Category: Ancient Web

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

Posted on by

Ancient Web: ANI-MATO Put a Professional Animator’s Workshop on the Web

Before every practical skill became a twelve-minute video padded to satisfy an algorithm, professionals put entire workshops on ordinary webpages.

Visit ANI-MATO

ANI-MATO is the site of Finnish professional animator Jan-Eric Nystrom. His homepage describes work spanning animated commercials, title and special-effects sequences for feature films, documentary animation, time-lapse, clay animation and even 3D work.

The interesting part is what sits around the biography.

The site contains practical pages on animation equipment and film technique, including instructions for building a simple animation stand, information on camera stands, film history and a Super 16 camera conversion. These are not vague motivational essays about creativity. They are workshop documents.

One surviving animation-stand page explains how to build a basic rig for attaching a camera and lights so somebody can actually shoot animation. The material dates back to the 2000s and early 2010s, yet people were still linking to its Super 16 conversion instructions in filmmaking discussions in 2024.

That is a good test of old technical writing: does somebody still need it after the website stops looking fashionable?

Expertise without a content funnel

Nystrom says he studied at Helsinki University of Technology and was active in its movie club, Montaasi, where members made a 16mm anamorphic film about making animation. The site grows naturally from that background. It feels less like a publication than the public shelf beside a working animator’s bench.

That kind of material ages differently from news.

A homemade animation stand does not become useless because the CSS is old. A description of film transport or camera modification can remain valuable years after the original hardware leaves stores. In fact, the older and stranger the equipment gets, the more likely the surviving enthusiast page becomes one of the few places that still explains it.

The site’s film-history material has even been cited in later academic work on audiovisual preservation.

That is why CacheRat’s 1,967 Ancient Web Domains research list keeps favoring specialist sites over generic nostalgia pages. The useful old web is full of people who knew how to do something and bothered to write it down.

ANI-MATO is one of those workshops.

The tools may be old. The instructions are still sitting on the bench.

Explore the ANI-MATO animation pages

Posted on by

Ancient Web: 440 International Kept Radio People and Radio History Searchable

Before LinkedIn, Facebook, searchable station databases, and a thousand talent directories, radio people had 440 International.

Visit 440 International

The surviving homepage is almost aggressively compact. It links to “440: Satisfaction,” a where-are-they-now directory; “Those Were the Days,” a daily history feature; favorite radio stories; links; buttons; and background material on the site. There is very little modern furniture around it. The content is the interface.

A 1996 trade-magazine article about 440 gives useful context for what the site was doing when the commercial Web was still young. It described a service built around radio-industry people, station histories, links, and a “Those Were the Days” feature. The name itself came out of radio culture: the article explains that Rolling Stones’ “(I Can’t Get No) Satisfaction” had been record number 440 at a station, and the number stuck.

The “Those Were the Days” pages are still scattered across the site. Pick a date and you can fall into short pieces about broadcasting anniversaries, popular music, television, performers, technology, and general history. One surviving page marks the beginning of commercial FM broadcasting in 1941. Others range much farther outside radio while retaining the voice of a broadcast-ready calendar feature.

The site was useful enough that later U.S. Census Bureau “Profile America” pieces cited 440 International’s calendar material as a source. That does not make every old trivia item sacred scripture, but it shows the site had become part of the information ecosystem well beyond its original radio audience.

The more historically interesting piece is the people directory. Radio careers are unusually difficult to trace because call signs change, stations change formats and owners, on-air names differ from legal names, and talent moves between markets. A specialist directory created by people who understood that world filled a real gap. It preserved the connective tissue around broadcasting that conventional histories often ignore.

The design is also a snapshot of how an industry directory worked before profiles became platform accounts. You did not “build your personal brand” inside 440. You looked people up, followed links, read stories, and moved on with your life.

That makes the site useful to researchers now. It preserves names, old station associations, links, media-history writing, and the vocabulary of radio culture in the period when the industry was first becoming searchable on the Web.

CacheRat’s 1,967 Ancient Web Domains research list contains many specialist directories whose original utility has faded but whose historical utility has increased. 440 International is a textbook example.

The radio business moved on. The old directory is still sitting there, quietly answering questions from another era.

Explore 440 International

Posted on by

Ancient Web: MicroMac Is a Frozen Catalog of Mac Upgrades That Once Made Sense

MicroMac’s website reads like a catalog from an alternate universe where buying a new computer was optional and your existing Macintosh could simply be modified until it became something else.

Visit MicroMac

The surviving pages are packed with accelerator cards, logic-board trade-ups, cache products, memory adapters, processor upgrades, and tools for Macintosh models that have been obsolete for decades. The product list includes names like Carrera040, Thunder, ThunderCache, Speedster040, PowerBoy, PowerMaster, SIMMdoubler, BigSIMM, and MacOpener.

Those names are fun. The technical details are better.

A Color Classic or Performa could be pushed into territory Apple never officially promised. One MicroMac FAQ describes replacing a logic board with an LC575 board, running 68040 processors at upgraded speeds, dealing with VRAM timing, installing System 7.1 through Mac OS 8, and going beyond Apple’s stated memory limit with MicroMac’s BigSIMM. The company’s product pages offered logic-board swaps for long lists of Performa machines and accelerators for systems built around 68030, 68040, PowerPC 601, 603, and 604 processors.

The software page is another small time capsule. Accelerator control panels and extensions were distributed as StuffIt archives and BinHex files. Utilities such as MODE32, Speedometer, and machine-ID tools sit alongside MicroMac’s own control software. Instructions casually assume the reader knows what an extension, control panel, System Enabler, and PRAM are.

This was a real hardware economy. Owners were not merely adding RAM or swapping storage. Vendors sold ways to change CPU speed, bus speed, cache behavior, processor family, logic board, video capability, and even the physical expansion possibilities of compact Macs.

The surviving site appears largely frozen around the turn of the millennium. Product pages still list telephone numbers, fax numbers, email sales contacts, and prices as if somebody might call tomorrow about a 68040 accelerator. That frozen commercial context makes the site especially useful. A modern collector database can tell you that a particular card existed. MicroMac shows how it was marketed, which machines it claimed to support, what problems customers were expected to encounter, and how the company explained the upgrade to buyers at the time.

That is the sort of detail CacheRat keeps finding in its 1,967 Ancient Web Domains research list. Old vendor sites can become accidental service manuals once the vendor, product category, and assumptions behind the products disappear.

MicroMac is not merely a store that forgot to close its website. It is documentation for a period when “Mac upgrade” could mean turning the computer you owned into a machine Apple never sold.

Browse the surviving MicroMac catalog

Posted on by

Ancient Web: Android Arts Is a One-Person Internet of Games, Hardware, and Drawings

Android Arts is what happens when somebody refuses to separate art, hardware, game design and personal opinion into different brands.

Visit Android Arts

The site belongs to Arne Niklas Jansson and sprawls across an absurd range of subjects: Zelda, Metroid, Wing Commander, Star Trek, Doctor Who, Warhammer 40K, Traveller, pixel art, art tutorials, electronics, Amiga hardware, console redesigns, KiCAD, tabletop concepts, game-development experiments and piles of sketches that do not need a commercial reason to exist.

There is even a 2002 archive linked from the front page.

This is not a portfolio in the modern sense. It is closer to a workshop with every drawer left open.

The homepage is the filing cabinet

Android Arts is especially useful because the homepage itself exposes the site’s internal history. Old projects remain linked beside newer ones. Failed or unfinished ideas are not necessarily cleaned away. A Famicube hardware concept can sit near a pixel-art tutorial, a Master of Orion Amiga port idea, a Star Trek redesign and an electronics project.

That mess is productive.

You can see how one interest bleeds into another: old games become art exercises, art exercises become hardware speculation, hardware becomes interface design, and all of it feeds back into new drawings.

The site’s philosophy is also stated unusually plainly. Jansson argues against an internet dominated by a few giant sites and says he would welcome the return of small quirky enthusiast pages. He notes that the site renders in the text-oriented Links2 browser. He also explains why the domain intentionally uses ordinary HTTP rather than HTTPS: the site has no scripts, ads, tracking or merchant functions, and he has little interest in certificate maintenance for a static personal site.

You do not have to agree with that security choice to appreciate how perfectly it fits the project.

This is a website whose owner still thinks about what the web itself should be.

CacheRat’s 1,967 Ancient Web Domains research list is largely a hunt for exactly this kind of material: independent sites where expertise and personality accumulated over years without being reformatted into disposable posts.

Android Arts has survived because it is not merely a portfolio.

It is Arne’s own little internet, and he kept adding rooms.

Dig through Android Arts

Posted on by

Ancient Web: The Victorian Web Started Before the Web

The Victorian Web is old enough that calling it an “old website” is slightly misleading. Its core project began before the World Wide Web existed.

Visit The Victorian Web

George P. Landow and collaborators at Brown University developed the project’s early hypertext material in the late 1980s using systems such as Intermedia and later Storyspace. The goal was not simply to digitize Victorian texts. It was to connect literature, history, art, religion, technology, politics, economics, and biography through links so students could move between related ideas instead of following a single textbook path.

That history explains the site’s structure. The Victorian Web is not arranged like a modern encyclopedia with one canonical article per topic. It grew as a network of what the project called lexias: smaller documents tied together by contextual links.

The transition to the Web is its own archaeological story. Project credits describe the collection growing dramatically in 1993 as material was imported and linked. In 1994 students added more than a hundred new documents. Marc Zbyszynski then began manually recreating in HTML the link menus that Storyspace had generated automatically, and in 1995 Landow built the HTML version’s layout and icons.

That means the site did not merely adopt hypertext because the Web made linking convenient. Hypertext was already the intellectual model. HTML became the delivery mechanism.

The result is enormous. The site covers authors such as Dickens, Ruskin, Tennyson, the Brontës, and many others, but it also branches into architecture, painting, sculpture, gender, religion, imperial history, social conditions, science, railways, and technologies of the nineteenth century. Whole books and primary documents sit beside essays, images, biographies, and contextual notes.

For a modern webmaster, the interesting lesson is that The Victorian Web was designed around relationships between information rather than feeds, engagement metrics, or a search box as the primary navigation device. Links are not decorative citations at the bottom. They are the architecture.

That architecture also makes the site a useful historical artifact in two directions at once. It documents the nineteenth century while preserving ideas about digital scholarship from the late twentieth century.

Researchers can use it for the Victorian material itself, but people studying hypertext, educational computing, early Web publishing, or digital humanities have another reason to care: the website embodies a design philosophy that predates the browser era.

The larger 1,967 Ancient Web Domains research list contains many old references. Very few can honestly say their information architecture was being worked out before the Web arrived.

Explore The Victorian Web

Posted on by

Ancient Web: Per-Bast Built an Egyptian Cat Goddess Shrine With IRC and XHTML

There is something extremely old-web about building an entire corner of your site around one Egyptian cat goddess and then attaching an IRC channel to it.

Visit Per-Bast

Per-Bast describes itself as an online shrine to Bast, the ancient Egyptian goddess commonly associated with cats. The author explains the name Per-Bast as the Egyptian city later known to the Greeks as Bubastis and uses the page to collect history, religious interpretation, links and personal devotional practice.

The important distinction is that this is not presented as a neutral academic encyclopedia.

The author openly identifies personal beliefs, recommends prayer and offerings, and separates those beliefs from material described as based on historical records. That makes the page easier to read honestly: it is a believer’s shrine with historical material, not an institution pretending to have no point of view.

A shrine that is also a website

The surrounding site architecture is half the fun.

The navigation includes places called The Portal, UFO, Town of ZZT, The Observatory, Library of Babel, Haunted House, Reptile House, KoshkaIRC, Wildcat Den and The Dock. At the bottom sit tiny buttons for things including “Best viewed on a real computer,” Autism Acceptance, Pale Moon, and XHTML/CSS validation.

Visitors are invited to leave comments in a guestbook or stop by the site’s IRC network.

That is a radically different social model from the modern web. The page is public, but community happens through deliberately chosen side channels rather than a universal comment widget owned by a platform.

The Bast page itself discusses the deity’s protective roles, cat imagery, the ancient city of Per-Bast, Herodotus’s account of festivals there, modern Kemetic practice, and the author’s own offering habits. Historical or religious claims on a devotional page should obviously be checked against specialist sources before being treated as settled scholarship.

But as a web artifact, the page is wonderfully clear about what it is.

CacheRat’s 1,967 Ancient Web Domains research list keeps turning up sites like this because the independent web gave narrow obsessions room to become complete environments.

Per-Bast is not a post about Bast.

It is a place about Bast.

The distinction is the whole point.

Enter Per-Bast

Posted on by

Ancient Web: Chicon 2000 Left an Entire Worldcon Online

Most event websites are disposable machinery. They sell tickets, explain where to park, survive the weekend, and eventually get replaced by a sponsor page. The website for Chicon 2000 somehow missed that memo.

Visit the Chicon 2000 website

Chicon 2000 was the 58th World Science Fiction Convention, held in Chicago from August 31 through September 4, 2000. The surviving site still contains the structure built to organize thousands of science-fiction fans around that event: registration information, hotels, travel pages, guest biographies, schedules, committee information, art-show material, masquerade information, business-meeting documents, and piles of fan culture that would normally vanish once everybody checked out of the hotel.

The homepage itself is an artifact. It offers frames and no-frames versions, uses table-based navigation, tiny image buttons, bright link boxes, an email subscription form, and a Chicago skyline image. One surviving version cheerfully reports that it is “Day 635, Chicon 2000,” which is a pretty good indication that nobody bothered teaching the event website how to stop counting.

The useful material goes far beyond design nostalgia. The site names guests including Ben Bova, Bob Eggleton, Jim Baen, Bob and Anne Passovoy, and toastmaster Harry Turtledove. The basic-information pages explain what a Worldcon was to a newcomer, how to budget for the trip, where people would stay, and what five days of panels, readings, autographs, and late-night activity looked like from the organizers’ point of view.

Then there are the stranger records that make a preserved event site worth keeping. Chicon’s filk section includes a songbook table of contents full of convention songs and fandom in-jokes. The site also preserves World Science Fiction Society constitutional material, including the rules governing Hugo Awards and future Worldcon selection. Older bid-era pages even document promotional trading cards produced while Chicago was campaigning to host the convention.

That combination matters. A later encyclopedia can tell you who won an award or how many people attended. The original website shows how the event explained itself before it happened, what organizers thought attendees needed to know, how fandom spoke to itself, and what infrastructure surrounded the formal program.

For researchers studying science-fiction fandom, convention culture, early event websites, or community organization before social media absorbed most of the logistics, this is primary-source material hiding in plain sight.

CacheRat’s 1,967 Ancient Web Domains research list contains plenty of similarly frozen event sites. They are often more valuable twenty years later than anybody designing them expected.

Chicon ended in September 2000. Its Web table is apparently still set.

Explore the surviving Chicon 2000 site

Posted on by

Ancient Web: Shorthand Shorthand Shorthand Keeps Pen Stenography Writing at Speed

There is a website that has spent more than a quarter-century insisting that pen shorthand is not dead, and it has the practice dictation to prove it.

Visit Shorthand Shorthand Shorthand

The site was launched in 1999 and is devoted to systems including Gregg, Pitman, and Speedwriting. Its host identifies as a Gregg Anniversary writer but explicitly welcomes users of other systems. The home page explains shorthand as rapid writing and places it in the office culture where stenographers and secretaries took dictation, transcribed letters, and made business communication move before word processors made that workflow largely invisible.

The useful material is practical rather than merely nostalgic. There are sections for learning shorthand, identifying which Gregg version someone writes, comparing shorthand with recording devices, building writing speed, and understanding modern systems. The dictation section provides practice material at 60, 80, 100, 120, and 140 words per minute.

That last section is particularly good Web archaeology because it preserves not just information about shorthand but the training process itself. The site reproduces old-style business dictation and explains testing conventions: valid pieces needed to run for a minimum amount of time, older Gregg tests could last five or ten minutes, and students were expected to check outlines and drill them before attempting speed work.

It is easy to treat shorthand as a quaint pre-digital office technology. The site resists that framing. It describes pen stenography as a personal note-taking tool, a hobby, and—in some countries and professions—something still used seriously. Its continued updates reinforce the point. This is not a museum page abandoned in 2001; the site was still being updated in 2026.

The design is old enough to feel like part of the subject. It is mostly text, simple links, small graphics, and pages organized around what a shorthand student actually needs. There is no account system, no course funnel, no “AI note assistant.” You click a link and practice writing faster.

For researchers, the site is useful because it bridges office history and living practice. It preserves terminology, teaching assumptions, speed-building exercises, and the distinctions between shorthand systems that broad historical summaries often flatten away.

It also demonstrates something modern web publishing routinely forgets: useful reference material does not need to be visually reinvented every three years. If the audience knows what it wants, durable navigation beats fashionable furniture.

CacheRat’s 1,967 Ancient Web Domains research list keeps surfacing sites like this—small specialist resources that remain useful because somebody never stopped caring about the subject.

Practice some shorthand at the original site

Posted on by

Ancient Web: Crystalinks’ Colosseum Page Still Calls a 2007 Event Recent

Sometimes you can date an old webpage without checking the copyright footer. The prose accidentally does it for you.

Visit the Crystalinks Colosseum page

Crystalinks’ page on the Roman Colosseum is enormous. It moves through construction, imperial history, medieval reuse, restoration, architecture, seating, spectacles, Christian traditions and even the plant species that have grown inside the ruins.

Then, deep in the page, it says “Most recently” before describing the Colosseum being illuminated after New Jersey abolished capital punishment in December 2007.

There it is. A little archaeological marker embedded directly in the sediment.

The page is still online, still readable, and still carrying the temporal assumptions of the era when that paragraph was written.

Static pages preserve their own age

Modern reference sites are constantly rewritten. Templates change. Dates get normalized. old phrasing disappears. A static personal page can preserve something different: not just information about its subject, but evidence of when the page itself stopped evolving.

Crystalinks has plenty of those traces.

There are character-encoding scars in the text. Long sections sit one after another without modern card layouts or collapsed accordions. Navigation at the bottom leads into Roman architecture, ancient civilizations, an alphabetical index, the Crystalinks homepage, a psychic-reading page, a book page, donations and advertising.

That mixture is pure old web. Encyclopedia, personal cosmology, commercial links and historical reference material all share the same domain because one site owner decided they did.

The Colosseum article itself should be treated as a secondary web reference, not as an academic source of record. Some claims are debated, and the page plainly contains material frozen from an older state. That is part of what makes it useful to examine as web history.

CacheRat’s 1,967 Ancient Web Domains research list is full of sites where the page is interesting for two reasons at once: what it says, and what its survival tells you about how people once published information.

Crystalinks did not need a headless CMS to hold a few thousand words about Rome.

It needed one long HTML file and apparently no compelling reason to rewrite “most recently” for nearly twenty years.

Perfect.

Read the old Crystalinks Colosseum page

Posted on by

Ancient Web: The Castlevania Dungeon Still Wants 800×600 and 64K Colors

The preserved CastleVania Dungeon does not ease you into the experience. It tells you the site is best viewed at 800×600 with 64K or more colors, offers frames or no frames, and recommends downloading a TrueType font called “Halloween” for added effect.

Enter the preserved CastleVania Dungeon

That is not retro styling added later. This was how ambitious fan sites actually introduced themselves.

Kurt Kalata’s original Castlevania site goes back to 1997. An early surviving page from March of that year lists the NES, Game Boy, SNES, Genesis, MSX, PC Engine, and arcade branches of the series alongside music, cheats, emulation, and whatever else could be assembled by one fan with server space and patience. The page counter itself says visitors had been arriving since February 12, 1997.

By the preserved 2005 version, the Dungeon had become much larger. Individual games received reviews, manuals, stage information, walkthroughs, passwords, secrets, endings, foreign-version notes, artwork, screenshots, and strange regional details. The Haunted Castle page, for example, includes cabinet photographs, flyers, jumper settings, instruction sheets, and commentary about an arcade Castlevania game the author says he did not even know existed when he started the site.

That is one of the pleasures of old fan sites: you can watch the author learn in public.

The Dungeon also had columns, fan contributions, music, forums, and long arguments about topics like emulation. Its “What’s New” page eventually explains that the site was heading into hiatus while Kalata worked on a newer project called Hardcore Gaming 101. That makes the preserved Dungeon more than a dead fan page. It is part of the prehistory of a later retro-gaming publication and book project.

Hardcore Gaming 101 eventually published a Castlevania book based partly on material from the Dungeon, heavily revised and expanded. The old website remains useful because it shows the earlier form of that work: how the information was organized for fans before it became a book.

Modern franchise wikis are broader and easier to search. They are also usually scrubbed of the individual author’s voice, page-design decisions, mailbag arguments, obsolete browser assumptions, and community archaeology that make a site like this historically interesting.

That is why old fan sites deserve preservation as sites, not just as piles of copied facts. CacheRat’s 1,967 Ancient Web Domains research list contains plenty of similar cases where the original presentation is half the artifact.

The Dungeon is still navigable. The Halloween font remains optional. Your monitor can probably handle 800×600.

Browse the CastleVania Dungeon backup