[tedsanders]

As a physicist, I am similarly skeptical. It seems quite unlikely to me that wireless power will take off within our lifetimes. However, if it does happen, I suspect it will be because of one of the following hypotheses:

(1) Electronics continue to become more energy efficient, to the point that a tiny wireless power trickle is sufficient to run them

(2) Perhaps computers will be an enabling technology, both in design (with E&M modeling, algorithm-assisted antenna design) and in operations (computer controlled beam steering)

Anyone care to share any other "post-mortem" hypotheses?

[Filligree]

On the assumption that it happens? Well, you covered the more likely ones. Let me add one more.

3) Advances in molecular manufacturing (Drexler-style!) allows us to construct atomically perfect transmitters and receivers. This, somehow, increases efficiency to north of 90%. And increases range, too.

[rybosome]

I think that one of the NVRAM solutions in development like Intel and Micron's 3D xpoint might have lower power consumption per GB than DRAM given the non-volatility and thus the obviated need to continually refresh the value. So if we were to see DRAM replaced by NVRAM (or at least augmented, where all but the most critical data were shuffled from DRAM -> NVRAM each time the screen locked), that might be a pathway.

I tried to find some numbers on this, but this is outside of my area of expertise. Would love to hear somebody more knowledgable weigh in.

[hackuser]

What about it makes you skeptical? Stop holding out on us ... :)