Category: Ancient Web

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

Posted on by

Ancient Web: Prem Subrahmanyam Built a Personal Fossil Museum

Prem Subrahmanyam’s fossil gallery is a small site with a very old-Web premise: I know something, I have collected examples, and I am going to put them online so other people can learn from them.

Visit Beneath the Calamites Tree

The page is called Beneath the Calamites Tree, after the giant Carboniferous horsetail relatives whose remains became part of the coal-forming vegetation of their era. The header image is not a stock illustration; Subrahmanyam says he created the restored Calamites forest himself in LightWave 3D.

From there the site branches into sections on trilobites, graptolites, fossil plants and a fossil-hunting FAQ. The specimens were collected by the author in parts of the southeastern and midwestern United States.

That combination—field collecting, photography, explanatory text and homemade graphics—is exactly why personal science pages from this era remain useful. They are not trying to be comprehensive museums. They show what one informed hobbyist actually encountered and thought was worth explaining.

Small subject, deep attention

The site is also a reminder that educational publishing did not require institutional scale. A collector could photograph a specimen, write up what it was, explain where it fit geologically and publish the whole thing with a few hand-coded links.

There is a practical difference between that and a generic encyclopedia page. Personal collections tend to preserve the details that matter to other collectors: what kinds of fossils turn up together, what they look like outside a museum display case, what novice hunters confuse them with, and what the author learned in the field.

The page’s copyright notice is another period detail. Subrahmanyam explicitly permits personal use of the images and educational reuse with credit while reserving commercial rights. It is the kind of plain-language sharing policy that was common before every site acquired a legal-document generator.

The site is not large, flashy or algorithmically important. It does not need to be. Its value comes from the fact that a real collector decided to document a corner of paleontology carefully enough that the work still survives more than two decades later.

That is a recurring theme in the old web: the useful page exists because somebody cared about a subject more than they cared about traffic.

Browse Prem’s original fossil gallery

Posted on by

Ancient Web: ThemeParks.ie Is a Personal World Atlas of Amusement Parks

Some people come back from vacation with a folder of photographs. The person behind ThemeParks.ie came back from years of amusement-park travel and built a world index out of them. The result is a personal photo archive spanning parks and travelling fairs across Europe, Asia, Africa, Australia, and North America.

http://www.themeparks.ie/

The front page is gloriously direct: country, park, link. No recommendation engine. No giant booking widget. No desperate attempt to predict what attraction you should visit next. You choose a place and start looking.

The archive includes famous parks, obscure local funfairs, closed parks, and places most English-language theme-park sites barely mention. Ireland gets particularly deep coverage, but the index ranges from Tokyo DisneySea and Europa-Park to amusement parks in Azerbaijan, Kuwait, Vietnam, Morocco, Egypt, and South Africa.

The old pages become more valuable when the park itself changes or disappears.

Bell’s Amusement Park in Tulsa, for example, is preserved as a photo page noting its 1951-2006 operation and former attractions. The site’s page for Folkestone’s Rotunda/Funland Park records a seaside park later sold for redevelopment and documents its restored 1920s side-friction coaster. Those are no longer just vacation photos. They are visual evidence of places that changed or vanished.

The archive also captures rides while they were being built. ThemeParks.ie documented the assembly of the CĂş Chulainn wooden coaster at what is now Emerald Park over a period of months. That gives coaster researchers something different from a finished promotional image: a progression of the physical structure taking shape.

The change log is another little historical gift. Entries stretch back through the mid-2000s and continue into the 2020s, documenting batches of photos from fairs and parks around the world. It is possible to watch one enthusiast’s travel history turn into a long-running record of the amusement industry.

This is why personal websites matter for preservation. Institutions tend to archive the famous ride, the major operator, or the official press release. Enthusiasts photograph the queue railings, the temporary fair layout, the weird kiddie ride, the sign that was replaced next year, and the coaster everyone forgot after the land became apartments.

http://www.themeparks.ie/

ThemeParks.ie is not a comprehensive database and does not pretend to be one. It is something more old-web than that: a person’s obsession organized carefully enough that everyone else gets to use it as a reference.

Posted on by

Ancient Web: The Shadowlands Built a Massive Sea Monster Index

The Shadowlands sea-serpent page is what happens when somebody decides that every lake monster, mystery carcass, old newspaper account, giant squid story and blurry photograph belongs in one place—and then actually does the work.

Visit The Shadowlands sea-serpent archive

The page describes itself as a collection of sea serpents and lake monsters, and by 2010 it had grown into a huge index of reports, photographs, videos and visitor submissions. Champ, Ogopogo, Chessie, Caddy, Loch Ness, giant octopus stories, alleged monster sharks, unidentified carcasses and historical sightings all live together on one enormous document.

This is not a scientific database. Claims are presented with wildly different levels of support, and some entries rely on eyewitness reports or recycled media stories. But the page does something modern social platforms are remarkably bad at: it keeps the material together long enough for a researcher to see the mythology as a body of culture.

The page is part scrapbook, part index, part argument

The author begins with a reasonable observation—that much of the ocean remains poorly explored—and uses examples such as the rediscovery of the coelacanth and the existence of giant squid to argue that unknown large animals are not impossible in principle. From there the page moves into far more speculative territory.

That transition is useful to observe. Cryptozoology often works in the space between two true statements: we have not cataloged every species, and eyewitnesses sometimes report strange animals. The difficult part is everything between those facts and a specific claim that a legendary creature exists.

The Shadowlands page does not resolve that problem. It preserves it.

The best material is often historical. The page reproduces descriptions attributed to nineteenth- and twentieth-century sightings, regional monster names, newspaper-era stories and sketches. It also reflects the period when amateur paranormal sites invited visitors to email photographs, video, scanned material and firsthand reports directly to the webmaster.

That human collection model matters. Before Reddit threads and Facebook groups, niche subjects accumulated around individuals willing to maintain link pages for years. The result could be messy, credulous and extremely useful as cultural evidence at the same time.

The page is also visually and structurally ancient in the best sense: giant table of contents, inline images, hundreds of links, repeated decorative graphics and no concern whatsoever for whether a phone likes it.

For anyone studying how folklore migrated onto the early web, The Shadowlands is not just about monsters. It is itself an artifact.

Explore the original Shadowlands monster page

Posted on by

Ancient Web: Poets’ Corner Has Been Transcribing Poetry Since 1994

Before every classic poem had a dozen scraped copies floating around search results, somebody had to actually type the damn thing in. Poets’ Corner, part of The Other Pages, began in 1994 and grew through exactly that sort of patient volunteer labor.

https://theotherpages.org/poems/

The collection describes itself as one of the oldest large text resources on the web. By its 2022 count it held roughly 7,459 works by about 800 poets, ranging from tiny inscriptions to book-length material. The project was created and maintained by a small group of editors, including Steve Spanoudis, Bob Blair, Jon Lachelt, and Nelson Miller, with additional contributions from volunteers around the world.

The numbers are interesting. The method is better.

The site grew through transcription, editing, correction, indexing, and hand-built HTML. That means it preserves a kind of web scholarship that is easy to overlook now: people taking printed material, verifying it, organizing it, and making it searchable without needing a commercial reason to do so.

Poets’ Corner includes detailed author indexes, shorter and longer reading selections, faces of poets, favorites, National Poetry Month material, subject indexing, and thousands of individual text files. Its editors deliberately kept the design simple so the pages would remain fast and usable across devices and slow connections.

That simplicity paid off.

In 2020, the site underwent a huge manual update to move old material to HTTPS, repair dead links, modernize formatting, and remove features that no longer worked. One update notes that the collection contained more than 2,400 files across 82 directories and that every file had to be touched by hand. That is the sort of maintenance work invisible to users right up until nobody does it and half the web disappears.

The site also has a clear archival philosophy. It focuses primarily on works believed to be in the public domain in the United States, asks users to verify copyright rules elsewhere, and prefers linking to maintained source pages instead of copying whole chunks of the site into abandoned mirrors.

For literary researchers, the value is obvious. For web historians, the site is useful for another reason: it shows how durable a modest architecture can be. A resource created in the era of WebCrawler and Magellan survived into the 2020s because the editors kept the underlying structure understandable.

https://theotherpages.org/poems/

Poets’ Corner is not impressive because it looks modern. It is impressive because somebody kept caring about the text for more than thirty years. That is a much rarer web technology.

Posted on by

Ancient Web: Reptoids Connected Real RAND Transit Papers to Secret Underground Trains

This is where old-Web research gets fun: the page is wrong in the interesting way. The Reptoids Research Center took real 1970s RAND papers about evacuated-tube high-speed transportation and folded them into claims about reptilian humanoids, black-budget tunnel systems and secret underground shuttle networks.

Visit the original Reptoids tunnel page

The real documents are the anchor. RAND researcher Robert M. Salter published The Very High Speed Transit System in 1972 as paper P-4874. The proposal described magnetically levitated vehicles moving through low-pressure underground tubes, with extremely high theoretical speeds and energy recovery during deceleration. Salter followed with Trans-Planetary Subway Systems: A Burgeoning Capability, P-6092, in 1978.

Those papers existed. Transportation archives still catalog them.

What the Reptoids page does next is something very different. It treats the existence of ambitious underground-transit proposals as supporting evidence for claims that a vast classified tunnel network was actually built and is used by intelligence agencies, military programs and favored corporations. The page also connects those claims to alleged reptilian-human encounters and the mythology surrounding places such as Dulce Base.

There is no comparable evidence on the page establishing that such a secret network exists.

A useful specimen of conspiracy construction

That gap between evidence and conclusion is precisely what makes the page worth preserving.

The pattern is familiar: start with a genuine technical document, patents or government research; establish that the underlying technology was seriously discussed; then slide from “this was proposed” to “this was secretly completed.” The factual material gives the speculative material weight it did not earn on its own.

The page is especially valuable because it preserves this process in a pre-social-media form. There is no recommendation algorithm feeding the reader increasingly dramatic claims. Instead there are hand-built links, scanned articles, patents, quotations and a site owner manually assembling a case.

It also reveals something about 1990s and early-2000s conspiracy culture: government documents themselves were trophies. Finding a RAND paper with a science-fiction-sounding title felt like discovering a loose thread in reality.

The funny part is that the actual RAND work is already interesting enough. Ultra-fast evacuated-tube transit, electromagnetic propulsion and coast-to-coast travel concepts do not need subterranean reptilians to be strange.

The Reptoids page is therefore useful on two levels: as a path into genuine transportation-history documents, and as a preserved example of how legitimate research was repurposed into Internet mythology.

Read the original Reptoids Research Center article

Posted on by

Ancient Web: Howard Wiseman Mapped Post-Roman Britain and Admitted the Maps Were Wrong

Historical maps often look more certain than the evidence behind them. A colored border gets drawn, a kingdom gets a name, and suddenly an argument becomes geography. Howard Wiseman’s old Ruin and Conquest of Britain site does something refreshing: it draws the maps anyway, then spends considerable effort explaining why the early ones are almost certainly wrong in detail.

https://howardwiseman.me/DECB/DECBmaps.html

The maps cover Britain at several points from the end of Roman rule through the tenth century, including reconstructions for roughly 400, 530, 585, 660, and 942. Wiseman uses them to place locations named in early sources and to visualize his reconstruction of political control during the extremely murky fifth- and sixth-century transition from Roman Britain to Anglo-Saxon England.

The useful part is not just the artwork. It is the argument attached to it.

Wiseman explicitly warns that drawing crisp boundaries for the fifth century requires more information than survives. He notes that some state names are later forms, some are based on doubtful traditions, and some labels are simply his own practical inventions. He also points out that the maps probably simplify a political landscape that contained more, smaller states than can now be reconstructed.

That level of self-criticism makes the site more useful, not less.

The larger project dates back to the 1990s and includes a roughly 20,000-word attempt to arrange early sources into a chronological narrative, tables, source discussions, scholarly publications, and a later “best estimate” reconstruction that deliberately relies on more trustworthy evidence. Wiseman separates possible history from reliable history rather than pretending the distinction can be eliminated.

The maps themselves are dense with places: Roman towns, civitates, forts, traditional battle sites, early kingdoms, rivers, and proposed boundaries. Notes walk through places associated with figures such as Gildas, Vortigern, Hengist, Cunedda, Urien, and the shadowy historical material later connected with Arthur.

For researchers, this is a good example of a personal website doing work that would now often be broken across journal articles, GIS projects, social posts, and subscription databases. The whole argument remains navigable as one person’s evolving research project.

It is also a useful warning about historical visualization. A map can be beautifully legible while the underlying history remains uncertain. Wiseman does not hide that problem behind the graphics.

https://howardwiseman.me/DECB/DECBmaps.html

The old web was full of enthusiasts willing to spend years building resources that were too specialized for a commercial publisher and too opinionated for an encyclopedia. This is one of the better surviving examples: detailed enough to be useful, and skeptical enough to tell you where not to trust it.

Posted on by

Ancient Web: A High Schooler Built a Museum of Blue Screens in Public

Long before social media turned every broken airport screen into a viral post, one teenager was already collecting them. Kevin’s Blue Screen of Death page is a gallery of Windows crashes appearing where computers were very much not supposed to look like computers: airports, ATMs, department stores, ticket machines, giant public displays and kiosks.

Visit Kevin’s Blue Screen of Death gallery

The page is more than a pile of screenshots. Each entry records the kind of machine, the Windows version when known, the location and a comment. There are Windows 9x crashes on public displays, NT 4.0 on kiosks, an ATM in Spain, an airport screen at Heathrow and even a giant Times Square display.

The commentary is unmistakably from the era. An ATM crash gets a joke about whether your money is safe. A ticket machine gets a crack about lost revenue. The site also separates genuine photographs from doctored BSOD pictures made purely for entertainment.

That distinction is useful now. The web has spent decades recompressing, reposting and divorcing images from their original context. Here, the collector actually tried to say what was real, what was staged and what was merely funny enough to keep.

A teenager’s joke became a historical record

The site’s own update log is almost as interesting as the gallery. Kevin says he built it in high school and, by 2012, could hardly believe it had remained online for more than eight years. When the original free subdomain disappeared, he did what many people from that generation eventually had to do: bought hosting and restored the site with the original filenames intact.

That preservation decision is why the page still works as a small museum of early embedded Windows. Public terminals in the late 1990s and early 2000s often exposed the operating system underneath when something went wrong. Instead of a branded recovery screen, the machine suddenly confessed: yes, this airport sign is just a PC.

The site also captures a transitional moment when Windows was escaping the desktop and showing up everywhere. A crash on your beige tower was annoying. A crash on a thirty-foot billboard was public theater.

Today, public systems often hide failure behind custom shells, remote-management layers or blank screens. Kevin’s collection preserves the period when the machinery beneath the interface was still very easy to catch with its pants down.

Browse the original BSOD collection

Posted on by

Ancient Web: Scirocco.org Preserved a Volkswagen Mailing List as a Technical Memory

A lot of useful automotive knowledge never came from service manuals. It came from somebody with frozen lug bolts, a leaking hose, three bad ideas, and an email list full of people who had already broken the same thing. Scirocco.org is a surviving piece of that older system of knowledge transfer, built around Volkswagen Scirocco owners talking directly to one another.

http://www.scirocco.org/

The domain dates back to the late 1990s, and contemporary references describe an international mailing list with hundreds of Scirocco owners. That matters because the Scirocco was already becoming an enthusiast car by then. Factory support was thinning out, unusual parts were getting harder to find, and increasingly the best person to ask about a strange bracket or cooling problem was another owner who had one sitting in the driveway.

The most valuable part of the surviving site is not polished editorial content. It is the mailing-list archive.

Messages from the early 2000s preserve exactly the kind of details that disappear when a community moves platforms: discussion of obscure factory skid plates, differences between early Mk1 bumpers, radiator and cooling-system troubleshooting, lighting swaps, dealer-only hoses, part availability, and the improvised repairs people used when normal documentation ran out.

That rawness is the point.

A modern car site usually tries to turn everything into a clean article, a video, an affiliate link, or a searchable question-and-answer page. A mailing list preserved the conversation itself. Advice is mixed with jokes, corrections, signatures, arguments, bad memories, and somebody inevitably trying to unsubscribe by emailing the whole group. It is less tidy, but it shows how the knowledge actually formed.

Scirocco.org also captures a particular phase of enthusiast Internet culture. Owners were not merely consuming information about a car. They were collectively maintaining an informal technical library while the cars aged underneath them. Somebody in one country could identify a rare part for somebody in another. A repair that took one person a weekend could save dozens of later owners from repeating the same mistake.

There is research value here beyond Volkswagens. Mailing-list archives are evidence of how specialized communities operated before Reddit, Facebook groups, Discord, and algorithmic feeds. The archive is chronological, verbose, sometimes ridiculous, and unusually resistant to the flattening that happens when old discussions are summarized into generic content.

For anyone restoring an early Scirocco, researching enthusiast communities, or studying how technical knowledge moved around the pre-social web, the old list is still worth digging through.

http://www.scirocco.org/

A 25-year-old email about a stubborn wheel or a weird piece of factory plastic may not look historically important. Wait until the last person who remembers that part is gone. Then suddenly it is documentation.

Posted on by

Ancient Web: Saint IGNUcius Turned Free Software Into Hacker Religion

Richard Stallman’s Saint IGNUcius page is one of those artifacts that makes more sense if you remember when hacker culture routinely mixed serious technical principles with jokes elaborate enough to become institutions. Here, GNU, Emacs and free software are recast as a mock religion complete with a saint, a church and a halo made from obsolete computer hardware.

Visit the Saint IGNUcius page.

The page is dated August 15, 2000, although the character itself had already become part of Stallman’s public appearances. Saint IGNUcius is introduced as a saint of the “Church of Emacs,” a running joke built around the old rivalry between Emacs and vi users and the quasi-religious devotion programmers sometimes show toward their preferred tools.

The costume’s halo is an old computer disk platter. That detail may be the entire philosophy of the early hacker Web in one photograph: repurpose dead hardware, make an extremely nerdy joke, then publish an explanation so strangers can get the reference.

The joke carries real software culture inside it

The page uses religious language to parody software evangelism. GNU becomes part of the saint’s identity. Emacs is treated as doctrine. Source code and software freedom become material for mock ceremonies and puns.

One recurring bit is the idea of a “Foobar Mitzvah,” in which a user symbolically enters the community by understanding or sharing source. The humor works because the underlying argument was serious: users should be able to study, modify and share the software they depend on.

That mixture was common in older programmer culture. Technical communities accumulated folklore, saints, demons, holy wars and deliberately overdramatic language around editors, operating systems and licenses. The jokes were social glue as much as entertainment.

It is also careful not to become an actual church

The page explicitly says the Church of Emacs is a joke and warns readers not to take it—or any church—too seriously. That clarification is useful because a screenshot of Stallman wearing a robe and disk-platter halo can travel much farther than the context around it.

The page’s copyright notice also reflects the culture it is promoting: verbatim copying is permitted as long as the notice is preserved. The joke itself is meant to circulate.

Twenty-six years later, Saint IGNUcius is worth remembering because software culture has become vastly more corporate and professionally packaged. Developer communities still produce memes, but fewer project leaders show up in homemade religious costume to preach editor jokes with a piece of computer storage strapped behind their head.

Whether that represents progress is left as an exercise for the reader.

See Saint IGNUcius in his original habitat, complete with the Church of Emacs explanation and the disk-platter halo.

Posted on by

Ancient Web: PlaneCrashInfo Built a Searchable History of Aviation Disasters

PlaneCrashInfo is the sort of site the modern web keeps trying to replace with dashboards, map widgets, infinite scroll and six layers of JavaScript. It does not need any of that. It is a hand-built aviation accident database, organized by year and category, with terse records that let the subject matter speak for itself.

Visit PlaneCrashInfo

The site says its database covers fatal civil and commercial airline accidents, fatal cargo and test flights, major military transport accidents, large helicopter accidents, fatal airship accidents, crashes involving famous people, and other noteworthy incidents. Each record follows a compact format: date, time, location, operator, flight number, route, aircraft type, registration, construction number, people aboard, fatalities, ground deaths and a short summary.

That structure matters. It makes the database useful to researchers without pretending that every event needs an interactive documentary wrapped around it.

The site dates itself to 1997 and its database stretches back to the earliest decades of aviation. The main navigation is gloriously old-Web: Database, 100 Worst, Unusual Accidents, Accident History, Statistics, Last Words, Accident Map, Famous Deaths and references. There is even an offer from the maintainer, Richard Kebabjian, to help look up information on older accidents if visitors provide enough detail.

A database before everything became an app

PlaneCrashInfo is also a good example of how much useful work independent site owners once put online without trying to turn the result into a platform. The database is plain enough that people have scraped it for research projects, while university libraries have linked to it as an open aviation data source.

The site also maintains statistical pages. One analysis breaks fatal accidents into broad primary-cause categories such as pilot error, mechanical failure, weather, sabotage and other causes. Whatever one thinks of simplifying complex accidents into one initiating cause, the important thing is that the methodology and sample criteria are stated rather than hidden behind a chart.

There is a darker reason the site remains useful: aviation accident history is fragmented. Official investigations are often distributed across different national authorities, decades and document systems. A personal database cannot replace primary records from agencies such as the NTSB or ICAO, but it can give researchers a fast starting point and enough identifiers to go hunting for those records.

That is the old web at its best: one person builds the index because the index ought to exist.

Browse the PlaneCrashInfo aviation accident database