The Internet and Computing
From a mechanical engine that never ran to a network that never sleeps
Computing did not begin with a chip or a screen. It began with a Victorian mathematician trying to remove human error from printed tables, and it grew through vacuum tubes, transistors, and silicon into the machines and networks that now sit in everyone's pocket. This timeline follows the hardware, the software, and the wires: the mechanical engines, the codebreaking computers of the Second World War, the transistor and the chip, the first networks, the personal computer, and the web that connected all of it. Artificial intelligence and video games have their own timelines here; this one covers the machines and the wires that made both possible.
Source healthshow
- Primary source11 events
- Peer-reviewed3 events
- Reputable source21 events
- General source1 event
62 of 62 checked source links loaded and matched the event’s key terms. This confirms the source is live and on-topic, not that it proves the claim, which is what reading it is for.
11 sources couldn’t be checked automatically, often a legitimate source that blocks automated readers. These are left out of the figure above rather than counted against it, and are worth reading directly.
No reader corrections reviewed yet. See something wrong? Every event page has a way to say so.
Every event names its strongest source; grades come from the domain and declared type. Last reviewed . See how trust works and the source registry.
Events
- 1834Well documented
Reputable source · 2 sourceswhy?
Best source: A Brief History
The domain "computerhistory.org" is on our Reputable source registry.Babbage conceives the Analytical Engine
Charles Babbage had already spent years on the Difference Engine, a mechanical calculator meant to print error-free mathematical tables. In 1834, with that project stalled, Babbage conceived a far more ambitious machine, later called the Analytical Engine. It borrowed the punched-card system from the Jacquard loom, which used cards to control weaving patterns, and applied the same idea to computation. The design split the machine into a 'Store,' where numbers and intermediate results were held, and a separate 'Mill,' where arithmetic was carried out, the same store-and-process split every computer still uses. It could perform all four arithmetic functions, and Babbage's notes describe conditional branching and looping (iteration) even though he had no name for these ideas yet.
Why it matters: No Analytical Engine was ever completed in Babbage's lifetime, but the design laid out the logical architecture, a separate memory and processor, conditional operations, and a form of programmability, that every later computer would rediscover independently. It set the template Ada Lovelace would use to write the first algorithm nine years later.
How we know: The Computer History Museum's Babbage Engine project maintains detailed historical pages drawn from Babbage's surviving notebooks, drawings, and the reconstructed Difference Engine No. 2, built from his original plans and now on public display.
Designer: Charles Babbage · Conceived: 1834 · Key components: The 'Store' (memory) and the 'Mill' (arithmetic unit) · Input method: Punched cards, adapted from the Jacquard loom
- August 1843Debated
Reputable source · 2 sourceswhy?
Best source: Ada Lovelace
The domain "computerhistory.org" is on our Reputable source registry.Ada Lovelace publishes the first algorithm
Ada Lovelace translated an Italian engineer's French-language memoir on Babbage's Analytical Engine and, at Babbage's suggestion, added her own notes. Her notes ran three times longer than the original memoir and were published in 1843. The last of them, known as Note G, laid out a step-by-step sequence of operations for the Analytical Engine to compute Bernoulli numbers, a set of numbers used in mathematical series. It is the first published description of a stepwise procedure written for a machine to execute. Lovelace also went beyond the arithmetic Babbage had in mind, writing that the Engine 'might act upon other things besides number' and could, in principle, compose music.
Why it matters: Note G is why Lovelace is often called the first programmer, a claim historians still debate given how closely she worked with Babbage on the ideas. What is not disputed is that she was the first to publish a concrete algorithm for a general-purpose machine and the first to describe computation as something that could go beyond arithmetic into symbols of any kind, an idea that would not become mainstream again for over a century.
How we know: The Computer History Museum's Babbage Engine project describes the Notes and Note G from its own historical record, and notes explicitly that 'biographers debate the extent and originality of Ada's contribution,' a genuine scholarly dispute rather than a settled fact.
Author: Ada Lovelace (Augusta Ada King, Countess of Lovelace) · Published: August 1843, in Taylor's Scientific Memoirs · Key contribution: Note G: an algorithm to compute Bernoulli numbers · Debated claim: Whether the algorithm was originally hers or developed jointly with Babbage
- 1890Well documented
Reputable source · 2 sourceswhy?
Best source: Making Sense of the Census: Hollerith's Punched Card Solution
The domain "computerhistory.org" is on our Reputable source registry.Hollerith's punched-card tabulator wins the census contract
The 1880 US census had taken more than eight years to fully tabulate by hand, and the Census Bureau ran a competition to find a faster method for 1890. Herman Hollerith, a former Bureau employee, won it with a machine that read data punched as holes in paper cards. Operators pushed spring-loaded pins through the card; where a hole let a pin pass through, it dipped into a small cup of mercury below and completed an electrical circuit, which advanced a dial on the tabulator counting that category. An average operator could process about 7,000 cards a day, roughly ten times faster than tallying by hand.
Why it matters: Hollerith's company later merged with others to form the Computing-Tabulating-Recording Company in 1911, which was renamed International Business Machines in 1924. Punched cards built on his design remained the standard way to feed data and programs into computers for the next seventy years, well into the era of mainframes.
How we know: The Computer History Museum's history of the punched card describes Hollerith's design, the mercury-cup circuit, and the processing speed from its own collection notes and historical records.
Inventor: Herman Hollerith · Year: 1890 US census · Mechanism: Punched cards read via pins completing a mercury-cup circuit · Later company: CTR (1911), renamed IBM in 1924
- 30 November 1936Well documented
Reputable source · 2 sourceswhy?
Best source: Turing Machines
The domain "plato.stanford.edu" is on our Reputable source registry.Turing describes the universal machine
Alan Turing published 'On Computable Numbers, with an Application to the Entscheidungsproblem' in the Proceedings of the London Mathematical Society, in installments dated 30 November and 23 December 1936. He was attacking a question posed by mathematician David Hilbert: whether a single mechanical procedure could decide, for any statement in formal logic, if it was provable. Turing answered by inventing an idealized device, an 'automatic machine' that reads and writes symbols on an infinite tape one cell at a time, following a fixed table of rules. He then described a 'universal machine' that could read a description of any other such machine off the tape and imitate it exactly, showing that one general-purpose device could in principle run any computation.
Why it matters: The paper proved that some mathematical problems have no mechanical solution at all, the halting problem, while simultaneously describing, in the universal machine, the theoretical blueprint for a stored-program computer that runs any program fed to it. Every general-purpose computer built since is, in the abstract sense Turing described, a physical approximation of that single idea.
How we know: The Stanford Encyclopedia of Philosophy's entry on Turing Machines and the IEEE's Engineering and Technology History Wiki both describe the paper's content and the 1936 date directly from the historical and scholarly record; the paper itself was named 'Turing machines' only afterward, by logician Alonzo Church's 1937 review.
Author: Alan Turing · Published: 30 November and 23 December 1936, Proceedings of the London Mathematical Society · Key idea: The universal machine: one machine that can imitate any other computing machine · Term coined by: Alonzo Church, in a 1937 review of the paper
Sources - 5 February 1944Well documented
Reputable source · 2 sourceswhy?
Best source: 1944
The domain "computerhistory.org" is on our Reputable source registry.Colossus breaks the Lorenz cipher at Bletchley Park
British telephone engineer Tommy Flowers spent eleven months designing and building Colossus at the Post Office Research Station in Dollis Hill, London, to help break the German Lorenz cipher (codenamed 'Tunny' by the British), used for high-level communications between Hitler and his generals. Colossus was delivered to Bletchley Park in late 1943 and early 1944 and was working by early February 1944. It used about 2,500 vacuum tubes to test possible cipher-wheel settings electronically, far faster than the manual methods that had taken four to six weeks per message. By the end of the war, ten Colossus machines and the 550 people operating them had decrypted 63 million characters of high-grade German communications.
Why it matters: Colossus was the first large-scale electronic digital computer, predating ENIAC by more than a year, though its existence stayed classified for decades and so it played little direct role in later computer design. The intelligence it produced is widely credited with shortening the war, including by revealing German troop positions ahead of the D-Day landings.
How we know: The National Museum of Computing, which holds a working Colossus rebuild at Bletchley Park, documents Flowers's design process, the delivery and operational dates, and the valve count from its own historical and technical records.
Designer: Tommy Flowers · Working by: Early February 1944 · Vacuum tubes: About 2,500 per machine · Target: The German Lorenz ('Tunny') cipher
Sources - 30 June 1945Debated
Primary source · 2 sourceswhy?
Best source: First Draft of a Report on the EDVAC
Cited as a "primary" source (no stronger domain match).Von Neumann's EDVAC report describes the stored-program computer
While ENIAC was still being built, mathematician John von Neumann joined the Moore School team and, in an incomplete 101-page document circulated on 30 June 1945, wrote the first widely-read description of a stored-program computer: a machine holding both its instructions and its data in one memory, encoded in the same binary form, so it could read, and even modify, its own program while running. This is the design later called the 'von Neumann architecture,' still the basic layout of most computers today. The report drew on discussions von Neumann had with Eckert, Mauchly, and Herman Goldstine, but named only von Neumann as author when it circulated.
Why it matters: The stored-program idea solved ENIAC's central weakness, that reprogramming meant physically rewiring the machine, by letting programs live in memory and be swapped as easily as data. Because von Neumann's report spread widely under his name alone, it also created a lasting credit dispute: Eckert and Mauchly maintained they had worked out the concept independently at the Moore School well before his report, and historians have since acknowledged their role even as the term 'von Neumann architecture' remains standard.
How we know: The Computer History Museum's own blog on 'the neverending quest for firsts' lays out the credit dispute explicitly, describing how the report's wide, uncredited circulation obscured Eckert and Mauchly's contributions for years before historians corrected the record.
Report author: John von Neumann · Circulated: 30 June 1945, by Herman Goldstine · Core idea: Instructions and data stored together in one binary memory · Credit dispute: Eckert and Mauchly say the idea was developed jointly at the Moore School before von Neumann's report
- 15 February 1946 (public dedication)Well documented
Peer-reviewed · 2 sourceswhy?
Best source: ENIAC, the First General-Purpose Digital Computer, Turns 80
The domain "spectrum.ieee.org" is on our Peer-reviewed registry.ENIAC runs at the University of Pennsylvania
J. Presper Eckert and John Mauchly led a team at the University of Pennsylvania's Moore School of Electrical Engineering building ENIAC, the Electronic Numerical Integrator and Computer, starting in 1943 to calculate artillery firing tables for the Army. It first ran in late 1945 and was formally dedicated on 15 February 1946. Using more than 17,000 vacuum tubes as electronic switches, ENIAC could execute conditional branches, letting it change its calculation path based on prior results, an 'if this, then that' capability that made it reprogrammable rather than fixed to one task. Reprogramming meant physically rewiring plugboards and setting switches, which could take days, but the machine itself performed calculations thousands of times faster than any electromechanical predecessor.
Why it matters: ENIAC is generally treated as the first general-purpose electronic digital computer, the point where computing left electromechanical relays behind for good. Its slow, physical reprogramming process is exactly the problem the stored-program concept, developed in parallel by the same Moore School team, was built to solve.
How we know: Penn Engineering's own historical account of ENIAC, published for its 80th anniversary, describes the build, the vacuum tube count, and the conditional-branching capability directly from the university's institutional record.
Built by: J. Presper Eckert and John Mauchly, University of Pennsylvania · Dedicated: 15 February 1946 · Vacuum tubes: More than 17,000 · Original purpose: Artillery firing table calculations for the U.S. Army
- 16 December 1947Well documented
Reputable source · 2 sourceswhy?
Best source: 1947: Invention of the Point-Contact Transistor
The domain "computerhistory.org" is on our Reputable source registry.Bell Labs demonstrates the point-contact transistor
At Bell Telephone Laboratories in Murray Hill, New Jersey, physicists John Bardeen and Walter Brattain, working under William Shockley, had spent weeks trying to build a solid-state amplifier out of semiconductor material instead of a fragile, power-hungry vacuum tube. On 16 December 1947, their work culminated in the first successful semiconductor amplifier: two closely spaced gold contacts, held in place by a plastic wedge, touching the surface of a small slab of high-purity germanium. Voltage applied to one contact modulated the current flowing through the other, amplifying the input signal up to 100 times. They demonstrated the device to Bell Labs leadership on 23 December 1947.
Why it matters: The transistor did the job of a vacuum tube using a fraction of the power, generating far less heat, and taking up a fraction of the space, without a filament that burned out. Every integrated circuit and microprocessor that followed is, at bottom, millions or billions of transistors descended from this first germanium point-contact device.
How we know: The Computer History Museum's Silicon Engine project describes the December 1947 date, the germanium and gold-contact construction, and the roughly 100-times amplification directly from its own historical account, and Bardeen, Brattain, and Shockley shared the 1956 Nobel Prize in Physics for the discovery.
Inventors: John Bardeen and Walter Brattain, under William Shockley · Date: 16 December 1947 (first success); demonstrated 23 December 1947 · Material: High-purity germanium with two gold point contacts · Recognition: 1956 Nobel Prize in Physics
- 14 June 1951Well documented
Reputable source · 2 sourceswhy?
Best source: UNIVAC I
The domain "census.gov" is on our Reputable source registry.UNIVAC I ships to the Census Bureau
Eckert and Mauchly, the engineers behind ENIAC, began building UNIVAC (Universal Automatic Computer) in 1948 through their own company, later acquired by Remington Rand. On 14 June 1951, officials from the Census Bureau attended a dedication ceremony at the Eckert-Mauchly Laboratory in Philadelphia. UNIVAC read data from magnetic tape rather than only punched cards, used vacuum tubes for processing, and could print results or store them back to tape. It was the first computer built as a commercial product for sale to any customer rather than a one-off machine built for a single research or military purpose.
Why it matters: On election night in 1952, a UNIVAC correctly forecast Dwight Eisenhower's landslide win from a sliver of early returns, an event CBS was initially too skeptical to broadcast; it introduced the American public to the idea that a machine, not just a person, could process data and make predictions. UNIVAC's commercial sale model, rather than one-off government contracts, established the computer industry as a business.
How we know: The U.S. Census Bureau's own history page documents the March 1951 contract signature and the June dedication ceremony; the Computer History Museum's 'This Day in History' entry documents the November 1952 election-night broadcast.
Built by: J. Presper Eckert and John Mauchly (Eckert-Mauchly Computer Corporation) · Delivered: Dedicated 14 June 1951, Census Bureau · Input: Magnetic tape, plus punched cards · Notable use: Correctly forecast the 1952 presidential election on live television
- April 1957Well documented
Reputable source · 2 sourceswhy?
Best source: Timeline of Computer History: 1957 (FORTRAN)
The domain "computerhistory.org" is on our Reputable source registry.FORTRAN ships as the first widely used high-level language
Programming a computer in the 1950s meant writing in machine-specific assembly code or raw numeric instructions, tedious and error-prone work. John Backus led a small team at IBM that spent from 1954 to 1957 designing FORTRAN (Formula Translation) and a compiler to convert it automatically into machine code for the IBM 704. The system shipped to IBM 704 customers in April 1957. Backus later said his motivation was simple: 'I didn't like writing programs, so I started work on a system to make them easier to write.'
Why it matters: FORTRAN let scientists and engineers write formulas in something close to ordinary mathematical notation instead of raw machine instructions, and its compiler proved that a program could translate high-level code into efficient machine code without giving up speed. It became the first high-level language to see broad, sustained industrial use, opening the door to the entire discipline of software separate from hardware.
How we know: IBM's own historical profile of Backus documents his role, the FORTRAN development timeline, and the direct quote about his motivation for building the compiler.
Lead designer: John Backus, IBM · Shipped: April 1957, for the IBM 704 · Full name: FORTRAN (FORmula TRANslation) · Significance: First widely used high-level programming language
Sources- IBM. John Backus · reference
- Computer History Museum. Timeline of Computer History: 1957 (FORTRAN) (1957) · secondary
- 1959Well documented
Reputable source · 2 sourceswhy?
Best source: August 1: US Navy Recalls Hopper to Head COBOL Effort
The domain "computerhistory.org" is on our Reputable source registry.Grace Hopper helps design COBOL
The US Navy recalled Captain Grace Murray Hopper, who had already developed one of the world's first compilers in the 1950s, to active duty to help develop a new programming language for business applications. Hopper had worked on the Mark I and Mark II computers at Harvard in the 1940s and had spent the years since building compiler technology that translated instructions written closer to English into machine code. In 1959, she played a central role in defining COBOL (Common Business-Oriented Language), designed with English-like syntax so that businesspeople, not only trained programmers, could read what a program was doing.
Why it matters: COBOL became, in the Computer History Museum's own words, 'probably the most successful programming language for business applications in history,' and decades-old COBOL code still runs banking and government systems today. Its English-like design showed that a programming language did not have to look like mathematics or machine code to be powerful.
How we know: The Computer History Museum's own institutional profile of Hopper and its 'This Day in History' archive both describe her Navy recall, her compiler background, and her role in COBOL's creation.
Key figure: Grace Murray Hopper, US Navy · Year: 1959 · Full name: COBOL (Common Business-Oriented Language) · Design goal: English-like syntax readable by non-programmers
- September 1958 to January 1959Well documented
Primary source · 2 sourceswhy?
Best source: Jack S. Kilby: Biographical
Cited as a "primary" source (no stronger domain match).The integrated circuit is invented twice in a year
In September 1958 at Texas Instruments, Jack Kilby demonstrated the first working circuit with all its components, transistor, resistors, and capacitors, built from a single piece of semiconductor material rather than wired together from separate parts. A few months later, in January 1959, Robert Noyce at Fairchild Semiconductor filed a patent for a different, more practical approach: components diffused directly into a silicon chip and connected by aluminum metal lines deposited on top, building on colleague Jean Hoerni's newly developed planar process. Noyce's design eliminated the fragile hand-wired connections Kilby's version still needed and could be mass-manufactured.
Why it matters: Kilby proved a single chip with multiple components was possible; Noyce's planar version made it manufacturable at scale, and that combination became the foundation of the entire semiconductor industry. Kilby shared the 2000 Nobel Prize in Physics for the invention; Noyce, who died in 1990 and was ineligible, is separately credited by the patent record and by companies he later founded, including Intel.
How we know: Kilby describes his own 1958 work at Texas Instruments in his Nobel Prize autobiography; the Computer History Museum's Silicon Engine project documents Noyce's January 1959 patent filing and the planar process it built on.
First working demo: Jack Kilby, Texas Instruments, September 1958 · Manufacturable version: Robert Noyce, Fairchild Semiconductor, patent filed July 1959 · Kilby's recognition: Shared 2000 Nobel Prize in Physics · Key technique: Jean Hoerni's planar process, used in Noyce's design
- 19 April 1965Well documented
Primary source · 2 sourceswhy?
Best source: Cramming More Components onto Integrated Circuits
Cited as a "primary" source (no stronger domain match).Moore projects the doubling of chip components
Gordon Moore, director of R&D at Fairchild Semiconductor, drew a line through five data points tracking how many components fit on an integrated circuit at the lowest cost per component, from 1959 through 1964. His paper 'Cramming More Components onto Integrated Circuits' was published in Electronics magazine on 19 April 1965. Extrapolating the trend forward, he projected that by 1975 a single chip would hold 65,000 components, implying the number was roughly doubling every year. A decade later, with the industry's actual progress in hand, Moore revised his own forecast to a doubling every two years.
Why it matters: The observation, later dubbed 'Moore's Law' by Caltech professor Carver Mead, became a self-fulfilling target that chipmakers organized their research roadmaps around for decades, driving the ever-cheaper, ever-denser processors behind every device from mainframes to smartphones.
How we know: The Computer History Museum's Silicon Engine project documents the original 1965 paper, the 65,000-component 1975 projection, and Moore's later revision to a two-year doubling directly from the historical record.
Author: Gordon Moore, Fairchild Semiconductor · Published: 19 April 1965, Electronics magazine · Original projection: 65,000 components per chip by 1975 · Term coined by: Carver Mead, Caltech
- 9 December 1968Well documented
Reputable source · 2 sourceswhy?
Best source: The Mother of All Demos
The domain "invention.si.edu" is on our Reputable source registry.Engelbart's 'Mother of All Demos' unveils the mouse
At the Fall Joint Computer Conference in San Francisco, Douglas Engelbart and his team from the Stanford Research Institute demonstrated their oN-Line System (NLS) to an audience of computer professionals. The demonstration gave the very first public look at the computer mouse, alongside hypertext linking between documents, real-time collaborative text editing, multiple on-screen windows with adjustable views, and shared-screen video conferencing between Engelbart on stage and a colleague back at SRI, thirty miles away. Engelbart had spent years pursuing a personal research goal of using computers to 'augment human intellect' rather than simply automate calculation.
Why it matters: Almost everything demonstrated that day, a pointing device, clickable links, overlapping windows, and video calls, became standard computing features only after Xerox PARC engineers, several of whom had worked with Engelbart, carried the ideas forward into the Alto computer in the early 1970s. Nothing shown was a product; it took over a decade for any of it to reach ordinary users.
How we know: SRI, Engelbart's own research institute, documents the 9 December 1968 date, the technologies shown, and the conference location in its own historical account of the demonstration.
Presenter: Douglas Engelbart, Stanford Research Institute (SRI) · Date: 9 December 1968 · Location: Fall Joint Computer Conference, San Francisco · System name: oN-Line System (NLS)
SourcesRelated timelines- History of Video Games → · The mouse and windowed interface Engelbart demonstrated became standard PC input methods years before graphical adventure and strategy games depended on them.
- 29 October 1969Well documented
Reputable source · 2 sourceswhy?
Best source: Day the Infant Internet Uttered its First Words
The domain "lk.cs.ucla.edu" is on our Reputable source registry.ARPANET carries its first message
UCLA professor Leonard Kleinrock and his student programmer Charley Kline set up a message transmission from the UCLA SDS Sigma 7 computer to a Stanford Research Institute computer 350 miles away, operated by programmer Bill Duvall. The plan was to remotely log in by typing 'login.' Kline transmitted the letters L and O; the system crashed before the G. The connection was fixed roughly an hour later and the full login completed. Kleinrock had spent years developing packet switching, a method of breaking information into small, independently routed pieces that get reassembled at their destination, and ARPANET (the Advanced Research Projects Agency Network) was the first network built to test the idea at scale, initially linking four university and research computers.
Why it matters: The accidental 'LO' is remembered as the internet's first words, but the more consequential achievement was proving packet switching worked over real distance between real computers. Every network built afterward, including the modern internet, routes data the same fragmented, reassembled way rather than by holding a dedicated circuit open the way a telephone call does.
How we know: UCLA's own historical record, maintained by Kleinrock's lab, documents the event from the IMP log kept at UCLA at the time, naming the participants, the computers involved, and the exact sequence of the crash.
Sender: Charley Kline, under Leonard Kleinrock, UCLA · Receiver: Bill Duvall, Stanford Research Institute · Date/time: 29 October 1969, 22:30 · Underlying technology: Packet switching
- Late 1971Well documented
Reputable source · 2 sourceswhy?
Best source: Ray Tomlinson
The domain "lemelson.mit.edu" is on our Reputable source registry.Tomlinson sends the first networked email
Ray Tomlinson, a programmer at Bolt, Beranek and Newman (BBN), had helped build ARPANET's networking software. In 1971 he adapted an existing local messaging program called SNDMSG, combining it with a file transfer program called CPYNET, so a message could be sent to a user on a different, remote host computer rather than only to someone logged into the same machine. To make that possible, he needed a way to separate the recipient's username from the name of the machine it lived on, so he picked the @ symbol, chosen mainly because it wasn't already used in usernames or in the TENEX programming environment. He tested the new capability by sending a message between two computers sitting only a few feet apart.
Why it matters: The user@host format Tomlinson invented for that test message is still exactly how every email address is written today. Email became, and long remained, the single most-used application on the networks that grew into the internet.
How we know: The Internet Hall of Fame's official biography of Tomlinson, written for his 2012 induction, documents the SNDMSG/CPYNET combination and his reasoning for choosing the @ symbol.
Inventor: Ray Tomlinson, BBN · Year: 1971 · Programs combined: SNDMSG and CPYNET · Key innovation: The @ symbol separating user from host
Sources - March 1971Well documented
Peer-reviewed · 2 sourceswhy?
Best source: Chip Hall of Fame: Intel 4004 Microprocessor
The domain "spectrum.ieee.org" is on our Peer-reviewed registry.Intel ships the 4004, the first commercial microprocessor
Japanese calculator maker Busicom hired Intel to build a set of chips for a new calculator line. Rather than design several fixed-purpose chips, Intel engineer Ted Hoff, with Stanley Mazor, proposed a single general-purpose processor chip that could be programmed for different tasks, working with Busicom engineer Masatoshi Shima to define what the chip needed to do. Federico Faggin, assisted by Shima, used his experience with silicon-gate MOS technology to fit the design's 2,300 transistors into a single 16-pin package. The first fully working Intel 4004 was delivered in March 1971 for Busicom's calculator prototype, and Intel began selling it to any customer that July.
Why it matters: The 4004 packed an entire central processing unit, the part of a computer that executes instructions, onto one chip for the first time, rather than spreading that logic across a circuit board of separate components. Every processor since, from early personal computers to the phone this article might be read on, is a descendant of putting a whole CPU on a single piece of silicon.
How we know: The Computer History Museum's Silicon Engine project documents Hoff's architectural proposal, Faggin's fabrication work, the 2,300 transistor count, and the Busicom origins from its own historical record of the chip's development.
Designers: Ted Hoff, Federico Faggin, Stanley Mazor, and Masatoshi Shima · Delivered: March 1971 (Busicom prototype); general sale July 1971 · Transistors: 2,300 · Original client: Busicom (calculator chipset)
SourcesRelated timelines- History of Video Games → · General-purpose microprocessors like the 4004 made it possible to build affordable dedicated game hardware later in the 1970s.
- March 1973Well documented
Peer-reviewed · 2 sourceswhy?
Best source: 50 Years Later, We're Still Living in the Xerox Alto's World
The domain "spectrum.ieee.org" is on our Peer-reviewed registry.Xerox PARC's Alto shows the first working GUI
Engineers at Xerox's Palo Alto Research Center built the Alto as a research machine to explore what they called personal, distributed computing rather than a product to sell. It combined a bitmapped display, roughly 600 by 800 pixels capable of showing varied fonts and layouts rather than only fixed characters, with a mouse and software that presented movable, overlapping windows and clickable icons, a genuinely visual way of using a computer rather than typing text commands. Its word processor, Bravo, is considered the first WYSIWYG ('what you see is what you get') editor, showing text on screen exactly as it would print. The system never sold in large numbers; Xerox gave away or sold a few thousand units mostly to research institutions.
Why it matters: Nearly every visual element of modern computing, the pointer, the window, the icon, the idea that a screen shows you what you will get on paper, debuted on the Alto years before Apple or Microsoft shipped anything comparable. Xerox's own 1981 commercial version, the Star, and later Apple's Macintosh both drew directly on what PARC had already built.
How we know: The Computer History Museum's own account of the Alto, in its Revolution exhibit, describes the mouse, the windows and icons interface, and the Bravo WYSIWYG editor from its historical and curatorial record.
Built at: Xerox Palo Alto Research Center (PARC) · First operational: 1973 · Key features: Mouse, overlapping windows, icons, bitmapped display · Word processor: Bravo, the first WYSIWYG text editor
- 22 May 1973Well documented
Reputable source · 2 sourceswhy?
Best source: NIHF Inductee Robert Metcalfe, Who Invented the Ethernet
The domain "invent.org" is on our Reputable source registry.Metcalfe and Boggs build Ethernet at Xerox PARC
Xerox's Palo Alto Research Center (PARC) had built the Alto, an early personal computer, and wanted every Alto in the building connected to each other, to a newly built laser printer, and to the outside world through ARPANET. Robert Metcalfe, with colleague David Boggs, adapted ideas from the University of Hawaii's ALOHAnet, which broadcast data over radio, into a wired version running over coaxial cable, with no central controller. A node listened to the cable and only transmitted when it sensed the line was quiet, a technique called carrier sense. Metcalfe circulated a memo describing the scheme on 22 May 1973, though Boggs later argued the true birthday was 11 November 1973, the day the system first actually worked.
Why it matters: Ethernet let any number of computers share a single cable without a central switch coordinating them, a design so durable that later versions running on twisted-pair and fiber cable still use the same core Ethernet framing, decades after coaxial cable itself disappeared from offices.
How we know: The National Inventors Hall of Fame's induction page for Metcalfe and the IEEE's own Milestone documentation for Ethernet describe the PARC origin, the Alto/printer/ARPANET connection problem, and Boggs's co-invention.
Co-inventors: Robert Metcalfe and David Boggs, Xerox PARC · Memo date: 22 May 1973 · Working system: 11 November 1973 · Technique: Carrier-sense multiple access over shared coaxial cable
- January 1975Well documented
Reputable source · 2 sourceswhy?
Best source: 1975
The domain "computerhistory.org" is on our Reputable source registry.The Altair 8800 launches the hobbyist personal computer
A small Albuquerque firm called MITS (Micro Instrumentation and Telemetry Systems), which had previously sold electronic calculator kits before a price war wiped out that market, put its new computer kit on the cover of Popular Electronics magazine's January 1975 issue. The Altair 8800, built around Intel's 8080 processor, sold for $297 unassembled or $395 with a case, came with just 256 bytes of memory, and had no keyboard or screen; users entered programs by flipping switches and read results from blinking lights. Orders poured in regardless, and MITS co-founder Ed Roberts, who designed the machine, is credited with coining the term 'personal computer.'
Why it matters: The Altair proved a market existed for computers bought by individuals rather than institutions, even one this stripped-down, and its open bus design became the S-100 standard that let a growing ecosystem of hobbyist hardware and software, including an early version of Microsoft's BASIC, plug into the same machine.
How we know: The Computer History Museum's 1975 timeline entry and the Smithsonian's National Museum of American History both document the Popular Electronics cover story, the price, and MITS's role in starting the personal computer era.
Maker: MITS (Micro Instrumentation and Telemetry Systems) · Designer: Ed Roberts · Price: $297 (kit) / $395 (with case) · Processor: Intel 8080
- April 1977Well documented
Reputable source · 2 sourceswhy?
Best source: The Apple II
The domain "computerhistory.org" is on our Reputable source registry.The Apple II debuts at the West Coast Computer Faire
Steve Wozniak designed the Apple II, and Apple, cofounded with Steve Jobs, unveiled it at the West Coast Computer Faire in San Francisco in April 1977. Unlike the Altair or Apple's own earlier Apple I, which required buyers to supply their own case, keyboard, and power supply, the Apple II arrived as a complete, ready-to-use unit that plugged into a television for its display and had BASIC built permanently into its memory. The floppy disk drive, added in 1978, and the VisiCalc spreadsheet program, released in 1979, later turned it into a genuine business tool rather than only a hobbyist machine.
Why it matters: The Apple II was promoted, and largely succeeded, as a computer for ordinary people rather than engineers, proving a mass consumer market existed for a fully assembled personal computer rather than a kit. It became one of the first personal computers to sell in large enough numbers to build a lasting software industry around itself.
How we know: The Computer History Museum's Revolution exhibit documents Wozniak's design, the finished-product packaging, and the later disk drive and VisiCalc additions from its own historical account.
Designer: Steve Wozniak · Debut: West Coast Computer Faire, April 1977 · Shipped: 10 June 1977 · Key later additions: Floppy disk drive (1978), VisiCalc (1979)
SourcesRelated timelines- History of Video Games → · The Apple II's color graphics and open architecture made it an early platform for a wave of commercial computer games.
- 12 August 1981Well documented
Primary source · 2 sourceswhy?
Best source: The IBM PC
Cited as a "primary" source (no stronger domain match).The IBM PC opens personal computing to third parties
IBM introduced the IBM 5150 Personal Computer on 12 August 1981, built around Intel's 8088 processor. Rather than build the machine entirely from proprietary IBM parts as it had with earlier products, IBM used off-the-shelf components and, unusually for the company, published detailed technical documentation describing the PC's open architecture. That decision let outside companies build compatible expansion cards, and eventually entire compatible computers, that would work with IBM's machine. The PC ran PC-DOS, IBM's version of the operating system Microsoft had bought the rights to and licensed to IBM while keeping the right to sell it to others as MS-DOS.
Why it matters: IBM's name gave personal computers instant legitimacy in businesses that had ignored the Apple II and Altair as hobbyist toys, while the open documentation it published let Compaq and others build IBM-compatible 'clones,' eventually built better and sold cheaper than IBM's own machines. That openness, replicated across the industry, made the Intel-plus-Windows PC platform, rather than IBM itself, the long-term winner.
How we know: IBM's own corporate history page documents the August 1981 launch, describes its open architecture, and states plainly that the PC 'streamlined business operations, spurred the development of the software industry and vaulted computing into the mainstream.'
Model: IBM 5150 Personal Computer · Launched: 12 August 1981 · Processor: Intel 8088 · Operating system: PC-DOS (IBM's licensed version of Microsoft's MS-DOS)
Sources- IBM. The IBM PC · primary
- HistoryOfInformation.com. IBM Introduces the IBM 5150 - The IBM PC · reference
- 1 January 1983Well documented
General source · 2 sourceswhy?
Best source: Final report on TCP/IP migration in 1983
Cited as a "reference" source (no stronger domain match).ARPANET switches to TCP/IP, creating the internet
ARPANET had run since 1969 on the Network Control Program (NCP), a protocol built only to connect ARPANET's own hosts, with no way to link it to other, separate networks. In March 1982, the US Department of Defense declared a new protocol suite, TCP/IP, its official standard and set a deadline: every host still using NCP after 1 January 1983 would lose access to the network entirely. Engineers coordinated a single global cutover, sometimes called 'flag day,' on that date, extensively tested in advance so the transition caused no major outages. TCP/IP's use of 32-bit addressing also allowed for roughly 4 billion possible host addresses, letting separate networks interconnect into one network of networks.
Why it matters: NCP could only ever be ARPANET's private internal protocol; TCP/IP was designed from the start to let independent networks connect to each other, which is the literal meaning of 'internetworking.' The 1 January 1983 cutover is the date many historians point to as the moment ARPANET truly became part of what we now call the internet.
How we know: The Internet Society's own report on the 1983 migration, based on a presentation by engineer Ron Broersma who participated in the transition, documents the DoD's 1982 standard declaration, the 1 January 1983 deadline, and the address-space consequences of the switch.
Old protocol: NCP (Network Control Program) · New protocol: TCP/IP · Cutover date: 1 January 1983 · Mandated by: US Department of Defense, standard declared March 1982
- September 1983Well documented
Primary source · 2 sourceswhy?
Best source: Overview of the GNU System
Cited as a "primary" source (no stronger domain match).GNU project begins the free software movement
Richard Stallman, a programmer at MIT's Artificial Intelligence Lab, made the Initial Announcement of the GNU Project on a Usenet newsgroup in September 1983. GNU, a recursive acronym for 'GNU's Not Unix,' aimed to build a complete operating system compatible with Unix but made entirely of free software, meaning anyone could run, copy, study, and modify it without a proprietary license restricting them. Stallman personally wrote core pieces including the GNU Emacs editor and a C compiler; by 1990, the project had built or found nearly every major component of an operating system except one, the kernel, the core program that manages a computer's hardware and resources.
Why it matters: GNU supplied the surrounding tools and philosophy, but the operating system it needed was completed only when Linus Torvalds released a compatible kernel called Linux in 1991, made free software the following year; combined, the two became the GNU/Linux system that today runs the majority of the world's web servers and underlies Android phones.
How we know: The GNU Project's own history page, published by the Free Software Foundation, documents the September 1983 announcement, the project's goals, and its own account of how Linux completed the missing kernel piece.
Founder: Richard Stallman · Announced: September 1983 · Full name: GNU ('GNU's Not Unix') · Missing piece completed by: Linus Torvalds's Linux kernel, 1991
- 24 January 1984Well documented
Reputable source · 3 sourceswhy?
Best source: Remembering Apple's "1984" Super Bowl ad
The domain "americanhistory.si.edu" is on our Reputable source registry.Apple launches the Macintosh with a Super Bowl ad
Apple aired a one-minute commercial called '1984,' directed by Ridley Scott, during Super Bowl XVIII on 22 January 1984, casting IBM's dominance of the computer industry as a dystopian Big Brother that the Macintosh would defeat. Two days later, on 24 January, Apple unveiled the Macintosh itself at its shareholders meeting. Drawing directly on ideas Steve Jobs had seen at Xerox PARC and on Apple's own underperforming Lisa computer, the Macintosh built in a mouse, a graphical interface with overlapping windows and menus, clickable icons, and cut-copy-paste editing, packaged as a complete, self-contained product rather than a kit or expandable box.
Why it matters: The Macintosh was the first computer to bring PARC's graphical interface ideas to a mass consumer audience rather than a research lab or a five-figure workstation like the Xerox Star. It set the visual template, icons, windows, a mouse, that both Apple's own later products and, eventually, Microsoft Windows would follow.
How we know: The Smithsonian's National Museum of American History documents the 1984 ad, the January 24 launch, and the Macintosh's PARC-derived and Lisa-derived design features from its own historical account and object collection.
Ad aired: 22 January 1984, Super Bowl XVIII · Ad director: Ridley Scott · Macintosh unveiled: 24 January 1984 · Key features: Mouse, overlapping windows, icons, cut-copy-paste
Sources- Smithsonian National Museum of American History. Remembering Apple's "1984" Super Bowl ad · reference
- Smithsonian Lemelson Center for the Study of Invention and Innovation. Remembering Apple's "1984" Super Bowl Ad · reference
- Computer History Museum. Timeline of Computer History: 1984 (Apple 1984 commercial) (1984) · secondary
- 12 March 1989Well documented
Primary source · 2 sourceswhy?
Best source: A short history of the Web
Cited as a "primary" source (no stronger domain match).Berners-Lee proposes the World Wide Web at CERN
Tim Berners-Lee, a software engineer at CERN, the European particle physics laboratory, wrote a proposal in March 1989 for what he called an information management system, meant to solve the practical problem of CERN's own scattered documentation, spread across incompatible formats and systems used by scientists visiting from around the world. Working with Belgian colleague Robert Cailliau, he formalized the idea as a management proposal in November 1990 and, by Christmas 1990, had defined the core pieces still used today, HTML for formatting documents, HTTP for transferring them, and the URL for addressing them, and had written the first web browser and server software himself. In 1991, he released the software, first to CERN colleagues and then, that August, to internet newsgroups worldwide.
Why it matters: Berners-Lee designed the web to run on top of the internet that already existed rather than building a new network, meaning anyone already connected to the internet could use it once they had the software, which is a large part of why it spread so quickly once released.
How we know: CERN's own history of the web documents the March 1989 proposal, the November 1990 formal proposal with Cailliau, and the 1991 software release directly from the institution where it happened.
Inventor: Tim Berners-Lee, CERN · Collaborator: Robert Cailliau (formal proposal, November 1990) · Initial proposal: 12 March 1989 · Public software release: 1991
- January 1993Well documented
Reputable source · 2 sourceswhy?
Best source: NCSA Mosaic
The domain "ncsa.illinois.edu" is on our Reputable source registry.Mosaic makes the web visual
Marc Andreessen and Eric Bina, programmers at the National Center for Supercomputing Applications (NCSA) at the University of Illinois, built Mosaic, a browser released as an early version in January 1993 and as version 1.0 that April, with Windows and Macintosh versions following that September. It was the first widely used browser to display images inline on the same page as text, rather than as separate windows or plain text placeholders, arranging pictures and words the way a printed magazine layout would. By December 1993, more than 5,000 copies were being downloaded every month, and the New York Times called it network computing's first 'killer app,' a program so useful on its own that it could create an entire new industry from nothing.
Why it matters: Berners-Lee's original web software worked, but Mosaic made it visual and easy enough for non-technical people to want to use, turning the web from a tool for physicists and researchers into something a much broader public would adopt within a few years. Andreessen left NCSA to found Mosaic Communications, later renamed Netscape, carrying the browser's lineage directly into the first browser wars.
How we know: The University of Illinois's own NCSA page documents Andreessen and Bina's roles, the inline-image innovation, and the download numbers that made Mosaic the web's first mass-adopted browser.
Creators: Marc Andreessen and Eric Bina, NCSA, University of Illinois · First release: January 1993 · Key innovation: Inline images shown with text on the same page · Later company: Mosaic Communications, renamed Netscape
- August 1998Well documented
Primary source · 2 sourceswhy?
Best source: How we started and where we are today
Cited as a "primary" source (no stronger domain match).Google is incorporated
Larry Page and Sergey Brin met at Stanford in 1995 and began building a search engine, initially called BackRub, that ranked web pages by treating links to a page as votes for its importance rather than only matching keywords in the page's own text, an approach that became the PageRank algorithm. In August 1998, Sun Microsystems cofounder Andy Bechtolsheim wrote them a $100,000 check on the spot after a demonstration, and Google Inc. was officially incorporated.
Why it matters: PageRank produced noticeably more relevant results than the keyword-matching search engines already on the market, and that quality gap is what let Google grow from a Stanford dorm-room project into the company that eventually became the web's default entry point and, later, the owner of one of the most valuable technology businesses in the world.
How we know: Google's own corporate history page, 'How we started and where we are today,' documents the Stanford meeting, the BackRub search engine, and the August 1998 incorporation directly.
Founders: Larry Page and Sergey Brin · Incorporated: August 1998 · Original name: BackRub · Key algorithm: PageRank
Sources - 1999Well documented
Primary source · 2 sourceswhy?
Best source: The Evolution of Wi-Fi Technology and Standards
Cited as a "primary" source (no stronger domain match).Wi-Fi reaches consumers with 802.11b
The IEEE ratified the original 802.11 wireless networking standard in 1997, transmitting at up to 2 megabits per second over unlicensed 2.4 GHz radio spectrum, too slow for most practical use. In 1999, the IEEE ratified 802.11b, raising the top speed to 11 megabits per second, still on the 2.4 GHz band, fast enough to be genuinely useful for general computing. Apple's commercial breakthrough that same year, its AirPort wireless base station bundled with the iBook laptop, was the first mass-marketed consumer product built around the new standard.
Why it matters: 802.11b's jump to 11 Mbps, more than five times the original standard's speed, is what turned wireless networking from a niche technology into something worth putting in an ordinary consumer laptop, and it was the platform that carried the word 'Wi-Fi' into everyday language.
How we know: The IEEE Standards Association's own account of Wi-Fi's evolution documents the 1997 and 1999 standards, their data rates, and Apple's AirPort launch as the technology's first major consumer product.
Original standard: IEEE 802.11 (1997), up to 2 Mbit/s · Consumer breakthrough standard: IEEE 802.11b (1999), up to 11 Mbit/s · First mass-market product: Apple AirPort base station and iBook, 1999 · Frequency band: 2.4 GHz (unlicensed)
- 10 March 2000 (Nasdaq peak)Well documented
Reputable source · 2 sourceswhy?
Best source: The Great Moderation
The domain "federalreservehistory.org" is on our Reputable source registry.The dot-com bubble bursts
Through the late 1990s, investors poured venture capital into internet startups on the promise of future growth rather than current profit, driving the Nasdaq Composite index up 86 percent in 1999 alone and to a peak of 5,048 points on 10 March 2000. Cheap credit after the Federal Reserve lowered interest rates following the 1998 collapse of hedge fund Long-Term Capital Management helped fuel the run-up; by December 1999, Nasdaq-listed stocks were worth 80 percent as much as the entire New York Stock Exchange, up from just 11 percent in 1990. The bubble then imploded. The Nasdaq fell more than 75 percent between March 2000 and October 2002, erasing more than five trillion dollars in market value, and companies with no sustainable revenue, including the online pet-supply retailer Pets.com, went out of business within months.
Why it matters: The crash wiped out a generation of internet startups and preceded the US recession of 2001, but the infrastructure built during the boom, fiber optic networks, data centers, and a large pool of engineers who had learned to build for the web, remained in place and underpinned the next decade's real growth once profitable companies like Google and Amazon proved the underlying business model could work.
How we know: Goldman Sachs's own institutional history and the Federal Reserve's essay on the Great Moderation both document the Nasdaq's rise and 75 percent fall, the Federal Reserve's rate policy, and the bubble's connection to the subsequent 2001 recession.
Nasdaq peak: 5,048 points, 10 March 2000 · Decline: More than 75%, March 2000 to October 2002 · Value lost: More than $5 trillion · Notable failure: Pets.com
Sources - 15 January 2001Well documented
Reputable source · 2 sourceswhy?
Best source: Wikipedia launches
The domain "history.com" is on our Reputable source registry.Wikipedia launches as an open, editable encyclopedia
Jimmy Wales and Larry Sanger had been running Nupedia, a free online encyclopedia whose articles were written by credentialed experts and passed through a slow, formal peer-review process. In its first six months, Nupedia had published only two articles. Sanger proposed adding a wiki, software that lets any visitor edit a page directly in a web browser, as a faster way to draft material for Nupedia's review queue. That side project, launched 15 January 2001 and named Wikipedia, a blend of 'wiki' and 'encyclopedia,' let anyone write or edit an article with no review at all. It had more than 20,000 articles across 18 languages by the end of its first year.
Why it matters: Open, unreviewed editing, the opposite of Nupedia's slow expert gatekeeping, is precisely what let Wikipedia grow so fast, and it remains the reason the site struggles with vandalism and accuracy even as it became the internet's most consulted general reference. Its scale forced SourcedStory and every serious reference project since to reckon with a tertiary source that is enormously useful as a map to citations and unusable as a citation itself.
How we know: HISTORY's own account of the January 2001 launch describes Nupedia's failure, the two-article count in its first six months, and Wikipedia's growth to 20,000 articles by its first anniversary.
Founders: Jimmy Wales and Larry Sanger · Launched: 15 January 2001 · Predecessor project: Nupedia (expert peer review, launched 2000) · Growth: More than 20,000 articles in 18 languages within a year
Sources- HISTORY (A&E Networks). Wikipedia launches · reference
- Internet Hall of Fame. Jimmy Wales · reference
- 4 February 2004Well documented
Reputable source · 2 sourceswhy?
Best source: December 30: Facebook Registers One Millionth User
The domain "computerhistory.org" is on our Reputable source registry.Zuckerberg launches thefacebook from a Harvard dorm room
Mark Zuckerberg, a Harvard sophomore, launched thefacebook on 4 February 2004 from his Kirkland House dorm room, running on a single rented server costing $85 a month. It began as an internal directory limited to Harvard students, letting them build profile pages and connect with classmates. Within 24 hours, 650 students had registered; within two weeks, that had grown to 4,300, and by the end of the first month, more than half of Harvard's undergraduates had signed up. Zuckerberg had built the site with roommates and classmates including Eduardo Saverin, Dustin Moskovitz, Andrew McCollum, and Chris Hughes.
Why it matters: Facebook expanded from Harvard to other universities and eventually anyone with an email address, and by the end of 2004 had already registered its millionth user, the beginning of a growth curve that made it, for over a decade, the dominant social network and the template every later platform copied and adapted.
How we know: The Harvard Crimson's own retrospective account, published by the university's student newspaper for the tenth anniversary, documents the exact launch date, dorm room, hosting cost, and early registration figures. The Computer History Museum separately records Facebook reaching its one millionth user by the end of December 2004.
Founder: Mark Zuckerberg, with Eduardo Saverin, Dustin Moskovitz, Andrew McCollum, and Chris Hughes · Launched: 4 February 2004, Kirkland House, Harvard · Early growth: 650 users in 24 hours; over half of Harvard undergraduates within a month · Milestone: One millionth user by 30 December 2004
- 14 March 2006 (S3 launch)Well documented
Primary source · 2 sourceswhy?
Best source: Our Origins
Cited as a "primary" source (no stronger domain match).Amazon launches S3 and EC2, starting the cloud
Amazon had learned, running its own retail business, how expensive and difficult it was to provision and manage computing infrastructure, which distracted engineering teams from building the actual product. On 14 March 2006, Amazon launched Simple Storage Service (S3), letting any customer store data on Amazon's own servers instead of buying and maintaining storage hardware; 12,000 developers signed up within a single day. Amazon followed a few months later, in a limited public beta announced 25 August 2006, with Elastic Compute Cloud (EC2), which gave customers on-demand access to computing power itself, letting them run applications on Amazon's servers rather than their own, paid for by usage rather than owned outright.
Why it matters: S3 and EC2 turned computing infrastructure into a utility a startup could rent by the hour instead of capital equipment it had to buy upfront, removing one of the biggest barriers to launching a new internet company and setting the pattern every other major cloud provider, including Microsoft Azure and Google Cloud, later followed.
How we know: Amazon Web Services documents its own founding motivation and the S3 and EC2 launch dates directly, describing the shift from owning hardware to renting infrastructure as a service.
S3 launched: 14 March 2006 · EC2 launched: Limited beta, 25 August 2006 · Parent company: Amazon.com · Core idea: Renting storage and compute as a metered service
SourcesRelated timelines- Artificial Intelligence → · The rentable, on-demand compute AWS pioneered later became the infrastructure that trained and served large-scale machine learning models.
- 9 January 2007Well documented
Primary source · 2 sourceswhy?
Best source: Apple Reinvents the Phone with iPhone
Cited as a "primary" source (no stronger domain match).Apple unveils the iPhone
Steve Jobs introduced the iPhone at the Macworld conference in San Francisco on 9 January 2007, describing it as three devices in one: 'a revolutionary mobile phone, a widescreen iPod with touch controls, and a breakthrough internet communications device.' Its defining feature was a new interface built on a large multi-touch display, controlled entirely by finger taps and gestures instead of a stylus or physical keyboard, which Apple had spent years developing. The phone went on sale that June, with a 4GB model priced at $499 and an 8GB model at $599.
Why it matters: The iPhone established the touchscreen slab, with no physical keyboard, as the default smartphone shape, a template nearly every competing phone maker adopted within a few years. Its later App Store, launched in 2008, turned the device into a platform other companies could build businesses on top of, rather than only a product Apple sold once.
How we know: Apple's own original newsroom press release from the January 2007 announcement documents the date, the multi-touch interface, and Jobs's description of the device directly.
Announced: 9 January 2007, Macworld, San Francisco · Announced by: Steve Jobs · Key interface: Multi-touch display · On sale: June 2007, from $499
SourcesRelated timelines- Artificial Intelligence → · The iPhone's always-connected sensors and later on-device chips became a major platform for consumer AI features.
- 10 July 2008Well documented
Primary source · 2 sourceswhy?
Best source: The App Store turns 10
Cited as a "primary" source (no stronger domain match).The App Store opens the mobile app economy
The original iPhone launched in June 2007 with only the fifteen applications Apple had built into it and no way for outside developers to add more. Apple opened the App Store on 10 July 2008 with 500 apps available at launch, built by third-party developers who had paid a $99 annual fee to join the iPhone Developer Program announced that March. Users downloaded roughly 10 million apps in the store's first weekend alone.
Why it matters: The App Store turned the iPhone from a fixed set of Apple-built features into an open platform where any developer could reach millions of users directly, and by its tenth anniversary in 2018, Apple reported that developers had collectively earned more than $100 billion through the store, establishing app sales and in-app purchases as a lasting business model rather than a novelty.
How we know: Apple's own tenth-anniversary newsroom release documents the 10 July 2008 launch date, the 500-app starting count, and the cumulative developer earnings figure directly.
Launched: 10 July 2008 · Apps at launch: 500 · Developer fee: $99 per year (iPhone Developer Program, announced March 2008) · Developer earnings by 2018: Over $100 billion
Sources - 9 February 2016Debated
Reputable source · 2 sourceswhy?
Best source: The Apple-FBI Debate Over Encryption
The domain "npr.org" is on our Reputable source registry.Apple refuses to unlock the San Bernardino iPhone
After the December 2015 San Bernardino terrorist attack, the FBI recovered an iPhone 5C that had belonged to one of the shooters, but could not access its data because of the encryption built into iOS. On 9 February 2016, the FBI announced it wanted Apple's help, and a court order followed asking Apple to build custom software that would let investigators bypass the phone's passcode limits and try unlimited password guesses. Apple CEO Tim Cook publicly refused, arguing that creating such a tool at all, even for one phone, would create a permanent capability that could be turned against any iPhone user's encryption in the future. The FBI dropped its legal demand after paying a third-party contractor more than $1.3 million to break into the phone without Apple's help.
Why it matters: The case became the highest-profile public argument over whether government agencies should be able to compel a technology company to weaken its own encryption for law enforcement access, a debate that intensified after Edward Snowden's 2013 disclosures of NSA mass surveillance programs and remains unresolved in law today.
How we know: NPR's ongoing coverage series on the dispute documents the FBI's February 2016 request, Apple's public refusal, and the eventual withdrawal of the case after the FBI paid an outside vendor to unlock the phone independently.
Phone: iPhone 5C belonging to a San Bernardino attacker · FBI request: 9 February 2016 · Apple's position: Refused, citing precedent risk to all users' encryption · Resolution: FBI paid a third party over $1.3 million to unlock the phone independently