In an age before semiconductors how do you efficiently convert 100+V DC to a few volts for something like a valve filament?
Transformers are simple and anyone who can design an electric motor can design one almost by definition, but they ONLY work on ac.
Motors also do better on AC for many things (You cannot run an induction motor on DC).
Finally the biggies, circuit protection for HV DC is a bear (And still somewhat unsolved), arcs do not self quench when the supply is DC, which is why the big HV DC links tend to be point to point so the protection can be on the AC side.
Three phase alternators are more efficient then DC machines, and control of a network of ac machines is quite a bit simpler then the equivalent probelem for the DC case (Remember no transformers in a DC system also no line reactors) how do you move power from point a to point b without creating an excessive voltage at point a?
Now there actually were DC supplies to building back in the day, early lifts were powered by separately excited DC machines as the speed control was easier then it is in an AC machine (Remember, this is before semiconductors), and in fact some old office blocks in the US only had the last of the DC ripped out a few years ago.
Other uses were arc lamp supplies in Cinemas and plating plants, neither exactly a domestic situation.
On a technical level, given the technology of the time it was never even a debate, for all that edison tried.
Regards, Dan.