Manufactuers do have an interests to have the best interests of the consumer in mind, where safety is concerned, but not necessarily where cost, sizing flexibility, etc., are concerned.
OneTinSloth is correct that some innovations will fade because consumers don't like them, but why and why not consumers like them depends a lot on the marketing. In the case of Biopace, Shimano is commonly understood to have made a major marketing error in how they pitched Biopace, accidentally/implicitly suggesting that it wasn't a good design for "serious" riders, and after that became "common knowledge" no one wanted them. I run them on my fixed and on the road bike that I keep in Michigan at my parents' place, where the terrain is relatively flat.

I agree that threadless is better all-around, with the possible exception of road bikes that won't see crazy torquing in a sprint.
I don't think that threadless is a bad idea that manufacturers are trying to cram down people's throats. It is, however, spec'd in places where it doesn't make the most sense, or spec'd in a way that doesn't make sense (without long steerer tubes that allow for adjustability). Which isn't a critique of the threadless system, though.

I do think that we'll continue to see quill stems

Also, a very good analysis of the pros and cons of fewer sizes but greater number of models. Thanks!

One very interesting thing is that the "sport road" or "comfort road" market is something that major manufactuers didn't really get into with much heart until Rivendell had really pushed the category. Rivendell makes expensive bikes, but their strong critique that the average road bike - which will never be raced - should be able to take wider tires, fenders, a rack, etc... that philosophy got picked up by enough people to make the manufacturers realize that it was a viable marketing strategy for lots of cyclists. Surly and Soma have piggybacked in here, too.