Category: Ancient Web

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

Posted on by

Ancient Web: Action Grandma Is One Joke and One Page

Action Grandma does not have an investor deck, a content strategy or a compelling reason to exist.

That is most of the appeal.

Visit Action Grandma

The entire Neocities site is essentially a fake-product gag. It announces that this is where visitors can get the latest Action Grandma news, describes Action Grandma as a medicine for grandmothers with arthritis, presents a supposed first user, and ends with an invitation to try it.

There is not enough material here to build a mythology around the site, and doing so would miss the point anyway.

It is a stupid little webpage.

The independent Web needs more of those.

Publishing once required almost no justification

One of the quiet losses in the platform era is the idea that a webpage can be made because somebody had a joke at 11:40 p.m. and knew enough HTML to put it online.

No account needs to gain followers. No recommendation engine has to decide whether Action Grandma has engagement potential. The creator does not need a posting schedule to keep the algorithm warm. The page can simply remain at the same URL, continuing to be exactly as unnecessary tomorrow as it was yesterday.

Neocities is modern infrastructure, so Action Grandma should not be mistaken for a preserved GeoCities page from 1998. What it preserves is a behavior associated with that older personal Web: building a standalone site around an idea far too small to become a conventional product or publication.

That behavior matters because small pages create texture. The Web becomes less interesting when every thought has to fit inside the same five social platforms and inherit the same interface, metrics and advertising machinery.

The original personal Web was full of shrines, jokes, collections, fake companies, fan pages and experiments whose authors were under no obligation to explain the business model. Some lasted twenty years. Some vanished before lunch. Both were legitimate uses of the medium.

CacheRat’s 1,967 Ancient Web Domains research list contains plenty of serious technical and historical resources. Pages like Action Grandma belong in the dig for a different reason: they preserve the Web’s right to be pointless.

Not every URL needs to become a brand.

Sometimes Grandma just needs action.

See Action Grandma for yourself

Posted on by

Ancient Web: Louis Kessler Has Been Keeping His Personal Website Since 1997

A personal homepage becomes something different when the owner keeps maintaining it for nearly thirty years. Eventually the “What’s New” page stops being a changelog and turns into archaeology.

Louis Kessler’s homepage is still online.

Kessler says his official site went live in January 1997, after a trial version appeared in August 1996. The first version already looked like the personal Web in miniature: genealogy, family research, cemetery photography, computer chess, software projects, Jewish history, and pages built with his daughters around cartoons they liked.

More interesting than the surviving homepage is the giant running history attached to it.

The site records old redesigns, traffic counts, awards, search-engine rankings, projects that moved to their own domains, pages that vanished, family sections removed when children grew up, and technologies abandoned after becoming annoying. In April 1999, for example, Kessler notes that he dismantled a Microsoft FrontPage Web and converted it back to a standard text-based site because FrontPage had become too difficult to control.

That sentence alone could be carved above the entrance to a museum of 1990s webmaster trauma.

The numbers also show what a successful personal reference site could become before “creator” was a job title. By the end of 1999, Kessler recorded more than 490,000 page hits. By 2007, the site had passed 3.6 million. His genealogy software links became one of the site’s major draws, eventually getting thousands of monthly visits and leading to a separate review project.

The old update log preserves little Web-era details that normally disappear: joining Web rings, adding JavaScript quizzes, receiving site awards, spam-proofing email addresses, checking rankings across Google, Yahoo, AlltheWeb and MSN, and discovering that a useful page could somehow be popular while still ranking badly because its title used the wrong phrase.

That is not just nostalgia. It is a long first-person record of how an ordinary independent website changed as the Web around it professionalized.

Kessler never needed to freeze the site in 1997 to preserve that history. He simply kept the history attached while moving forward.

Browse Louis Kessler’s surviving personal site and then dig into its old update pages. CacheRat’s 1,967 Ancient Web Domains research list contains more personal sites where decades of incremental changes tell a better Internet story than a screenshot of one old homepage ever could.

Posted on by

Ancient Web: Brian Lemay’s Cartoon Factory Still Teaches Animation the Old Web Way

The old Web is full of sites that are effectively one-person trade schools. Brian Lemay’s Animated Cartoon Factory is one of them.

Visit Brian Lemay’s Animated Cartoon Factory

The homepage describes the site as an educational resource for animation, animated cartoons, cartooning, animation design, layout, storyboarding, computer animation, and training. It is not trying to hide what it is behind a lifestyle brand. It is an animation teacher’s website.

The structure is wonderfully direct. There are learning materials, books, downloadable PDFs, a section on sculpting animation characters with oven-bake clay, an animation-history timeline, and information about video lectures and training.

The current page says it was last updated in January 2022, but much of its visual language belongs to an earlier Internet. That makes the site useful in two directions at once: the subject matter is animation history and technique, while the website itself is a surviving example of how specialists used to package knowledge online.

Lemay’s career also gives the material context. Secondary biographical sources credit him with decades of animation work and teaching, including layout work on television animation. The important point for this site is not celebrity. It is that the pages were built by somebody working in the field and teaching the craft.

A website can be a textbook shelf

The site does something that modern course platforms often make unnecessarily complicated: it exposes the subject directly.

There is no requirement to create an account before discovering what topics are covered. There is no fake scarcity timer. There is no endless funnel of “free masterclass” pages designed mainly to harvest an email address. The visitor can browse the subject, inspect the material, and decide what is useful.

That openness has preservation value too. A page explaining character sculpting or linking an animation-history timeline remains independently discoverable. The knowledge is not buried inside a proprietary course dashboard.

For modern webmasters building educational sites, this is worth remembering. Courses can be products, but the public index itself can still teach something. A good specialist website should leave a trail of useful material even for the person who never buys anything.

CacheRat’s 1,967 Ancient Web Domains research list contains many sites created by people who simply knew a subject and decided the Web was a reasonable place to explain it.

Brian Lemay’s Animated Cartoon Factory is a clean example of that model still hanging around.

Explore the original animation site

Posted on by

Ancient Web: Matt Kane Built a Modern Site Like a 1990 HyperCard Stack

Not everything in an Ancient Web dig has to be ancient. Sometimes a modern developer rebuilds an older computing idea carefully enough to show why it worked in the first place.

Visit Matt Kane’s site

Matt Kane’s homepage looks like a black-and-white HyperCard stack escaped from an early Macintosh and somehow acquired links to GitHub, Bluesky and modern web tooling.

That is deliberate.

Kane’s own “HyperSite” explanation says the design is based on a 1990 HyperCard stack. HyperCard was Apple’s pre-Web hypermedia system built around linked cards, interactive elements and the HyperTalk scripting language. Kane says it was where he first learned to code useful things around 1991, and that the Web therefore felt familiar when he encountered it a few years later.

The site’s footer beginning its copyright line at 1991 is part of the joke and autobiography. This is not a claim that mk.gg has been sitting untouched on the public Web since 1991.

The retro interface is running very modern machinery

The interesting part is how seriously the reconstruction is treated.

Kane says the fonts, icons and graphics are based on originals from the 1980s, including work associated with Susan Kare. He credits recreated Chicago and Geneva fonts, HyperCard graphics, Infinite Mac and other sources that helped him reproduce the environment. Flying toaster animation appears because apparently no reconstruction of old Macintosh culture is legally complete without toast crossing the screen.

Underneath all that, the site is built with Astro and, according to Kane, ships zero client-side JavaScript.

That contrast is useful. Retro web design is often reduced to putting a tiled GIF behind a page and calling it 1996. mk.gg instead borrows an older interaction model: cards, simple controls, monochrome graphics, visible navigation and a small number of understandable elements. The result feels old because the conceptual vocabulary is old, not because the CSS is deliberately broken.

Kane is not a nostalgia-only developer either. His bio describes more than 25 years building for the Web, work on modern frameworks and developer tools, and a current role at Cloudflare. The old interface is being recreated by somebody who knows exactly what contemporary web stacks can do and intentionally chooses restraint on his personal site.

CacheRat’s 1,967 Ancient Web Domains research list mostly points backward. This one is useful in the opposite direction. It demonstrates that old interface ideas can still be technically coherent rather than merely decorative.

HyperCard never became the Web.

But when you look at a page like this, the family resemblance is hard to miss.

Explore mk.gg

Posted on by

Ancient Web: The Thalion Webshrine Rescued an Unreleased English Ambermoon

Thalion Software died in 1994. Its Web afterlife turned out to be much longer than the company itself.

The Thalion Webshrine still survives on the Exotica network.

Thalion was a German developer and publisher formed in 1988 by people with roots in the Atari ST demoscene. Its catalog covered action games, racers, simulations, and RPGs, but the studio never became a commercial giant. By 1994 it was gone.

A few years later, fan Alex Holland began assembling a site dedicated to the company and its games. In a later recollection, Holland said he had met an ex-Thalion developer while working at Argonaut in 1997. That encounter pushed him to collect every scrap of Thalion material he could find and build a website by hand in a text editor on an Amiga.

Then preservation stopped being theoretical.

Thalion’s RPG Ambermoon had shipped commercially in German on the Amiga, but an English translation had been developed and never released before the company collapsed. While Holland was contacting former Thalion people, the rumor of a completed English version kept surfacing. In 1998, Thalion co-founder Erik Simon sent him what Holland described as the only existing disk set of the English beta.

The Webshrine put it back into circulation.

That is an unusually pure example of what fan sites accomplished before game preservation became a recognized discipline. The page was not waiting for a museum, publisher reissue, official anniversary campaign, or preservation foundation. A fan contacted the people who made the games, gathered surviving files and documentation, and put the material where other people could find it.

Contemporary Usenet posts show the site already serving that role around 2000. Players looking for Thalion titles were pointed directly to the Webshrine, and Holland himself posted updated addresses as hosting changed. Years later, the archive still gets cited when researchers discuss cancelled ports, development history, and the English Ambermoon beta.

The preservation chain even continued beyond the original site. Former Thalion developer Jurie Horneman later recovered old Amberstar and Ambermoon source material from backups, with members of the surviving Thalion community helping convert and preserve it.

So this is more than a fan page about a dead company. It became part of the reason pieces of that company still exist.

Visit the Thalion Webshrine and remember that a lot of software history survived because one person cared enough to email the right retired developer. CacheRat’s 1,967 Ancient Web Domains research list contains more sites where fandom quietly turned into preservation before anybody gave it an official name.

Posted on by

Ancient Web: Mario Kart 64 Shortcuts Were Once Traded by Email

Before speedrunning had polished leaderboards, video proof standards, Discord servers, and frame-by-frame analysis, Mario Kart 64 players were already tearing courses apart and passing discoveries around by email.

David Wonn’s comprehensive Mario Kart 64 shortcuts page is still online.

The page reads like field notes from the first generation of serious glitch hunters. Each shortcut has instructions, an inventor when one is known, an approximate discovery date, and often a story explaining how the trick moved through the community.

The dates are the best part. Many discoveries cluster in 1997, essentially right on top of the game’s North American release. Luigi Raceway wall jumps, Kalimari Desert lap tricks, Toad’s Turnpike warps, Frappe Snowland lap skips, Choco Mountain wall jumps, and Wario Stadium abuse were being tested while most players were still learning the tracks normally.

Wonn did not merely collect the final tricks. He kept provenance.

For Luigi Raceway, he credits Jeos Oreamuno and explains how he reconstructed the trick after receiving a description by email. For Toad’s Turnpike, Wonn records the reveal date as March 12, 1997 and links to his original email describing the discovery. Other entries distinguish between tricks he found independently, shortcuts posted first on N64HQ, rumors that had not yet been reproduced, and techniques discovered by several people around the same time.

That is primitive but recognizable speedrunning methodology: establish the claim, reproduce it, separate rumor from confirmed technique, preserve who found what, and argue about whether the resulting time should count.

The page even has a dedicated rumors section. A supposed shortcut could live there while people experimented. If it turned out to be repeatable, it graduated into the legitimate list. If it failed long enough, it disappeared.

No database schema was required. HTML and reputation did the job.

Wonn’s page also preserves a social layer modern gaming databases often flatten away. He remembers local tournaments, Nintendo Power coverage, secret techniques held back for years, arguments over credit, and the excitement of discovering that Lakitu’s rescue logic could be weaponized against the course itself.

This is why an old strategy page can matter after every shortcut has been rediscovered on video. The modern clip shows how. The old page remembers when people first realized it was possible.

Read through the original shortcut archive, especially the comments under each technique. CacheRat’s 1,967 Ancient Web Domains research list has more gaming pages from the period when discoveries traveled from controller to email inbox to hand-written HTML.

Posted on by

Ancient Web: The Werme Family Homepage Is What a Personal Website Used to Mean

A modern personal site is often a portfolio, newsletter signup, or carefully narrowed personal brand. The Werme Family Home Page is from the older school: put the things you care about online and let the homepage grow around them.

Visit the Werme family site

An older version of the homepage cheerfully apologizes for the lack of “a flash animation with cacophonous music” and then immediately starts branching in every direction.

There are family trips. Bicycle rides. Quilting. Colonial reenactment. Weather. Climate. GPS. Geocaching. Sundials. New Hampshire politics. Data visualization. Court issues. Snow depth. Car crashes. Old photographs. Software engineering. A Pamela Smart page. A state-budget browser. A 1978 blizzard account.

This is not bad information architecture so much as a faithful map of a person’s brain.

Ric Werme describes himself as a software engineer and “hacker – in the best sense of the word.” He also says his computing roots reach back to ARPANET work at Carnegie Mellon, where he implemented FTP and Telnet for PDP-10 systems. Elsewhere the site contains weather data and programming experiments, including plots inspired by Edward Tufte’s ideas about information display.

Paula Werme’s sections add legal work, quilting, family history, hiking, and other projects. The site also preserves long-form accounts of trips, including a 2003 cross-country bicycle journey and references to Ric’s 1974 ride from Palo Alto toward Montana.

The homepage is the archive

The important thing here is not any single subject. It is the refusal to split a life into separate branded channels.

A weather observation can sit next to family photographs. A software anecdote can sit next to a bicycle trip. A page about a snowstorm can remain useful twenty years later because nobody decided it was “off topic” and deleted it during a content refresh.

That gives the site enormous archaeological value. Personal websites like this preserve the incidental material that institutional archives rarely collect: what somebody was learning, fixing, arguing about, measuring, photographing, and thinking about at a particular time.

The site even includes a Lynx inspected badge and the declaration that no Microsoft programs are necessary to appreciate it. That is almost a manifesto for durable Web publishing.

For modern webmasters, the lesson is not that every homepage should become this sprawling. It is that longevity often comes from owning the material and keeping old pages addressable. A site can be messy and still be extraordinarily useful.

CacheRat’s 1,967 Ancient Web Domains research list is full of this kind of personal web: strange mixtures that would never survive a social platform’s demand for one topic, one audience, and one kind of post.

Explore the Werme family homepage

Posted on by

Ancient Web: Morel Madness Left a Trail Back to Larry ‘Tree’ Lonik

Sometimes the interesting thing about an old page is that the page itself is starting to disappear while its subject remains scattered across the Web.

Visit the Morel Madness Larry Lonik page

That is the situation with this Morel Madness page. It does not currently retrieve reliably enough to pretend we can inventory every heading and image. What survives around it, however, is unusually specific.

A 2020 episode description from the podcast Bringing Owls to Athens calls MorelMadness.com a primitive fan site dedicated to Larry “Tree” Lonik. That name is not an Internet invention. Publisher records identify Larry Lonik as a Michigan morel enthusiast and author whose books included The Curious Morel. Google Books traces that title back to 1984 and describes it as a guide to hunting, cooking and understanding morels in the Great Lakes region.

So the dead-looking fan page points to a real offline subculture.

Mushroom hunting had its own information underground

Long before every hobby had a subreddit, enthusiasts built pages around the person who knew where the strange thing grew, how to recognize it and what to do with it afterward.

Lonik fit that role unusually well. Later publisher material describes him as a morel-hunting legend who shared his knowledge through books and videos. Contemporary and retrospective references place him inside Michigan’s spring mushroom culture, where morel hunting was not merely food gathering but a seasonal ritual with festivals, lore, recipes and carefully guarded hunting spots.

That context makes the old Morel Madness page more interesting than a celebrity fan page. It represents the Web attaching itself to a regional knowledge tradition that already existed in print, clubs, festivals and word of mouth.

There is an important archaeological limitation here: because the original page is not currently rendering reliably, we should not invent its exact contents from secondary descriptions. We can verify the target of the fandom and the larger body of work around him. The missing page itself remains a gap.

That gap is part of the story.

CacheRat’s 1,967 Ancient Web Domains research list is useful precisely because old URLs often become clues rather than destinations. A failing page can still identify a person, book, community or forgotten obsession worth following into other surviving sources.

The modern Web is very good at answering “what is a morel?”

The older Web was better at introducing you to the guy everybody in Michigan apparently knew as Tree.

Check the original Morel Madness page

Posted on by

Ancient Web: Admiral Furashita Built an Entire Japanese Navy That Never Existed

Some hobby websites preserve history. Others preserve the alternate history somebody spent decades inventing because the real fleets were apparently not complicated enough.

Admiral U. Furashita’s Fleet is still docked at CombinedFleet.com.

The page is a catalog of fictional Imperial Japanese Navy vessels created by Craig Burke for his Victory Through Seapower World War II wargaming universe. They did not exist, and the site makes that clear immediately. What makes the collection wonderful is how seriously the joke is documented after that.

There are battleships, battlecruisers, carriers, aircraft cruisers, monitors, seaplane carriers, submarines, and designs that sound just plausible enough to make an incautious reader wonder whether Imperial Japan really considered building them. Each entry can include displacement, armament, aircraft capacity, speed, line drawings, operational reasoning, and the kind of development history normally reserved for actual naval vessels.

Burke’s pen name is part of the gag. “Admiral U. Furashita” is a phonetic joke built around what people tended to say about his imaginary ships. He explains that he had been hooked on Pacific naval warfare since childhood, after reading his father’s copy of Eugene Burns’s Then There Was One. His father had served aboard USS Enterprise as a junior radar officer during early Pacific operations.

From there, decades of naval reading and wargaming turned into a fleet that existed nowhere except paper, counters, imagination, and eventually HTML.

That last step is what makes the page interesting to CacheRat. The old Web gave deeply committed hobbyists a place to publish material that was too specific for magazines and too strange for conventional reference books. There was no requirement that a subject have a mass audience. If five hundred people on Earth wanted detailed specifications for a fictional Japanese catamaran aircraft carrier, somebody could simply build the page.

And because the site is still there, the surrounding hobby culture survived with it.

The fleet also carries a useful warning for modern browsing: not every authoritative-looking table on the Web describes a real object. Here the fiction is labeled and the joke is intentional. Twenty-five years of copying information out of context can make that distinction surprisingly important.

Go browse Furashita’s impossible fleet and enjoy a Web page built by somebody who clearly knew enough real naval history to make fake naval history properly. CacheRat’s 1,967 Ancient Web Domains research list contains more corners where hobbies got far deeper than any algorithm would have recommended.

Posted on by

Ancient Web: SSQQ Put a Houston Dance Studio Online in 1998 and Never Really Left

Most small-business websites eventually get rebuilt until every trace of their earlier selves disappears. SSQQ went the opposite direction. Its old dance-studio web presence became a giant memory palace and is still online.

Visit SSQQ

SSQQ was a Houston social dance studio run by Rick Archer. Archer’s own history says he began learning computers in 1997 and that the SSQQ website went online in late 1998, built with help from David Schroeder and Sylvia Key.

The timing mattered. Instead of treating the site as a digital business card, SSQQ used the Web for actual operations and community life. The studio published a newsletter, posted class information, collected photographs, documented Halloween parties, kept romance and wedding stories, and eventually offered online registration.

Archer writes that the studio threw away its traditional mailing list and shifted class advertising onto the Internet. By 2001, SSQQ had online registration. In a 2004 site introduction, he described a large operation with Western, Salsa, Swing, Ballroom, Whip/West Coast Swing, and miscellaneous classes.

The website kept expanding because the business kept generating stories.

It became more than a dance-studio site

SSQQ’s archive includes travel, cruises, romances, newsletters, class history, party photos, and long personal writeups. The current homepage says the studio was the largest independent dance studio in the country around the start of the millennium, that Archer sold it in 2010, and that SSQQ Travel eventually organized dozens of group cruises involving thousands of passengers.

Whether a visitor cares about dance is almost beside the point now. The site has become a record of how a local community used the early commercial Web.

That is what makes SSQQ valuable to a webmaster in 2026. The site did not succeed because it looked technologically impressive. It succeeded because the website was tied directly to what the community was doing offline.

A newsletter sent people back to the site. Party photographs gave students a reason to browse after an event. Registration saved people time. Wedding stories turned social proof into permanent local folklore. Travel pages extended the community beyond the building itself.

It is crude by modern design standards and vastly more human than most current small-business templates.

The lesson is straightforward: a useful website can accumulate value for decades if it records the life around the business instead of merely describing the services offered this week.

CacheRat’s 1,967 Ancient Web Domains research list contains many sites that accidentally became historical archives by refusing to delete yesterday every time they redesigned tomorrow.

SSQQ is one of the better examples.

Explore the surviving SSQQ site