Take a scenario. Give it best-case figures. See if it works. If it doesn't, then it's impossible. If it misses the mark by just a bit, it's a remote possibility. If it works, try the worst case figures and the average case ones. Of course, you can also start with the worst case ones and show that it works.

I like to approach problems this way. Plus, one can have dreams, can't they?

Once you've shown it can theoretically work, it's time to consider the side effects: best/worst. With the above, someone now controls a multi-gigawat orbiting power source.

Upsides: well, it can power useful stuff just about anywhere on (two third of) earth when there's nothing else to do. That would be really useful to compensate renewables, or other grid fluctuations. One could even imagine making airplanes slower of faster depending on demand.

Downsides: well, somebody really does control that :)