming more "fit" ... the "goal" is survival are reproduction not fitness\adaptation.