Category: Ancient Web

The most ancient times of the internet: 90s, early 2000s.

Posted on by

Ancient Web: Pagat Has Been Cataloging Card Games Since 1995

If somebody teaches you a card game at a kitchen table, there is a decent chance the rules exist on Pagat in a version assembled from people who actually play it.

Visit Pagat

John McLeod founded the site in 1995 with a simple goal: document traditional card and domino games so players could discover unfamiliar games and preserve the rules of ones they already knew.

Three decades later, the structure has become enormous without becoming mysterious.

You can browse alphabetically, by number of players, by region, by mechanism, by objective, or by equipment. There are sections for solitaire, dominoes, proprietary games, invented games, children’s games, poker variants, and traditional games that rarely make it into mass-market rule books.

Folk knowledge with an index

Card-game rules are slippery historical material.

Families change them. Regions use different names. A game moves across borders and picks up local variations. Sometimes two groups use the same name for noticeably different games.

Pagat’s value is not merely that it posts instructions. McLeod explicitly credits readers from around the world for sharing information about the games they play and helping expand the site.

That turns the website into something between a reference manual and a long-running collection of living folk rules.

The name “Pagat” comes from the lowest trump in the Central European game Tarock. That is exactly the sort of fact you wind up learning because somebody maintained a specialist website for thirty years.

The site was still actively updated in September 2026.

From a webmaster’s perspective, Pagat is also a good example of why deep classification matters. A giant subject does not have to become unusable if readers can approach it from several sensible directions. “I have four people and a normal deck of cards” is a perfectly valid navigation problem, and the site has an index for that.

This is one of the sites gathered in CacheRat’s larger Ancient Web research list for people who want the source corpus rather than individual guided stops.

Explore the 1,967 Ancient Web Domains research list

If you vaguely remember a card game from childhood, learned something strange while traveling, or simply need something better to do with a deck than another round of poker, Pagat is still doing its job.

Browse the card-game rules at Pagat

Posted on by

Ancient Web: Daniel Stenberg’s History of IRC

IRC history is messy in exactly the way real Internet history usually is: university machines, borrowed code, network splits, arguments, improvised fixes, and people naming infrastructure after whatever was happening around them.

Read Daniel Stenberg’s History of IRC

Daniel Stenberg’s page pulls that history into one long chronological account. He says he gathered information from multiple sources to verify stories and dates, and the page remains open to corrections through email, issues, and pull requests.

The story begins in Finland in summer 1988.

Jarkko “WiZ” Oikarinen wrote the first IRC client and server while working at the University of Oulu. The first server was tolsun.oulu.fi. Other universities began running servers, and by November 1988 IRC had spread across the Internet.

The network immediately became a network problem

By the middle of 1989, Stenberg’s history records about 40 servers worldwide. Then came disagreements, experiments, splits, reconnects, and the recurring discovery that letting humans share a network generates administrative problems almost immediately.

The page walks through EFnet, early splinter networks, Undernet, RFC 1459, and DALnet.

Some details are especially good because they explain why features appeared.

Undernet worked on timestamps, routing changes, and channel services partly because netsplits and channel takeovers were a recurring headache. DALnet pushed further into nickname and channel ownership and user protections.

Those features were not theoretical protocol decoration. People were annoying one another at scale, so somebody wrote software.

Stenberg also records IRC’s use during the Gulf War in early 1991, when live reports drew more than 300 concurrent users. That number looks tiny in 2026 and enormous when you remember what a distributed real-time Internet chat system looked like at the time.

For modern network builders, the history is useful because it shows protocol development under social pressure. Identity, moderation, routing, naming, abuse, and governance problems did not suddenly appear with social media.

We have been arguing with strangers through network software for a very long time.

This page is one of 1,967 sites in CacheRat’s broader Ancient Web research corpus.

Explore the 1,967 Ancient Web Domains research list

If you have ever typed /join, watched a netsplit, or wondered why IRC has so many oddly named historical networks, Stenberg’s account is worth the detour.

Read the full History of IRC

Posted on by

Ancient Web: W3C’s Timeline of How the Web Happened

The Web became so ordinary that it is easy to forget there was a period when somebody had to explain what a browser was, name the project, buy the machine, and convince other people to run servers.

Read W3C’s “A Little History of the World Wide Web”

This old W3C timeline compresses the story from early hypertext ideas through 1995, and its best feature is how procedural the history feels.

The page starts before the Web itself: Vannevar Bush’s Memex idea, Doug Engelbart’s NLS, Ted Nelson coining “hypertext,” and Tim Berners-Lee’s 1980 ENQUIRE program at CERN.

Then the familiar pieces start clicking into place.

The Web before it was inevitable

In March 1989, Berners-Lee circulated “Information Management: A Proposal” at CERN. In 1990 he began writing a browser/editor on a NeXT machine and settled on “WorldWideWeb” for the program.

The timeline records alternatives for the project name too: Information Mesh, Mine of Information, and Information Mine.

We got lucky.

By Christmas 1990 the WorldWideWeb browser/editor and line-mode browser were demonstrable. In 1991 the software began spreading more widely. By November 1992 the timeline says there were 26 reasonably reliable Web servers.

Then things accelerate.

NCSA Mosaic arrives in 1993. CERN declares the Web technology freely usable with no fees payable to CERN. The number of known HTTP servers climbs. Netscape’s future founders leave NCSA in 1994. The first international Web conference is overwhelmed with applicants. The W3C is founded in October 1994.

Seen as a timeline rather than a legend, the Web looks less like a single invention and more like a chain of decisions, releases, ports, meetings, arguments, and people deciding to run another server.

The page itself was created around 1995 and retains the compact style of an early reference document. It does not need animation to make its point. The links are the point.

This is exactly why the Ancient Web corpus is useful to webmasters and researchers: old pages sometimes preserve the original organizational memory better than a modern retrospective does.

Explore the 1,967 Ancient Web Domains research list

The W3C page is still one of the quickest ways to watch the Web go from an internal CERN proposal to something recognizably global.

Return to W3C’s Web history timeline

Posted on by

Ancient Web: Retrotechnology and the Art of Keeping Old Computers Useful

Herb Johnson has a useful definition of “retrotechnology”: repair, restoration, and reuse of old technology, especially the digital computers of the 1970s and 1980s.

Visit Retrotechnology

That definition immediately separates the site from a gallery of old beige boxes.

Johnson is interested in keeping the knowledge around the hardware alive: how systems worked, how they are repaired, what documentation survives, who built them, and what older methods can still teach people working with newer machines.

The site says its most substantial computing collection and core interest is S-100 personal-computing equipment from the 1970s and 1980s. There is also extensive material on Digital Research and Gary Kildall, whose work on CP/M sits underneath a large chunk of personal-computer history.

Preservation with a soldering iron

In 2009, Johnson began a series specifically about restoring, repairing, and preserving vintage computers. The pages document his work, other people’s work, and collaborative projects.

That practical emphasis matters.

It is one thing to record that a machine existed. It is another to preserve enough technical context for somebody to diagnose one when a forty-year-old board stops behaving.

The site also openly mixes preservation with commerce. Johnson sells hardware, parts, and services to keep the operation going. There is no need to pretend that historical work exists in a magical world without storage costs, parts costs, shipping, or time.

The homepage was revised in May 2026. That is another good sign of a genuinely living old-Web resource: the presentation may belong to an earlier Internet, but somebody is still maintaining the information.

For a webmaster, Retrotechnology is a useful reminder that depth beats polish when a site owns a narrow subject. A person searching for an S-100 board, CP/M history, an old Macintosh part, or restoration notes does not need a cinematic landing page. They need the person who kept the weird manual.

CacheRat’s Ancient Web research list contains 1,967 domains with this kind of material: specialist knowledge, hobby work, personal collections, and parts of Internet history that rarely fit neatly into modern platforms.

Explore the 1,967 Ancient Web Domains research list

Retrotechnology is not merely about old computers. It is about retaining enough knowledge that “old” does not automatically become “dead.”

Explore Herb Johnson’s Retrotechnology site

Posted on by

Ancient Web: The Handheld Game Museum Never Threw Anything Away

Before a phone could emulate half the history of videogames, portable gaming meant dedicated plastic machines with one job and, frequently, a speaker capable of annoying an entire car.

Visit the Electronic Handheld Game Museum

Rik Morgan’s Handheld Game Museum catalogs those machines by manufacturer, with sections for Atari, Bandai, Casio, Coleco, Entex, Gakken, Mattel, Milton Bradley, Nintendo, Radio Shack, Tiger, Tomy, VTech, and a long tail of companies most people stopped thinking about decades ago.

The front page alone feels like an inventory sheet that escaped onto the Web.

That is a compliment.

The weird details matter

The site goes beyond photographs of game fronts. It has manuals, repair information, internal photographs under “Game Guts!”, books and magazines, rarities, prototypes, and even a section tracking handheld games that appeared in movies.

The update history reaches back through year-by-year pages to 2000. In 2017, Morgan wrote that he had sold his collection to the National Videogame Museum in Frisco, Texas, but the site remained online and continued receiving smaller updates. The homepage records additions as recently as September 2024.

That is a useful preservation story on its own.

A physical collection can change hands while the collector’s accumulated catalog stays public. The photos, model names, manuals, odd variants, and years of notes remain searchable instead of becoming a binder in somebody else’s building.

There is also a fossil from the older social Web sitting right on the homepage: the Handheld Web Ring. The page plainly notes that the ring no longer works and leaves it there “for nostalgia only.”

Correct decision.

For modern researchers, little leftovers like that matter. They show how independent sites used to connect themselves before feeds and recommendation algorithms became the default map of the Internet.

The Handheld Game Museum came from CacheRat’s larger Ancient Web corpus, a list of 1,967 surviving or historically interesting domains for people who would rather explore the Web directly.

Explore the 1,967 Ancient Web Domains research list

If you remember a strange LCD football game, tabletop arcade, watch game, or electronic toy and cannot remember who made the thing, there is a decent chance somebody documented it here.

Browse the Electronic Handheld Game Museum

Posted on by

Ancient Web: Phil’s Old Radios Has Been Online Since 1995

Phil’s Old Radios says it launched in 1995, which means the website itself is now old enough to qualify for the sort of careful restoration work it documents.

Visit Phil’s Old Radios

Philip I. Nelson built the site around antique radios, televisions, and the work required to bring old sets back to life. It includes a large gallery, restoration articles, beginner material, book and magazine recommendations, parts sources, and special features about radio collecting.

The useful part is that the site does not merely show polished pictures of finished hardware.

It explains the work.

Repair knowledge that survives with the machines

Old electronics are a problem for preservation because keeping the physical object is only half the job. Capacitors fail. Insulation deteriorates. Tubes become harder to source. Previous owners make repairs ranging from clever to criminal.

A restoration article captures decisions that a photograph cannot.

That gives Phil’s Old Radios value beyond nostalgia. Somebody approaching an unfamiliar set can learn what to inspect, where to find parts, and what other collectors have already discovered. The beginner section even covers basics such as finding, pricing, testing, and repairing vintage sets.

The website is also a useful example for modern webmasters. Its information architecture is obvious from the front page: Gallery, Beginners, Information, Restoration, Features, Building, Sitemap, Contact, About, News.

No archaeological excavation of the navigation is required before beginning the archaeological excavation of the radios.

The homepage identifies the site as one of the earliest websites devoted to antique radios and TVs and carries a copyright range beginning in 1995. Three decades later, the basic purpose has not changed much. It exists to document a subject and help people who care about it.

That kind of continuity is part of what Ancient Web is looking for.

Phil’s Old Radios is one of the sites pulled from CacheRat’s larger collection of 1,967 classic and independent Web domains assembled for researchers, webmasters, preservation work, and general digging.

Explore the 1,967 Ancient Web Domains research list

If an old chassis full of dust and questionable capacitors sounds like a reasonable afternoon, the restoration section can consume one quickly.

Explore Phil’s Old Radios

Posted on by

Ancient Web: The Jet-Powered Beer Cooler

Some old websites preserve important standards or vanished software. This one preserves a man building a jet engine because his beer was too warm.

Visit the Jet Powered Beer Cooler

Simon Jansen’s page dates the project to 2001, and the premise is laid out with the kind of clarity modern product teams spend three meetings trying to achieve: an all-day session in the shed requires beer, and warm beer is unacceptable.

Ice was apparently insufficient. Refrigeration was apparently too ordinary. Jansen remembered that an expanding liquid draws heat from its surroundings, looked at an LPG cylinder, and arrived at the obvious next step.

Burn the gas through a homemade jet engine.

Extremely unnecessary engineering

The engine used a car turbocharger for the compressor and turbine, with a homemade combustor completing the arrangement. Jansen then used the rapid LPG consumption to chill the cylinder sitting in water with the beer cans.

According to the page, the engine could run up to 100,000 RPM, with exhaust temperatures around 500°C and noise above 125 dBA. After roughly five minutes, the beer went from 11°C to 2°C.

There was also a PVC pipe involved at one point.

It went badly.

The page became popular enough to hit Slashdot in July 2001, prompting Jansen to add a few notes for the sudden audience. They include a recommendation against using the same technology to cool an overclocked AMD processor and the compact update: “Sorry Mum.”

This is exactly the kind of thing the personal Web was good at. A person had an idea, documented the build, took pictures, wrote down what failed, and published it without first determining whether it had a scalable market.

Was it essential knowledge? No.

Was it useful engineering documentation mixed with personality, photographs, physics, and glorious misuse of workshop equipment? Absolutely.

The broader CacheRat Ancient Web list contains 1,967 sites from this same archaeological dig: serious references, technical archives, personal obsessions, and pages that mostly exist because somebody thought the Internet needed them.

Explore the 1,967 Ancient Web Domains research list

The cooler page is still online, complete with the warnings, measurements, bad PVC decision, and plans for an afterburner/sausage sizzler.

Read the Jet Powered Beer Cooler page

Posted on by

Ancient Web: The Coffee Pot That Became a Webcam

Long before webcams became background furniture, a group at the University of Cambridge pointed a camera at a coffee pot because walking down the hall to find an empty pot was apparently too much wasted effort.

Visit the Trojan Room Coffee Machine page

The page that survives is wonderfully matter-of-fact about the whole thing. A video capture board in an Acorn Archimedes grabbed one frame every second. The machine JPEG-encoded the image, and the Web server requested frames through an RPC mechanism developed in the Computer Laboratory.

The coffee pot itself was already being watched before Cambridge had a World Wide Web server. Once the Web arrived, the local convenience hack became something much larger: people outside the building could check the status of a coffee machine they could not possibly drink from.

That is early Web culture in miniature.

A useful problem came first

Nobody started with a marketing plan for “live streaming.” There was coffee in another room. People wanted to know whether there was any left. A camera and network connection solved it.

The surviving page still explains the technical plumbing instead of polishing the story into legend. It mentions MSRPC2, MSNL, the Acorn Archimedes, JPEG encoding, and even the fact that the lights in the Trojan Room were not always switched on.

In November 1994, BBC Radio Cambridgeshire visited to report on the service. The lab did what a lab full of networked computers in 1994 would naturally do: connected a radio to a workstation, relayed the broadcast across the local network, recorded it digitally, and put that online too.

For a webmaster in 2026, the page is a nice reminder that memorable Web projects often begin with an extremely specific problem. The technology serves the thing. The thing does not exist merely to justify a technology stack.

This site came from the larger corpus we are working through for CacheRat’s Ancient Web project. The full list contains 1,967 old, technical, personal, institutional, and generally strange corners of the Web.

Explore the 1,967 Ancient Web Domains research list

The original coffee machine is no longer dispensing status reports to the Internet, but Cambridge kept the page and the explanation.

See the Trojan Room Coffee Machine page

Posted on by

Ancient Web: The Old Calculator Museum

Some websites look old because somebody installed a retro theme. The Old Calculator Web Museum looks old because Rick Bensene has been building the thing since 1997 and never needed to turn it into something else.

Visit the Old Calculator Web Museum

This is a museum of actual electronic and electromechanical calculators, not calculator apps and not a gallery of product shots scraped from somewhere else. Bensene maintains working machines, documentation, advertisements, manuals and a few early computers at a private museum in Oregon. Whenever practical, the hardware shown on the site is kept operational.

That alone would make the collection useful. The website makes it better.

Written the old way

The museum is built in basic HTML. Bensene says he writes the pages himself using vi.

No advertising network. No giant publishing framework. No page assembled by a small industrial accident of JavaScript after you arrive.

He also makes a point of saying that none of the site’s content is AI-generated. On a website that began in 1997, that statement feels less like a trend and more like continuity.

The exhibits cover machines from companies including Wang, Canon, Casio, Commodore, Friden and MITS. Specifications sit next to photographs, historical notes, scans and links into increasingly specific corners of calculator history.

MITS is a good example of why this material matters. Before the company became famous for the Altair 8800, it sold electronic calculator kits. That little connection between calculators and the early personal-computer industry is easy to flatten into a sentence in a timeline. Here it gets room to breathe.

More than hardware

The museum also preserves material that tends to disappear even when the machines survive: advertisements, manuals, technical documents and recollections from people who worked in the electronic-calculator industry during the 1960s and 1970s.

Hardware can survive on a shelf. The stories behind it usually need somebody to keep copying them forward.

For webmasters, the site is also a reminder that useful pages do not expire because their markup is unfashionable. A clear hierarchy, stable links and specific information can remain useful for decades.

This is one stop from CacheRat’s larger Ancient Web research list, a collection of 1,967 old, independent, technical and generally strange corners of the Internet.

Explore the 1,967 Ancient Web Domains research list

The calculator museum is still there, still useful, and still being maintained.

Browse the Old Calculator Web Museum and its historical articles