Overview
Everything that reaches a screen or a speaker travels the same chain: something captures a pattern of light or air pressure, something encodes it, something carries it across distance, and something reconstructs it at the far end. That chain took roughly a century and a half to assemble, and almost none of it was built by the people whose names are attached to it in popular accounts. This guide follows the chain link by link — what problem each stage solved, what it made possible downstream, and why the question "who invented this?" almost always has a worse answer than the question "what problem were they actually solving?" It is the argument; the dates, in order, are on the broadcast technology timeline.
What You Need
- Where to start: the timeline for chronology, then any single link below
- No technical background — every term is explained where it first appears
- Companion guides: the histories of film, television, radio, recorded sound, colour and codecs
Steps
Capture came first, and it came twice
Two independent problems were solved within twenty years of each other, and neither inventor was thinking about the other. In 1878 Eadweard Muybridge set a battery of trip-wired cameras beside a track and froze the successive phases of a galloping horse, proving that movement could be decomposed into still frames and replayed. In 1860, Edouard-Leon Scott de Martinville had already inscribed airborne sound as a visible trace on lampblacked paper — the phonautograph wrote sound down seventeen years before anyone played it back. That gap matters: recording and reproduction are different inventions, and conflating them is the single most common error in this history. Everything downstream depends on a substrate to record onto, which is why George Eastman putting emulsion on flexible transparent roll film is a more load-bearing event than most of the famous ones.
The moving image becomes a machine
"Who invented cinema" is a definitional question rather than a factual one, and the honest answer depends entirely on what you count. Louis Le Prince shot moving-picture sequences in Leeds in 1888 — the earliest that survive — and then vanished from a train in 1890 before he could demonstrate them publicly. William Friese-Greene patented a chronophotographic camera in 1889; the myth built around him was demolished by Brian Coe in the 1950s and 60s, and is now being partly rebuilt, with Peter Domankiewicz arguing in 2024 that Coe himself over-corrected. At Edison's laboratory, William Kennedy Laurie Dickson settled 35mm film at a 1.33:1 frame — a decision still governing frame geometry more than a century later — under a credit policy that assigned the work to Edison. The Lumieres' 1895 screening at the Grand Cafe was not the first film ever projected, but it was the first commercially exhibited programme to a paying public, which is why it stuck.
Waves in the air, and the most misreported ruling in the field
No single person invented radio. Heinrich Hertz demonstrated electromagnetic waves in the laboratory; Branly, Lodge, Popov and Jagadish Chandra Bose developed detection; Tesla and Lodge established tuned circuits; Guglielmo Marconi built the first working long-distance wireless service and the first commercial business around it. Priority claims have been advanced for all of them, frequently along national lines. The 1943 US Supreme Court decision is routinely cited as settling this, and it did not: the Court invalidated the broad claims of Marconi's four-circuit tuning patent as anticipated by Stone, Lodge and Tesla, and expressly declined to rule on who invented radio — stating that Marconi's original patent "is not here in question." Courts rule on patents, not on invention. That distinction recurs so often in this history that it is worth holding onto as a rule.
Broadcasting is a different invention from radio
Transmitting a signal and running a service are separate achievements, and the second is usually credited to the first. Reginald Fessenden established continuous-wave voice modulation, though the famous 1906 Christmas Eve broadcast rests on a single source: his own account, written in 1932. Ernst Alexanderson built the high-frequency alternator that could generate a continuous wave powerful enough to matter. But it was Charles Herrold, running scheduled voice-and-music transmissions on an announced timetable to a known audience, who turned apparatus into broadcasting. Edwin Armstrong then supplied three of the foundational circuits — regeneration, the superheterodyne receiver, and wideband FM — and spent much of his life in patent litigation over them.
Seeing at a distance
Television has the densest priority dispute in media technology, and two of its most important figures built nothing. Paul Nipkow patented the spiral scanning disc in 1884 and almost certainly never built one; the patent lapsed unexploited, yet the disc became the standard scanning element of every mechanical television system that followed. In 1908 A. A. Campbell-Swinton published the full architecture of electronic television in Nature and likewise never built it. John Logie Baird gave the first demonstration of live moving images with tonal gradation, on a mechanical line that turned out to be a dead end. On 25 December 1926 Kenjiro Takayanagi displayed a television image on a cathode ray tube in Hamamatsu — months before Farnsworth, and almost unknown in Anglophone accounts. Philo Farnsworth built the image dissector and won priority at the US Patent Office in 1935; RCA appealed and lost, then licensed. And yet the tube that actually scaled was Vladimir Zworykin's charge-storage iconoscope, built on a principle Kalman Tihanyi had patented and RCA had bought. Farnsworth won the case. Zworykin's approach won the industry.
Standards are where broadcasting actually gets decided
A demonstration is not a service. Fritz Schroter patented interlaced scanning in 1930 — sending odd and even lines in alternate passes to halve the bandwidth — and it is still present in broadcast standards a century later. Isaac Shoenberg led the EMI team that chose 405 lines interlaced, the first high-definition public television service. Hidetsugu Yagi and Shintaro Uda produced the directional antenna that made VHF reception practical, and the order of those two names is itself a small case study in how credit travels: Yagi published in English, Uda did the experimental work. Colour then had to be added without breaking the millions of monochrome receivers already sold — a constraint that produced PAL, SECAM and NTSC as three different compromises with the same problem.
Colour, and the hardest credit question in the corpus
Louis Ducos du Hauron set out subtractive three-colour photography in 1869; almost everything afterwards is an implementation of it. Rudolf Fischer patented chromogenic development — dye formed in the emulsion by colour couplers — which is how essentially all colour film works. At Technicolor, Leonard Troland filed the core imbibition-printing patents while Herbert Kalmus drove the company from two-colour additive to three-strip. Natalie Kalmus was credited as colour director on almost every Technicolor feature for two decades. Whether that credit reflected her expertise or a contractual arrangement arising from her 1922 divorce settlement is genuinely contested; both readings are currently in print, the Color Advisory Service was a staffed department, and this site states both and stops there.
Recording changes what broadcasting is
Magnetic tape was developed in Germany — Fritz Pfleumer coated iron oxide onto paper strip, and AC bias, discovered around 1940, lifted the result from dictation quality to broadcast quality. Jack Mullin rebuilt captured Magnetophons with new electronics and demonstrated broadcast-quality tape in America, which is technology transfer functioning as invention. Then in 1956 the Ampex team solved rotary transverse scanning, writing video across the tape rather than along it, and broadcasting stopped having to be live. The team was six people; Charles Ginsburg led it and Fred Pfost hand-built the rotating heads. One of the six was Ray Dolby, whose later noise-reduction work made his name a brand — which is why "Ray Dolby invented videotape" is one of the most persistent errors in this field.
The digital chain, and the people standards erase
Harry Nyquist established the sampling condition, Claude Shannon founded information theory, and David Huffman derived optimal prefix-free coding as a graduate term paper to avoid sitting a final exam. In 1972 Nasir Ahmed conceived the discrete cosine transform, the mathematical core of JPEG, MPEG and effectively every codec since; the 1974 paper had three authors, and of the third, T. Natarajan, no biographical record appears to survive anywhere. That absence is not an oversight in this guide — it is the clearest illustration of the pattern. Standards-body authorship means the document is the invention, so there is no inventor to credit: the MP3 contributors, the JPEG committee and Ahmed himself sit inside standards that name no individual.
Why the credit lands where it does
Four mechanisms explain most of the misattribution in this history, and they are worth keeping separate because each calls for different language. Assignment: employment contracts hand inventions to the employer, so the work becomes "an RCA invention" — this took Dickson, Blumlein and the EMI team. Explicit anonymity: BBC Radiophonic Workshop staff were contractually uncredited, so Delia Derbyshire could not be named for realising the Doctor Who theme even when the composer wanted her to be, and Daphne Oram's film work went out anonymously on the same basis. Standards-body authorship, as above. And fame back-propagation, where someone famous for later work absorbs credit for earlier team effort. A fifth is simply linguistic: Takayanagi, Tihanyi, Popov and Bose published in Japanese, Hungarian, Russian and colonial-Indian contexts, and Anglophone histories largely did not follow.
Pro Tips
- Five different years can all be called "the invention" — conceived, patent filed, patent granted, first public service, first commercial sale. Zworykin filed in 1923 and the patent issued in 1938. Always ask which one a date refers to.
- Invented, patented, demonstrated and commercialised are four different achievements. Nipkow patented without building; Campbell-Swinton published without building; Marconi commercialised without originating the components.
- A court ruling is a legal outcome, not a historical finding. Both the 1943 Marconi decision and the 1934 de Forest decision are routinely misreported as settling who invented what.
- When an invention emerges from a large institution, ask who held the patents rather than who did the work — they are frequently different people.
- Independent simultaneous invention is the norm here, not the exception. Treat a lone-genius story as a claim requiring evidence.
Knowledge Base
What You'll Learn
The chain is easier to hold onto as a set of problems than as a set of names. Each link below existed because the previous one created something that could not yet be moved, stored or reconstructed. Two companion pages carry the people and the chronology: the A–Z of broadcast technology pioneers, filterable by strand and by gender, and the timeline of broadcast technology.
The chain, in one pass
Capture turns light or air pressure into a recorded pattern — cameras, microphones, film stock, sensors. Encoding puts that pattern into a form that survives transport: modulation, scanning, sampling, compression. Transmission moves it — wire, spark, carrier wave, satellite, fibre, packet. Reconstruction turns it back into something a person can perceive: a projector, a loudspeaker, a cathode ray tube, a liquid crystal panel. Every figure in this discipline worked on one or two of those links, and the interesting arguments are almost always about the joins between them.
Why the same thing gets invented repeatedly
Independent simultaneous invention is normal in this history. Electronic television was described by Campbell-Swinton in 1908, demonstrated in part by Takayanagi in 1926 and by Farnsworth in 1927, and made manufacturable by Zworykin in 1931 — four people, four countries, one idea whose time had come because the enabling components had arrived. AC bias for magnetic tape was found in Germany and independently by Marvin Camras in the United States. When several people solve a problem within a few years of each other, the useful question is what changed to make it solvable, not which of them was first.
The canon and the overlooked
This site profiles both, and marks which is which. The canonical figures are not wrongly famous — Marconi really did build the first working long-distance service, and Farnsworth really did win his interference. But alongside them sit people whose contribution is equally documented and who are routinely omitted: Tihanyi, whose charge-storage patent RCA bought; Takayanagi, first to a working electronic display; Bose, decades early on millimetre waves; Derbyshire and Oram, contractually forbidden their own credit; Natarajan, of whom nothing survives at all. Presenting the omission as part of the history, rather than quietly fixing it, is the point.
FAQ
Q: Who invented television?
A: No one person. Nipkow patented the scanning disc in 1884 without building it; Campbell-Swinton described the electronic architecture in 1908 without building it; Baird gave the first demonstration of live moving images with tonal gradation on a mechanical system; Takayanagi displayed the first electronic television image on a cathode ray tube on 25 December 1926; Farnsworth built the first working electronic camera tube and was awarded priority in 1935; Zworykin developed the charge-storage tube that the industry actually adopted. Which of those counts as "inventing television" depends entirely on the definition you choose.
Q: Did the Supreme Court rule that Tesla invented radio?
A: No. In 1943 the Court invalidated the broad claims of Marconi's four-circuit tuning patent as anticipated by earlier work of Stone, Lodge and Tesla. It expressly declined to rule on who invented radio, noting that Marconi's original patent "is not here in question." The claim that the ruling restored Tesla's priority is a widely repeated misreading, and Tesla had already died months before the decision.
Q: Why do so many of these people not get credit?
A: Four repeating mechanisms: employment contracts assigning inventions to the employer; explicit anonymity policies, of which the BBC Radiophonic Workshop is the clearest case; standards-body authorship, where the document is the invention and no individual is named; and fame back-propagation, where someone known for later work absorbs credit for earlier team effort. Language adds a fifth — work published in Japanese, Hungarian or Russian largely did not reach Anglophone histories.
Q: What is the difference between recording and broadcasting?
A: Recording stores a signal so it can be reproduced later; broadcasting sends it to many receivers at once. They developed separately and met late. Until Ampex solved rotary transverse scanning in 1956, television was overwhelmingly live, because there was no practical way to record a video signal at broadcast quality.
Q: Is this history settled?
A: Not entirely. The Friese-Greene case is actively contested in current scholarship, with a 2024 argument that the mid-century debunking itself over-corrected. The basis of Natalie Kalmus's Technicolor credit is genuinely disputed. Where competent people disagree, this site states the disagreement rather than picking a side.