The thesis of the article is not "lets conclude whatever I want"
The foundational assumption is that (a) historically people have just picked the values for the Drake Equation that "felt right", and this is wrong, biased, and stupid, and (b) we should not be surprised if the Drake parameters fall in any order of magnitude. So let's Monte Carlo in a log-scaled distribution (roughly). Thus, given what we actually do know, median civilizations per galaxy is 0.32, average is 27 million.
Interestingly, this suggests that optimistic Drake Equation parameters must be wrong, lest they create an empirically impossibly-populated galaxy.
This matters because with a battery, it's Very Hard(tm) to get the energy to discharge at a high enough rate to cause catastrophic effects, like for example why Mike Tyson can punch with ~1600 Joules (uncited, popularly referenced), about the same as a 5.56mmx45mm round fired from an M16/AR-pattern rifle. Because the Joules are distributed over space (bigger cross section) and time (slower impact), Mike Tyson's punch doesn't immediately shred your tissue and bones.
However, it's Very Easy(tm) to get a flywheel to have a tiny mechanical failure that causes the entire thing to release all its kinetic energy rapidly. Additionally, there's no conversion penalty with the flywheel. So the battery under catastrophe can and generally does shed its energy relatively slowly as rapidly-dissipating heat (see videos of cell phone batteries on fire), whereas a rotating wheel that just fell of its drive shaft is... markedly different.
Conclusion: Even if the flywheel only holds enough energy for a single charge, it would have to be buried far underground and still might create a crater in the gas station asphalt it sits underneath when (not if) mechanical failures happen.
Therefore, this is dumb and won't happen. Please help me understand where I'm wrong :)
The foundational assumption is that (a) historically people have just picked the values for the Drake Equation that "felt right", and this is wrong, biased, and stupid, and (b) we should not be surprised if the Drake parameters fall in any order of magnitude. So let's Monte Carlo in a log-scaled distribution (roughly). Thus, given what we actually do know, median civilizations per galaxy is 0.32, average is 27 million.
Interestingly, this suggests that optimistic Drake Equation parameters must be wrong, lest they create an empirically impossibly-populated galaxy.