pplpod

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

On July 1, 1970, Xerox chief scientist Jack Goldman founded the Palo Alto Research Center, placing it 3,000 miles from the copier company's East Coast headquarters. Bob Taylor filled its computer science lab with engineers leaving the nearby Stanford Research Institute as government funding dried up. In under a decade that team produced bitmap graphics, the graphical user interface, the Bravo what you see is what you get editor, Smalltalk and object-oriented programming, Ethernet, and the laser printer, then fused them into one machine: the Xerox Alto.

This episode is the biography of an institution that drew the blueprint for personal computing and watched others sell it. Executives who made fortunes leasing giant copiers to purchasing departments had no way to market a desk-sized computer to individuals. The distance that freed the researchers also left them speaking a language headquarters could not hear, and the talent walked out the door to Apple and beyond.

• The mouse was invented by Douglas Engelbart at SRI. PARC's contribution was weaving it into the Alto's overlapping windows, networking, and printing.
• Xerox's own attempt, the 8010 Star workstation of 1981, was expensive and aimed at office administrators. It sold only about 25,000 units.
• David Liddle and Charles Irby left in 1982 to found Metaphor Computer Systems, which reached nearly $40 million in annual revenue by 1987 before IBM acquired it. Other researchers took the interface work to the Apple Lisa and Macintosh.
• The great fumble story is incomplete: Xerox understood the laser printer immediately and built a multibillion dollar business on it, alongside PARC advances in LCDs and optical discs.
• PARC later pursued ubiquitous computing with the location-aware PARCTAB, became a subsidiary in 2002, and was donated to SRI International in April 2023, returning to the institution it once recruited from.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

In 1979, Brigham Young University graduate student Bruce Bastian and computer science professor Alan Ashton wrote a word processor for a Data General minicomputer owned by the city of Orem, Utah, and kept the rights. Their company, Satellite Software International, moved the program to the IBM PC in 1982, and because no C compiler existed for the machine, they wrote it entirely in x86 assembly language. That forced choice made WordPerfect faster than WordStar and, by 1990, the owner of 60 percent of the word processing market.

This episode follows the rise, the collapse, and the odd afterlife. WordPerfect won devotion with Reveal Codes, function key commands, safe macros, and free phone support, then stumbled badly when the world moved to Windows. A sale to Novell and a fire sale to Corel followed. Yet the software endures in law offices and government agencies, where exact formatting and decades of backward compatibility still matter more than fashion.

• The founders each held 49.9 percent of the company and gave office manager Pete Peterson 0.2 percent to act as tiebreaker.
• Reveal Codes split the screen to show every formatting tag in the text stream, much like HTML, so a stray bold code could be found and deleted instead of guessed at.
• WordPerfect macros lived outside the document, so files could not carry the macro viruses that later plagued Word. Support callers were entertained by hold jockeys while they waited.
• The first Windows version in 1991 was unstable and had to be installed from DOS, and Alt F4, a text blocking command in WordPerfect, closed the program in Windows. The company claimed Microsoft withheld API specifications.
• Novell paid $1.4 billion in 1994 and sold to Corel for $186 million in 1996. Corel closed the Orem offices in 1998 and moved development to Ottawa.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Teachers and librarians once warned that a website anyone could edit was where facts went to die. Two decades later, federal courts cite it and scientific journals ask authors to update it. This episode goes back to the year 2000 and Nupedia, a slow, expert-reviewed encyclopedia that demanded credentials and a multi-step peer review. In January 2001, Jimmy Wales and Larry Sanger launched Wikipedia as a feeder project, a public scratch pad for rough drafts. The scratch pad swallowed the encyclopedia it was built to serve.

Behind the growth sat a theory and a license. Wales, then running the web portal Bomis, drew on Friedrich Hayek's idea of spontaneous order, betting that scattered local knowledge would organize itself. Richard Stallman's GNU Free Documentation License made sure nobody could wall the work off. The episode then examines what came after the boom: stalled editor growth, edit wars settled without a boss, a damaging hoax, structural bias, and a co-founder turned critic.

• By 2007 the English edition passed two million articles, overtaking the Yongle Encyclopedia compiled in Ming dynasty China in 1408. Today there are over 68 million articles in more than 300 languages.
• Growth then plateaued as the obvious topics ran out, anti-vandalism bots and formatting rules scared off newcomers, and search engines began answering questions without sending visitors to the site.
• Researchers map conflict through mutually reverting edit pairs. English editors fought hardest over George W. Bush, anarchism, and Muhammad, while German editors clashed over Croatia, Scientology, and 9/11 conspiracy theories.
• In May 2005 an anonymous editor falsely linked journalist John Seigenthaler to the Kennedy assassination, and the claim stood for four months. The scandal produced strict rules for biographies of living persons.
• A draft page for physicist Donna Strickland was rejected in May 2018 for lack of media coverage, five months before she won the Nobel Prize. In October 2025 Larry Sanger published his Nine Theses, arguing the site has abandoned neutrality.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

In the early 1870s, British portrait photographer Harry Pointer posed his cats roller skating and sitting at tiny tea tables, then printed joke captions beneath them on souvenir cards people traded in their parlors. This episode argues that his Brighton Cats series holds the full DNA of the internet meme, and follows the idea across 150 years to the lolcat: a cat photo stamped with blocky white text in a deliberately broken dialect. The format is sillier than it looks and more rule-bound than anyone expected.

Linguists who assumed lolspeak was baby talk found consistent grammar instead, with regularized verbs, phonetic spelling, and fixed templates such as the cat in your fridge eating your food. The episode traces the form from Caturday threads on the anonymous image board 4chan to the January 2007 launch of I Can Has Cheezburger, and then to its strangest offspring: a crowdsourced Bible translation, a working programming language, and a taxidermy sculpture in a major museum.

• Victorian taxidermist Walter Potter built dioramas of stuffed kittens attending a wedding, part of a craze for animals staged in human scenes.
• After Time journalist Lev Grossman dated the earliest lolcat to 2006, readers pushed him to correct it to 2005. Caturday.com was registered on April 30, 2005.
• The American Dialect Society named lolcat a runner-up for 2007 Word of the Year. It lost to Googleganger.
• Fans built a mythology around Ceiling Cat as a watching god and Basement Cat as his dark rival, then cast them as God and Satan in the LOLCat Bible Translation Project.
• LOLCODE is a functioning esoteric programming language with commands like I HAS A and KTHXBYE, and a 2016 taxidermy sculpture of Ceiling Cat now sits in the San Francisco Museum of Modern Art's permanent collection.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

More than a decade before the World Wide Web, two graduate students, Tom Truscott and Jim Ellis, linked Unix machines at Duke and the University of North Carolina over ordinary phone lines. Locked out of the government-funded ARPANET, they used the UUCP protocol and late-night modem calls to pass messages from server to server, a store and forward design that worked like a digital bucket brigade. This episode tells how that poor man's ARPANET became the first global town square, with no CEO, no central servers, and no advertising algorithms.

Usenet gave the internet words like flame, spam, and sock puppet, turned the FAQ into a tool for saving bandwidth, and hosted the first announcements of the World Wide Web, Linux, and the Mosaic browser. Then users bent a text-only system into a file-sharing giant. The episode follows the encoding tricks that made that possible, the 2008 pressure campaign that pushed major providers to drop it, and the paid, hidden network that now moves more data than ever.

• The Big Eight hierarchies, including comp, sci, rec, and talk, organized discussion. Tim Berners-Lee announced his hypertext project on alt.hypertext in 1991, and Linus Torvalds asked for feedback on comp.os.minix.
• To push 8-bit files through a 7-bit ASCII network, users converted binaries to text with uuencode and later yEnc, split them into hundreds of posts, and repaired missing pieces with PAR files built on Reed-Solomon error correction.
• In 2008, New York Attorney General Andrew Cuomo pressed internet providers over illegal material, and Time Warner Cable, Verizon, and Sprint cut or restricted their Usenet feeds.
• Google bought Deja News in 2001 and folded its archive into Google Groups, then ended support for new Usenet content in February 2024.
• Daily traffic grew from roughly 4.5 gigabytes in December 1996 to more than 274 terabytes by February 2024. Commercial providers now advertise over 12 years of retention on more than 60 petabytes of storage.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

In January 2026, someone cleaning out a closet at the University of Utah found a nine-track magnetic tape holding a surviving copy of Unix version 4. This episode treats that discovery as the opening of a biography: the life of an operating system that began as an unsanctioned project at Bell Labs in the late 1960s. Ken Thompson, Dennis Ritchie, Doug McIlroy, and Joe Ossanna had grown tired of Multics, the overbuilt time-sharing system from MIT, Bell Labs, and General Electric, and retreated to a spare PDP-7 to build something small enough to understand.

What they made spread for reasons both technical and accidental. Rewriting the kernel in C in 1973 freed it from any one machine, and a philosophy of small tools, plain text, and pipes made it endlessly reusable. Because a consent decree kept AT&T out of the software business, universities got the source code for $150. The episode follows the Unix Wars, the trademark lawsuits, and the free descendants that finally carried the design everywhere.

• Brian Kernighan coined the name as a pun on Multics: Unics, for Uniplexed Information and Computing Service, later respelled Unix.
• Writing a kernel in a high-level language was considered impractical in 1973, and C itself was still evolving. The payoff was the first truly portable operating system.
• AT&T charged companies $20,000 for a license but gave universities the source for $150, which let Illinois researchers tie Unix to the ARPANET and Berkeley build BSD.
• After the Unix Wars among System V, BSD, Xenix, AIX, SunOS, and HP-UX, the rights passed to Novell in 1992 and the trademark ended up with The Open Group, while the name itself appeared on diapers and hairdryers.
• Richard Stallman launched GNU in 1983 and Linus Torvalds released the Linux kernel in 1991. In a 1999 interview, Dennis Ritchie called the Linux phenomenon delightful.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

In 1995, connecting a printer, modem, or joystick to a PC meant a tangle of parallel ports, serial ports, PS/2 plugs, and game ports, plus driver floppies and hand-assigned IRQ channels that crashed the machine when two devices collided. Seven companies, Intel, Microsoft, IBM, Compaq, DEC, NEC, and Nortel, set out to replace all of it with one plug and play, hot-swappable port that carried its own power. This episode traces the Universal Serial Bus from that fix to the cable that now runs nearly everything on your desk.

The story turns on a design choice: a tiered star topology with a single host in charge and up to 127 peripherals that speak only when spoken to. Putting all the intelligence in the computer made mice and keyboards almost free to build, and that economics let USB outlast the technically superior FireWire. The same trusting, low-cost design also left an opening for BadUSB sticks that pose as keyboards and USB killers that fry a motherboard with its own electricity.

• USB 2.0 reached 480 megabits per second in 2000 but ran half-duplex, like a walkie-talkie. USB 3.0 arrived in 2008 with five extra wires, full duplex, and 5 gigabit SuperSpeed.
• The 8b/10b encoding in USB 3.0 kept sender and receiver clocks in sync at a cost of 20 percent overhead. A later 128b/132b scheme cut that to about 3 percent.
• Names like USB 3.2 Gen 2x2 were engineering descriptions, not marketing, and the standards body eventually switched to plain speed labels such as USB 20Gbps.
• USB-C dropped the old Type-A and Type-B roles in favor of configuration channel pins that negotiate who is host and who supplies power, scaling from 2.5 watts in the original spec to 240 watts at 48 volts.
• A USB killer charges internal capacitors from the port's 5 volts, steps up past 200 volts, and fires it back down the data lines. Unlike Ethernet, USB was built with little electrical isolation to keep chips cheap.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

In 2007, David Karp and Marco Arment built Tumblr in a two-week gap between consulting contracts, and it gained 75,000 users in two weeks. By 2011, half its users were under 25. With no public follower counts and a tag field that people filled with rambling private commentary, it became a haven for niche communities, especially LGBTQ youth. In May 2013, Yahoo bought it for 1.1 billion dollars in cash. Six years later it sold for less than 3 million.

This episode traces how that value vanished. Yahoo bought an audience that mocked advertising and wrote down 712 million dollars by 2016. Verizon inherited the site in 2017 and, after Apple pulled the app from its store in late 2018, banned all adult content and handed enforcement to a crude algorithm. Traffic collapsed, Automattic picked up the remains in 2019, and the platform now loses 30 million dollars a year while its archive is sold as training data.

• Nearly 170,000 people signed a petition against the Yahoo sale, and users reblogged sponsored posts with sarcastic tags until the campaigns became jokes.
• The automated filter looked for skin tones and curves, so it flagged raw chicken, sand dunes, and a line drawing of a vomiting unicorn while real violations slipped through.
• By March 2019, Tumblr had lost an estimated 30 percent of its traffic.
• A 2022 settlement with the New York City Commission on Human Rights found the filter had disproportionately hit LGBTQ users and required algorithmic reviews.
• With more than half a billion blogs and roughly 172 billion posts, Tumblr began opting users into data sales to AI companies including Midjourney and OpenAI.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

The transistor is the most manufactured artifact in human history: by 2018, more than 13 sextillion had been made. It replaced the thermionic triode, a glass vacuum tube invented in 1907 that had to boil electrons off a heated plate, ran hot, and burned out like a light bulb. Physicist Julius Edgar Lilienfeld patented a solid-state alternative in Canada in 1925, but the pure semiconductor crystals needed to build it did not exist. The breakthrough came at Bell Labs in December 1947.

This episode explains how the device was invented, how it became microscopic, and how a speck of silicon does arithmetic. John Bardeen and Walter Brattain succeeded after William Shockley's first design failed, and the three shared the 1956 Nobel Prize in Physics. A furnace accident in 1955 then opened the way to printing transistors by the million. The story ends with the weaknesses of solid-state electronics and the places where vacuum tubes still win.

• Shockley's field-effect attempt was blocked by electrons trapped at the crystal surface, so Bardeen and Brattain pressed two gold point contacts directly onto germanium.
• Bell Labs engineer John R. Pierce coined the name by combining transfer and resistor. In France, Herbert Mataré and Heinrich Welker independently built a similar device in 1948 and called it the transistron.
• Carl Frosch and Lincoln Derick accidentally grew a silicon dioxide layer on a wafer in 1955, which led to photolithography and to the MOSFET, invented by Mohamed Atalla and Dawon Kahng in 1959.
• The MOSFET accounts for roughly 99.9 percent of all transistors, and a single Apple M3 Max chip holds about 92 billion.
• A cosmic ray can flip a transistor from zero to one, a single event upset, which is one reason tubes survive in broadcast transmitters, satellites, and guitar amplifiers.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Before the transistor, a portable radio was a heavy box of hot, fragile vacuum tubes that needed two separate batteries, one just to heat the filaments. That changed after December 1947, when William Shockley, Walter Brattain, and John Bardeen demonstrated the first working transistor at Bell Laboratories. The first commercial transistor radio, the Regency TR-1, arrived in November 1954 from an unlikely partnership between Texas Instruments and an Indianapolis maker of TV antenna boosters. It cost 49.95 dollars and sounded thin and tinny.

This episode follows the pocket radio from costly novelty to cultural force. American firms proved the idea but could not make it cheap. Masaru Ibuka and Akio Morita could, paying 25,000 dollars for a license to the Bell Labs patent, building their components in house, and renaming their company Sony. Their TR-63 of 1957 arrived just as the baby boom, teenage spending money, and rock and roll converged, and it took music out of the family living room for good.

• Texas Instruments had a manufacturing yield of only 20 percent at launch, so four of every five transistors were thrown away, and the TR-1 used just four of them.
• The TR-63 was slightly too big for a shirt pocket, so Sony allegedly gave its salesmen custom shirts with oversized pockets. It also popularized the 9-volt battery.
• Prices fell to about 15 dollars by 1962, and by 1964 no transistor radio was built entirely from U.S. components.
• At night, AM signals from stations in the American South bounced off the ionosphere to Jamaica, where musicians reworked New Orleans rhythm and blues into ska and later reggae.
• As circuits shrank, radios turned into novelties shaped like a Jimmy Carter peanut, Charlie the Tuna, and the egg ship from Mork and Mindy.