Sticky Light - non imaging photodetector

I am going to have a hard time not making one of these.

I surmise the dot is always moving in a tiny circle around a central point. The instantaneous brightness combined with the phase of the dot gives it a local map of reflectivity. This is combined with synthetic inertia and sometimes external forces (a gravity like preference in some demos) to produce the movement.

The circle rate is quite high. It looks like at least 4 times the frame rate of the youtube video, maybe much more. I suppose that means some sort of coil arrangement for mirror aiming.

I'm sorry. I don't get it. Why is this same / similar technique posted for at least the third time on Hacker News? Has it improved in the last two weeks or are the people posting news not reading Hacker News themselves?

I can understand bringing it to attention (again), but somehow it gets annoying too. I don't like to read the same ol' same ol' news, especially if it is on the same website. Not to be a troll or anything, but could the people who post first look if something similar is posted?

Any ideas on what a "non-imaging photodetector" actually is? There's very little in Google that's not connected to this video, and all of those references simply use the term without going into it.