In November 1877, Thomas Edison pressed a needle against a rotating cylinder wrapped in tin foil and spoke the words "Mary had a little lamb." The foil deformed under the vibrating stylus as it traveled along the groove. When he traced the same path again, the foil's bumps pushed the needle back in the same pattern, and his voice came back out. He had built the first device in history that could record and play back sound.
The phonograph worked. It also had a serious problem: tin foil wore out within a few plays. The indentations were too shallow and too fragile to survive repeated use. Edison believed in the cylinder as the form factor and spent years refining it. He eventually moved to wax cylinders, which were more durable and produced cleaner audio. By the late 1880s, wax phonograph cylinders were commercially available, primarily marketed as office dictation machines. Entertainment was not the initial pitch.
The Disc That Shouldn't Have Won
Emile Berliner had a different idea. In 1887, he patented a device he called the gramophone, and the key innovation was the shape of the medium: a flat disc instead of a cylinder. Berliner's stylus moved laterally through a spiral groove on a horizontal platter rather than tracing helical grooves on a rotating cylinder. The disc could be stamped from a master in large quantities using a simple pressing process. The cylinder could not.
The cylinder sounded better. Edison's cylindrical format had a more uniform groove velocity — the stylus moved at the same speed relative to the groove regardless of where it was in the track. On a disc, the inner grooves pass under the stylus more slowly than the outer grooves, which distorts audio quality near the end of the record. This is a physical property of the disc format that has never been fully solved. Edison's engineers knew this and used it as an argument for cylinders throughout the format war that followed.
Berliner's discs won anyway, for reasons that had nothing to do with audio fidelity. A disc was flat. It could be stored in a sleeve. Sleeves could be stacked. A shelf of record sleeves was a fundamentally better storage solution than a shelf of cylindrical tubes. The disc was also cheaper to manufacture: a single metal master could stamp thousands of identical copies from shellac. The cylinder required more complex tooling and produced fewer usable units per master. Industrial economics favored the disc at every step.
Alexander Graham Bell and the Wax Improvement
The gap between Edison's 1877 tin foil phonograph and the commercial wax cylinder era is not a straight line. Chichester Bell — Alexander Graham Bell's cousin — and Charles Tainter worked at the Volta Laboratory in Washington in the early 1880s and developed a significant improvement: replacing tin foil with a wax-coated cardboard cylinder that was cut rather than embossed. The stylus carved the groove rather than deforming the surface, producing deeper, more durable grooves with better audio quality.
They called their device the graphophone. Edison, characteristically, refused to license or collaborate. He went back to his laboratory and produced his own improved wax cylinder phonograph in 1888. The result was a commercial market with two incompatible but similar competing cylinder formats, which confused buyers and slowed adoption. This is a recurring pattern in the history of recorded media: competing format wars that delay the market and are eventually resolved less by technical quality than by distribution and manufacturing economics.
Victor, Nipper, and Mass Culture
By the early 1900s, Berliner's flat disc had won the format debate. Eldridge Johnson founded the Victor Talking Machine Company in 1901 and licensed Berliner's patents. Victor's master stroke was not technical. It was the image of a white terrier named Nipper sitting in front of a gramophone horn, head cocked to one side, listening to a recording. The painting by Francis Barraud was titled "His Master's Voice." Victor licensed it and made Nipper the face of recorded sound.
The gramophone horn amplified the needle's vibrations through a purely acoustic mechanism — no electricity involved. The sound quality was thin by modern standards, but it was recognizably a human voice or musical performance, which was astonishing to people who had never heard recorded audio. Gramophones spread through upper-middle-class households in the first decade of the twentieth century, then moved steadily downmarket as prices fell and catalog depth increased.
Electric Recording Changes Everything
The 1920s brought the next discontinuity: electric recording. Before 1925, recordings were made by having performers crowd around a large acoustic horn that converted vibrations directly into groove movement through mechanical coupling. Loud instruments went in the back, quiet instruments went in the front, and the spatial arrangement of the ensemble was determined by acoustic balance rather than musical preference. Bass response was poor because the horn mechanism could not efficiently capture low frequencies.
Western Electric developed condenser microphones and electronic amplification in the early 1920s and licensed the technology to major record labels starting in 1925. The improvement was dramatic. Electric recording captured a wider frequency range, required less acoustic compromise in performance arrangements, and allowed engineers to balance levels in ways that weren't possible with purely mechanical systems. The records pressed from electrically recorded masters sounded noticeably better than those pressed from acoustic masters, and consumers noticed.
The LP and the Speed Wars
Shellac records played at 78 revolutions per minute, a speed that had become standard through industry convention in the early 1900s. The format held roughly four minutes of audio per side, which set the length expectation for popular songs and classical movements for decades. Symphony recordings required multiple shellac discs and required listeners to interrupt playback to change sides and discs at inconvenient intervals.
Columbia Records introduced the LP — long-playing microgroove record — in 1948, playing at 33⅓ RPM and holding up to 23 minutes per side. The slower speed and finer groove allowed much longer recordings on a single 12-inch vinyl disc. A symphony now fit on two sides of a single record. The format was also vinyl rather than shellac, which was quieter, less brittle, and lighter.
RCA Victor responded by introducing the 45 RPM single in 1949. The two formats targeted different markets: the LP was for classical music and album-oriented material; the 45 was for pop singles. Both used vinyl. Both eventually displaced shellac. Stereo recording arrived commercially in 1958, adding directional separation to both formats. The vinyl record industry built around these two speeds and two diameters — 12-inch LP and 7-inch single — and that basic configuration remained stable for thirty years.
Why the Disc Survived
The vinyl revival that began in the 2000s and accelerated through the 2010s is often described as nostalgia, and nostalgia is part of it. But the disc's remarkable commercial longevity from 1887 through the early 1990s, and its partial survival afterward, reflects something more durable than sentiment. The manufacturing economics that first gave the disc its advantage over the cylinder never entirely disappeared. A vinyl record is a physical artifact with specific sonic characteristics that digital formats replicate but do not fully reproduce. Whether those characteristics are features or bugs is a matter of preference, not objective measurement.
The disc won the format war against the cylinder for reasons of manufacturing and storage convenience. It then survived threats from eight-track tape, compact cassette, and CD primarily because its installed base — the record libraries people had accumulated over decades — created switching costs that new formats had to overcome. Music streaming finally succeeded in displacing vinyl as the dominant format by removing the friction of physical media entirely, not by being better in any acoustic sense.
Edison's cylinder sounded better. The disc won anyway. The lesson is not that quality doesn't matter. It's that distribution, manufacturing economics, and storage convenience often matter more.
Tracking something you're building? builds.anethoth.com is a public build ledger for software projects in progress — proof that a product is really being built.