Category: Ancient Web

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

Posted on by

Ancient Web: Kitty’s PageWorks Taught the Web to Wiggle in 1996

Before websites had design systems, they had people making tiny animated mailboxes one GIF at a time.

Kitty’s PageWorks is one of those workshops still sitting where it was left.

Visit Kitty’s PageWorks

The site says PageWorks was established on February 1, 1996 by Kitty Roach. Its surviving map is a catalog of what web graphics once meant: animated images, buttons and icons, holiday graphics, kids’ pages, e-cards, transparent-background tips, banners, character art, and tutorials explaining how the images were made.

There is even a section on the dos and don’ts of copyright and direct linking.

That last part is wonderfully period-specific. Early webmasters routinely discovered that embedding somebody else’s image directly from their server meant borrowing not just the picture but the other person’s bandwidth bill.

PageWorks taught the culture around the graphics as much as the graphics themselves.

The web before stock libraries swallowed everything

Today, a site owner can buy an icon pack, drop in an SVG library, use an animation framework, or ask software to generate an entire visual system.

In 1996, somebody had to make the dancing letter.

PageWorks also makes the economics visible. Roach advertised freelance animation and web-design work at $35 an hour, offered licensing for original graphics, and maintained pages for commercial banners and buttons.

The site map recommends Netscape Navigator or Microsoft Explorer with 250 colors or more.

That one line dates the experience more effectively than a copyright footer ever could.

The web was still being designed around machines where available color depth, browser brand, download time, and file size were immediate creative constraints. Small GIFs were not merely a nostalgic aesthetic. They were a rational technology for making a page move without making the visitor wait forever.

The surviving site is useful precisely because it preserves the working environment around early web art. It is not a retrospective explaining what people supposedly did. It is the actual shop floor, with pricing, file advice, links, licensing, animation examples, and browser recommendations still attached.

CacheRat’s 1,967 Ancient Web Domains research list keeps turning up pages like this because the practical web often ages better than polished corporate redesigns.

A corporate homepage gets replaced.

A lady teaching you how to make a transparent GIF in 1996 might accidentally survive thirty years.

Explore Kitty’s PageWorks

Posted on by

Ancient Web: Nindestruct Put Its 1998 Nine Inch Nails Geocities Site Back Online

A lot of old fan sites survive only as screenshots. Nindestruct went further and put its own 1998 Geocities incarnation back on the Web as something you can actually click through.

Visit the restored 1998 Nindestruct site

The modern Nindestruct project describes that original page as its Geocities-era version and has gradually expanded into a broader Nine Inch Nails preservation site. The Web Archive section restores old fan pages, official subsites and material from what fans once casually called the “NINternet.” Some Flash-era pages are even run through Ruffle so they can function in current browsers without asking the visitor to install a plugin that died years ago.

The preserved material is valuable because it captures how music fandom worked before social platforms centralized it. A 1990s Nine Inch Nails page might contain magazine transcriptions, imported CD information, fan art, MIDI files, WAV downloads, fonts, bootleg lists, chat rooms, newsgroup-style forums and enormous hand-curated link pages. One restored page advertises a different full-length live song every Sunday, distributed as a zipped WAV file. Another describes “ten thousand links” to other Nine Inch Nails resources.

That sounds excessive until you remember that the links were the network. There was no universal profile system, no algorithmic feed and no platform account connecting everybody. A fan site earned its usefulness partly by knowing where the other fan sites were.

Nindestruct also preserves contemporary news pages. Old 1998 entries read like a mixture of music reporting, rumor control, fan correspondence and personal weblogging. They record what fans knew at the time, not what a later encyclopedia decided was important. That distinction makes them useful primary artifacts even when the writing is messy or the information was later corrected.

For a webmaster, the restoration work is just as interesting as the fandom. Preserving an old Web experience sometimes means more than saving HTML. Flash, image maps, obsolete media formats, broken external links and vanished domains all need different treatment. Nindestruct’s approach keeps enough of the original behavior to make the pages understandable.

The broader 1,967 Ancient Web Domains research list contains plenty of dead fan neighborhoods. This one is unusual because somebody came back with tools and rebuilt part of the street.

Explore Nindestruct’s 1998 Web archive

Posted on by

Ancient Web: Keno.org Still Carries a 1990s Rolling Stones Message Board

Keno.org is the kind of site that accidentally became a museum of its own operating history. It still contains a Rolling Stones fan community built around old-school message boards, pages from a retired Web-design business, personal material, archives, and enough hand-maintained navigation to show how one person’s domain could become a small town.

Visit Keno.org

The most interesting piece is Gasland, the site’s Rolling Stones message-board community. Keno’s own history explains that it grew out of the collapse of earlier Stones forums and was built using Matt’s old wwwboard, a free CGI message-board program. That detail matters because it places the site in the era before Reddit, Facebook groups and hosted forum platforms absorbed this kind of fan conversation.

The board eventually moved beyond the original CGI setup, but the archives survived. A current archive page still calls it “The Oldest Stones Message Board on the Net” and exposes year and month filters, text search, message IDs, thread IDs and old posts. Gasland itself closed in 2024, yet the material remains readable instead of vanishing with the community.

Keno.org also preserves its former commercial side. A Web-design page carries a copyright beginning in 1998 and lists the services that once defined a small webmaster business: site design, redesign, domain registration, hosting, search-engine placement, CGI/Perl, JavaScript, ASP, database work and meta tags. The page now plainly says the webmaster retired from Web design in 2014.

That combination makes the domain more interesting than a normal fan site. You can see both the culture and the machinery: fandom, hand-built pages, CGI boards, search-engine promotion, hosting and the assumption that one webmaster could reasonably operate all of it.

Modern platforms make communities easier to start, but they also tend to erase the infrastructure from view. Keno.org shows the opposite model. The webmaster, domain, archive and community history all remain tangled together, and that is precisely why the old material still has context.

The larger 1,967 Ancient Web Domains research list contains many sites like this where the page history is as useful as the nominal subject.

Keno.org is still online, complete with old navigation, board archives and the remains of a Web business that started when “webmaster” could genuinely describe one person’s whole job.

Explore Keno.org

Posted on by

Ancient Web: This Anti-Santa Page Turned Christmas Into Spiritual Warfare

There was a period when building a website meant you could finally publish the argument everyone at church had already heard three times.

Some people used that power cautiously.

Others investigated Santa Claus.

Visit “Santa Claus: The Great Imposter”

The page, credited to Terry Watkins of Dial-the-Truth Ministries, argues that Santa Claus is not an innocent Christmas character but part of a spiritual attack on children.

This is not a short opinion piece.

It is a full dossier.

The page moves through Bible passages, child evangelism, Satan, occultism, popular culture, Santa traditions, and eventually the famous linguistic observation that SANTA can be rearranged into SATAN.

That anagram is treated not as a mildly amusing coincidence but as evidence worth taking seriously.

The page also name-checks Britney Spears, *NSYNC, and Harry Potter as examples of perceived attacks on children. That combination places the rhetoric squarely inside the cultural atmosphere of the late 1990s and early 2000s, when evangelical panic, chain-email logic, and the rapidly expanding web frequently collided.

Before social media, every argument needed its own URL

What makes the page interesting now is not whether its conclusions are convincing.

They are claims made by the author, and some of its historical and occult connections are interpretive rather than established fact.

The artifact matters because it shows how ideological publishing worked before Facebook feeds, YouTube channels, and algorithmic outrage machines made distribution automatic.

If you wanted to persuade strangers that Santa was spiritually dangerous, you acquired hosting, wrote hundreds of paragraphs, formatted Bible verses, added images, cited books, and hoped somebody found the page through a search engine or forwarded the URL.

The result was strangely durable.

A social-media rant can disappear into a timeline within hours. A static doctrinal page can sit at the same path for decades, preserving not just the argument but the exact tone of the era that produced it.

That is one reason CacheRat’s 1,967 Ancient Web Domains research list contains material far beyond tasteful museum pieces. Web archaeology includes technical manuals, fan pages, serious scholarship, and occasionally somebody spending an extraordinary amount of HTML on the proposition that rearranging one letter exposes Christmas.

The old web gave everyone a printing press.

It did not install an editor.

Read the original anti-Santa page

Posted on by

Ancient Web: Victor Bryant Turned Decades of Ceramic History Into Web Tutorials

Ceramic Studies looks like somebody decided that the history of pottery was too complicated to leave to a few museum captions and then spent years building the missing textbook directly on the Web.

Visit Ceramic Studies

The tutorials are the work of Victor Bryant, whose background makes the site’s strange breadth make sense. His profile describes training in chemistry and geology, formal art study at St Martin’s School of Art, teaching and museum lecturing, and later study of studio pottery. The result is a site that treats a ceramic object as several things at once: an artwork, an archaeological object, a chemical process and a piece of manufacturing technology.

The pages move through long stretches of ceramic history rather than stopping at a single period. A Roman pottery tutorial discusses coarse ware, red gloss ware, relief decoration, lead glazes, fritware and mass-production techniques. The China material follows developments from early dynasties through high-fired wares, celadons, sancai glazes and porcelain. Bryant regularly inserts “Potter’s Notes” alongside the history, which is a useful reminder that these objects were made by people solving physical problems with clay, heat, minerals and kilns.

The construction is just as revealing as the subject. Many pages still expect frames, browser Back buttons, thumbnail images and separate image panes. Instructions explain where the reader should click and which scrollbar to use. Some pages carry revision dates from the early and mid-2000s, while copyright notices reach back to 1994.

That design would be considered awkward now. It also works remarkably well as a preservation container because the content is mostly ordinary HTML, text and images. Twenty years later, a reader can still follow the argument without installing an app, accepting a cookie wall or discovering that an interactive presentation platform was discontinued in 2017.

For researchers, Ceramic Studies is useful because it retains Bryant’s organization and terminology rather than flattening the material into a generic encyclopedia entry. For webmasters, it is a demonstration of what happens when a knowledgeable person builds around subject structure instead of a content-management template.

Sites like this are exactly why CacheRat keeps a larger 1,967 Ancient Web Domains research list. The valuable part is often not flashy history. It is one expert quietly placing years of accumulated knowledge where strangers can still find it.

Explore Victor Bryant’s Ceramic Studies tutorials

Posted on by

Ancient Web: BoxArtDen Treats Packaging Art Like Museum Material

A cardboard box is designed to be thrown away.

That is exactly why somebody eventually has to become weirdly serious about saving pictures of it.

Visit the BoxArtDen gallery

BoxArtDen is a free, non-profit gallery organized around material that commercial history tends to discard first: package illustrations, model-kit art, brochures, posters, advertisements, restoration examples, artist information, and even “box art that never was.”

The gallery structure matters.

This is not merely a folder of JPEGs dumped onto a server. The site separates original collections from reference material and restoration work, making the images useful for people trying to identify packaging, compare versions, study artists, or reconstruct what a damaged box originally looked like.

That is a different mission from product photography.

A modern store photographs a package because the package helps sell the object inside it. A preservation site photographs the package because the package itself became the object worth studying.

Commercial trash becomes design history

Packaging art sits in an awkward historical category.

It can be culturally important while remaining physically disposable. Boxes get crushed. Instruction sheets disappear. Posters fade. Retail displays are thrown out when the promotion ends. Even collectors who save the underlying object may discard the packaging to save space.

Then thirty years pass and somebody wants to know which illustration appeared on a particular release, who painted it, what typography was used, or how one regional edition differed from another.

Suddenly the garbage matters.

The web was exceptionally good at creating volunteer museums for material like this because a collector did not need permission from an institution to begin cataloging. A scanner, some hosting, and an unreasonable tolerance for classification could produce a reference collection available worldwide.

The downside is that these archives can be almost as fragile as the cardboard they document. Domains expire. Gallery software breaks. Image hosts vanish. Search engines stop surfacing old specialist sites.

That is why CacheRat’s 1,967 Ancient Web Domains research list keeps turning up collections whose real value is not flashy design but concentrated human sorting.

BoxArtDen preserves the layer of commercial art people were originally expected not to preserve.

Which is usually how archaeology starts.

Browse the BoxArtDen gallery

Posted on by

Ancient Web: InterGuru Still Converts Address Books From Eudora, Pine, Elm, and AOL

Moving an address book from one mail program to another used to be enough of a technical problem that somebody built an entire Web service around it. InterGuru says it has been doing exactly that since 1996, and the converter is still sitting there surrounded by the names of mail programs that now read like a software archaeology checklist.

Visit InterGuru’s E-Mail Address Book Conversions

The page converts contact lists between formats used by Eudora, Pine, Elm, Pegasus, CompuServe, Spry, Lotus cc:Mail, Lotus Notes, Microsoft Internet Mail, Netscape, AOL, Claris Emailer, T-Online, LDIF files, CSV-style databases and several other formats.

That list is the interesting part.

Today contacts are usually buried inside a cloud account and synchronized behind the scenes. In the 1990s and early 2000s, changing mail clients could mean figuring out which program stored names in which proprietary file, exporting what you could, and praying the destination application understood it. A Web-based converter was not a novelty. It solved an irritating migration problem that ordinary users actually encountered.

InterGuru’s interface still reflects that world. You choose a source format, choose a destination format, push Continue, and follow the instructions. There is no attempt to disguise the machinery behind a giant onboarding flow. The page assumes you know whether your contacts came from Pine, Eudora, Netscape 3, Lotus Notes, or a Unix .mailrc file.

One surviving subpage even preserves a 1998 update about a small utility for extracting names and addresses from Microsoft’s Exchange personal address book and generating text files suitable for import elsewhere. The language is casual, practical and unmistakably from the era of people swapping little conversion programs because somebody needed one.

For webmasters, the durability lesson is almost embarrassingly simple: a narrowly useful page can outlive the software ecosystem it served. InterGuru is now partly a reference document for dead mail clients, but the original problem is still understandable because the site names the formats, applications and workflow precisely.

That kind of specificity is why the larger 1,967 Ancient Web Domains research list is useful. Old pages often preserve the connective tissue between technologies that modern histories skip.

InterGuru is still online, still recognizable, and still asking questions like whether your address book came from Netscape 3.

Explore InterGuru

Posted on by

Ancient Web: Virtue Ventures Taught the Web How to Make Medieval Clothes

Cynthia Virtue’s website is what happens when a serious hobbyist is allowed to build outward in every direction instead of being forced into a content category. Virtue Ventures contains medieval clothing research, fabric advice, historically inspired home decoration, SCA material, family pages, odd products, and enough personality to make the whole thing unmistakably personal.

Visit Virtue Ventures

The strongest material is the collection of Medieval Clothing Articles, focused largely on clothing and accessories of the High Middle Ages in Europe. Virtue was deeply involved with the Society for Creative Anachronism and says she had been active in it since 1983. Her own biography describes costuming as a major interest, especially the gloriously impractical hats of the period, along with years of teaching, research, and service.

This was not a site that merely collected pictures of medieval clothes. The articles were written for people trying to understand and make things. Other historical-costuming pages continued linking to Virtue’s material years later, which is one of the better tests of whether an old tutorial actually became useful infrastructure.

The surrounding pages make the site even more representative of its era. There is Fabric for Bachelors, written for intelligent people who somehow reached adulthood without learning how clothing material works. There is A Modern Medieval Home Companion. There is an SCA feast survey. There are personal notes, family material, and a small commercial corner featuring designs and jokes that make considerably more sense after spending time on the rest of the site.

The homepage notes that the site had reached as many as 90,000 page views per month by February 2005. That is the author’s own figure, but it gives some scale to a project that still feels like somebody’s personal notebook rather than a media property.

Virtue also left the interruptions visible. After the birth of twins in 2002, the site went through a long update hiatus before she began adding material again. That is another feature modern publishing systems tend to hide: old personal sites often show the actual life of the person maintaining them. The archive does not pretend content was produced by an immortal department called Editorial.

For a modern webmaster, Virtue Ventures demonstrates how subject authority can grow from specificity rather than branding. The useful pages explain things the author had actually studied, made, taught, and argued about. Search engines may have changed repeatedly, but other people kept linking to those pages because they solved real problems.

It is the same reason the CacheRat 1,967 Ancient Web Domains research list remains worth rummaging through. A personal site can look miscellaneous from the outside while containing the best surviving explanation of one very specific thing.

Virtue Ventures is still exactly that sort of place.

Explore Cynthia Virtue’s medieval clothing and other articles

Posted on by

Ancient Web: Someone Photographed GeoCities’ Server Cage in 1999

GeoCities is usually remembered as blinking text, neighborhood names, MIDI files, and homepages about somebody’s cat.

Someone also took a camera into the room where all of that actually lived.

Visit the GeoCities Cage at Exodus Communications

The page says the photographs were taken on May 7, 1999 inside the GeoCities cage at Exodus Communications’ “Wyatt” data center.

That turns an abstract web service into a room full of hardware.

The photos and captions describe rows of Sun Ultra 2 systems attached to MTI JBOD storage arrays. The author estimates roughly 120 clusters and about 12 TB of storage. SCSI cables, Ethernet, console lines, and power snake through the racks. Four Cisco Catalyst 5500 switches handle public and private network traffic.

The page even explains a change in GeoCities’ architecture.

Earlier systems depended heavily on disks attached directly to individual servers. If a web server failed, the locally attached data became unavailable. GeoCities was beginning to move toward Network Appliance file servers mounted over NFS, trading some of that fragility for more centralized, fault-tolerant storage.

That is an unusually useful snapshot because it catches a giant consumer website while the infrastructure underneath it was still visibly evolving.

Twelve terabytes once held a civilization

Today, 12 TB can fit on a couple of consumer hard drives.

In 1999, the page describes an entire fenced server environment built around that scale of storage. Rows of machines sat on raised floors. Cooled air was routed through the racks. Thick bundles of Ethernet cable crossed the cage. Steel rods tied racks together because Silicon Valley earthquake preparation was not theoretical.

It is one thing to say GeoCities hosted millions of personal pages. It is another to see the physical apparatus required to keep those pages alive.

The old web was never weightless. Every dancing baby, fan shrine, guestbook, and JPEG had to exist on spinning disks somewhere.

Pages like this are why CacheRat’s 1,967 Ancient Web Domains research list matters. The best archaeological finds are not always the famous websites themselves. Sometimes they are tiny side pages documenting the machinery behind them.

GeoCities eventually became internet mythology.

This page shows the rack bolts.

See the GeoCities server-cage photographs

Posted on by

Ancient Web: Eric’s Dodge Truck Pages Became a Registry for Machines Nobody Documented

If you own a common classic car, somebody has probably published a glossy book about it. If you own an obscure 1950s Dodge truck and need to figure out exactly what the rust-covered frame number means, you may end up at Eric’s Classic Dodge Truck Pages instead.

Visit TownWagon.com

The site concentrates on older Dodge trucks that sit outside the usual muscle-car and collector-car spotlight. Its material includes 1954–1966 Town Wagons and Town Panels, 1957–1960 Power Giant Power Wagons, and registries covering several generations of Dodge and Plymouth trucks stretching back into the 1930s.

The owner registries are the key old-Web feature. Instead of merely presenting factory specifications, the site gathers surviving vehicles and the people who own them. That turns a personal enthusiast page into a loose research database: what still exists, which serial numbers have surfaced, what configurations owners have found, and which details recur across supposedly obscure models.

The information has been useful outside the site itself. Old truck discussions point people toward TownWagon pages when serial numbers are difficult to locate or decode. In one case, Eric described photographs and frame-number information that helped owners know where to scrape through grease and corrosion. Vintage-vehicle publications and enthusiast forums have also cited the registries when discussing production numbers and identification.

This is the kind of work that manufacturers rarely maintain forever. A company may preserve broad corporate history, but the person standing in a garage with an unidentified sixty-year-old truck needs something much more specific: where is the number stamped, what did this model look like, which years used this detail, and has anybody else found one like it?

Enthusiast sites fill that gap because their priorities are different. Nobody has to justify the traffic potential of a page devoted to a Power Giant frame stamp. The information exists because somebody needed it and figured somebody else eventually would too.

For researchers, owner registries also create a snapshot of the surviving population of machines at the time entries were collected. They are not official production records, and they should not be treated as complete censuses. They are something else: evidence assembled from actual vehicles by a community that knew what it was looking at.

That kind of specialist infrastructure appears repeatedly in CacheRat’s 1,967 Ancient Web Domains research list. The Web was very good at letting a handful of people build the reference book they wished existed.

Somewhere, somebody is still scraping sixty years of road grime off a Dodge frame. TownWagon may tell them where to scrape.

Explore Eric’s Classic Dodge Truck Pages and registries