The incandescent lamp did not appear out of nothing in 1879. Humphry Davy had run a platinum wire and an arc lamp in 1802; from the 1840s onward Warren de la Rue, Frederick de Moleyns, James Bowman Lindsay and others had looked for ways to make light from a wire heated in a vacuum. The problem was not invention but economy and endurance: a high enough vacuum, a thin yet resistant filament, and a lifetime well beyond a few hours. In 1878 in England the chemist Joseph Swan demonstrated a lamp with a carbonised paper filament before the Newcastle Chemical Society; by 1879 maturity was close on both sides of the Atlantic.

At the Menlo Park laboratory in New Jersey, Edison decided that the question was not a single lamp but an entire system that could function around it: generator, meter, fuse, switch, socket, wiring from plug to filament. On 21 October 1879 he ran a lamp with a carbonised cotton thread for about 13.5 hours; on the night of 31 December 1879 he opened Menlo Park to the public and lit dozens of bulbs at once. Thousands of visitors came by special train. By 1880 a bamboo filament pushed life into the hundreds of hours. Swan and Edison merged in Britain in 1883 as the "Edison & Swan United Electric Light Company" — as so often, the commercial result of parallel invention was partnership.

On 4 September 1882 the Pearl Street Station opened in Manhattan: six "Jumbo" generators, an initial 85 customers, direct current at 110 volts. It was the first moment that generation, distribution, and consumption worked together as a system inside a city. But DC could not be economically carried over long distances. In the same years Nikola Tesla's polyphase alternating-current motors and George Westinghouse's transformer-based distribution offered an alternative. Through the late 1880s and 1890s the Edison camp ran a campaign portraying AC as dangerous (the AC-powered electric chair was part of it); but the lighting of the 1893 Chicago World's Fair by Westinghouse–Tesla and, in 1895, the Niagara Falls hydroelectric station sending AC current to Buffalo, decisively closed the contest.

Electric lighting was not a layer of comfort on top of existing life; it redefined the length of the day. Factory shifts extended into the night, street lighting reduced crime and accident rates, theatres, cinemas, and shop windows produced a night economy. The current in the same wire soon powered the motor, the oven, the refrigerator, radio, the lift, and ultimately the computer. Global electricity consumption today reaches tens of thousands of terawatt-hours a year; the birth night of that infrastructure, both its usefulness and its limit, was a single Menlo Park demonstration tied to a carbonised thread. The parallel Swan–Edison and parallel DC–AC contests show once more that technology can rarely be reduced to a single hero or a single design.