Category: Ancient Web

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

Posted on by

Ancient Web: Giangrandi Kept an Engineering Notebook Online Since 2002

Giangrandi’s website looks less like a publication schedule and more like the notebook drawer of somebody who keeps building things and remembers to write down what worked.

Visit Giangrandi’s engineering site

The homepage divides the collection into electronics, mechanics, optics, and software and algorithms. The site says it was created in October 2002. More than two decades later, new technical material is still appearing.

The electronics section alone is a deep bench.

There are radio transmitters, SWR meters, crystal ladder filters, antenna tools, Geiger counters, fluorescent-tube experiments, transformer calculations, heat-sink notes, home PCB etching, power-supply troubleshooting, and a GPS-disciplined grandfather clock. Some pages are project builds. Others are explanations or measurement techniques that grew out of practical work.

The dates matter because they show accumulation rather than a one-time site launch. Articles from the mid-2000s sit beside updates from the 2010s and 2020s. The page structure stayed recognizable while the knowledge base kept growing.

Personal technical publishing ages unusually well

A lot of engineering information does not need a redesign every three years.

A formula for calculating a heat sink does not become easier to understand because the page has infinite scroll. A circuit diagram does not benefit from a recommendation algorithm. If a page explains the measurement, labels the components, and keeps its images intact, it can remain useful for a very long time.

Giangrandi’s site understands that almost accidentally by simply continuing to be a website.

The author also documents conventions. The electronics pages explain European schematic habits, notation, current direction, resistor symbols, and the abbreviated engineering notation used for component values. That small bit of context prevents the kind of confusion that happens when a diagram travels without its surrounding explanation.

The software section has the same practical flavor: a moon-phase calculator, coordinate conversions, an Easter calculator, a random-string generator, and a JavaScript clock with vintage displays. None of this needs to be bundled into an app store. A URL is enough.

For modern webmasters, the site is a useful argument for boring durability. Its hierarchy is shallow. Pages are individually linkable. Topics are grouped by what they are, not by when an algorithm thinks you should see them. Old material remains beside new material without becoming an archaeological layer hidden behind a CMS search box.

CacheRat’s 1,967 Ancient Web Domains research list is full of personal technical sites for a reason. People who build things often document the details that commercial documentation leaves out.

Twenty-four years later, those details are still sitting there.

Explore Giangrandi’s electronics, mechanics, optics, and software pages

Posted on by

Ancient Web: The-Pope.com Kept a Field Guide to HTML’s Extinct Tags

The-Pope.com is mostly known for religious material, but buried inside it is something far more relevant to Internet archaeology: Griff Ruby’s documentation of obsolete HTML and the browser wars that produced it.

Open the site’s old HTML-source page

Ruby’s broader Lost Tags of HTML material treats extinct markup like technological fossils. Tags appeared, disappeared, competed, became browser-specific, or were replaced by cleaner standards. A Web developer working today can look at <blink> or <marquee> as punch lines. On old pages, they were once perfectly ordinary tools.

The history matters because HTML did not arrive as one finished language and then proceed neatly through numbered versions. Netscape, Microsoft, browser vendors, standards groups, and individual developers pulled it in different directions. Some tags existed because one browser company shipped them first and dared everyone else to catch up.

Ruby documented that mess rather than pretending the Web evolved according to a clean architectural diagram.

The work has been noticed outside the site. The Web Ahead highlighted The Lost Tags of HTML in 2015, and later scholarship on Web semantics has cited Ruby’s page as a reference on abandoned markup.

Old HTML is easier to understand when somebody remembers why it existed

A specification can tell you that a tag is obsolete. A historical page can tell you what people were trying to accomplish with it.

That distinction becomes important when maintaining or restoring old sites. Ancient source code contains tags, attributes, layout hacks, browser checks, and invalid structures that make much more sense once you remember the environment they were written for. Developers were not necessarily incompetent. They were targeting Netscape Navigator, Internet Explorer, Mosaic descendants, and standards that were still moving underneath them.

This is one reason old personal sites remain useful even when their main subject is unrelated to computing. Somebody tucked a detailed technical essay into a corner twenty years ago, search engines mostly forgot about it, and it survived because the domain survived.

The exact source page here is suitably strange-looking. The surrounding HTML history is the real prize.

CacheRat’s 1,967 Ancient Web Domains research list keeps turning up these accidental archives: information preserved not by a dedicated museum but because somebody never deleted the directory.

For anyone who has ever opened a 1990s page and wondered what the hell the source code was doing, Ruby’s work is worth the detour.

Explore the surviving HTML material on The-Pope.com

Posted on by

Ancient Web: Mick Cooper Turned a Personal Homepage Into a Portsmouth Memory Bank

Mick Cooper’s homepage begins the old-fashioned way: here are the things I care about, here are some links, and if you care about any of them too, come in.

Visit Mick Cooper’s homepage

The subjects immediately start branching. Music. Portsmouth Football Club. Family history. British actors. MIDI files. Local schools. Songwriting. The Titanic. Banjo books. The Portsmouth music scene.

Cooper explains that he came to computers later in life and kept using an Atari 1040ST with Cubase 3.1 because he knew it well. He also mentions an Acorn RISC PC, a Roland G800 keyboard, and eventually a more conventional PC setup.

That detail tells you a lot about the site.

This is not a portfolio built to establish a brand. It is a personal filing cabinet that gradually became public history.

The local music pages are the real rabbit hole

Cooper performed around Portsmouth for decades under his own name and as Mickey Morgan. His music timeline records bands, venues, equipment, dates, and gigs stretching back into the 1960s. One page lists appearances almost date by date, including work with the band Heaven and shows during a period when the local live circuit intersected with much larger acts.

The larger Portsmouth material goes beyond autobiography. Cooper and collaborators gathered memories, photographs, venue histories, band information, newspaper research, and other fragments of the city’s music scene. The result preserves exactly the level of history that tends to evaporate first: not only famous performers, but local halls, resident bands, one-night bookings, old posters, and people who remember what happened there.

That kind of record is difficult to reconstruct after the participants are gone.

It is also a good example of why personal websites matter. A modern profile page might comfortably hold Cooper’s biography. It would be much worse at holding decades of unrelated research, family material, scanned ephemera, MIDI files, local history, and several enthusiastic side quests without demanding that everything be squeezed into one content format.

The old homepage had no such problem. Add another link.

For a researcher, the site is useful because the material remains in context with the person who collected it. For a webmaster, it demonstrates the underrated power of a hand-maintained index page: when the subject expands, the navigation expands with it.

CacheRat’s 1,967 Ancient Web Domains research list keeps turning up sites like this because local knowledge often survives on the web through one person’s refusal to throw anything away.

That is not bad information architecture.

Sometimes it is the archive.

Explore Mick Cooper’s site and Portsmouth history pages

Posted on by

Ancient Web: MARVUNAPP Built an Encyclopedia for Marvel’s Forgotten Characters

Marvel has created enough characters over the decades that some of them barely survive a single story, a handbook entry, or the memory of one reader who refuses to let an extremely minor villain die twice. The Appendix to the Handbook of the Marvel Universe was built for exactly those people.

Visit the Naga entry in the Marvel Universe Appendix

The site is commonly known as MARVUNAPP. Jeff Christiansen launched the project in 2001, and it grew into a giant catalog of Marvel characters, places, races, alternate realities, and other material too obscure for ordinary reference books to handle comfortably.

By 2017 the Appendix celebrated its 10,000th entry.

That number explains the project better than any mission statement. MARVUNAPP was not trying to summarize Spider-Man for somebody who had never read a comic. It was trying to answer questions about the person who appeared in three panels in 1974 and was never mentioned again.

The master-list URL CacheRat found points to Naga, the ancient Lemurian ruler associated with the Serpent Crown and Namor mythology. It is a typical Appendix page: creator and appearance information, history, powers, relationships, clarifications, and references stitched together from scattered comics.

The larger importance is the structure around that page.

Before fandom wikis swallowed the search results

MARVUNAPP belongs to an earlier model of fan knowledge. A small group of deeply invested people chose a scope, developed conventions, researched primary material, argued over continuity, and published pages manually.

The result became influential enough that later Marvel databases, discussion forums, and comic researchers routinely cite it. Some contributors also crossed into official Marvel handbook work, blurring the line between fan reference and professional reference in an unusually productive way.

That does not mean every interpretation on the site is beyond dispute. Comic-book continuity is basically an industrial process for manufacturing edge cases. What matters is that the Appendix records its reasoning and points back toward appearances rather than flattening everything into a paragraph with no provenance.

Modern Web platforms are very good at making enormous databases. MARVUNAPP is interesting because people began building one before the obvious platform for it existed.

It grew because somebody kept asking, “Who was that guy?” and refused to accept “nobody remembers” as an answer.

Explore the Naga page and the surrounding Appendix

Posted on by

Ancient Web: WeirdCrap Preserved the Web’s Chick Tract Parody Scene

There was a period when one recognizable piece of American print culture could appear in a laundromat, bus station, college hallway, convenience-store bathroom, or your mailbox: the tiny evangelical comic known as a Chick tract. Naturally, the early Web responded by making parodies of them.

Visit WeirdCrap’s Chick parody archive

WeirdCrap.com became one of the places where those parodies collected. The site’s Chick section linked and hosted spoof tracts that copied the visual grammar of Jack Chick’s originals—small black-and-white comic panels, blunt moral certainty, escalating disaster—and redirected it toward satire.

The result is useful Internet archaeology because this was not just one joke page. Chick-tract parody became a recognizable Web form. Other old sites and zines listed WeirdCrap alongside collections of parody tracts, and individual pieces such as Loved to Death continued to be cited years later.

That matters because the Web did not invent remix culture after social media arrived. People were scanning, redrawing, rewriting, linking, and distributing recognizable formats long before a meme template came with a share button.

The Chick tract was almost mechanically suited to that treatment. Its visual language was consistent enough that a parody could be recognized immediately, even when the message had been turned inside out.

A small archive of a larger Internet habit

There is also something appropriately old-Web about WeirdCrap itself. The site name does not pretend to be an institution. It sounds like exactly what somebody would type into a browser after deciding the Internet needed a permanent shelf for material that did not fit anywhere respectable.

That shelf now has historical value.

Pages like this preserve more than jokes. They show what people were reading, what formats were culturally recognizable, how copyright-consciousness differed, and how satire circulated when distribution meant personal domains, zines, mirrors, and hand-made link pages.

The surviving section is not a neutral archive of Jack Chick’s work, nor does it need to be. It documents one particular Internet reaction to it.

CacheRat’s 1,967 Ancient Web Domains research list contains plenty of sites like this: not official history, but evidence of what people actually did online.

WeirdCrap kept a little piece of that ecosystem intact.

Explore the old Chick parody section

Posted on by

Ancient Web: Beautiful Dragons’ Nutrition Chart Became Bearded-Dragon Forum Infrastructure

Some web pages become important without ever becoming famous. Beautiful Dragons’ nutrition chart is a good example: for years, bearded-dragon owners answered feeding questions by posting the same URL.

Visit the Beautiful Dragons nutrition chart

The page is essentially a specialist reference table. Foods are organized and color-coded so a keeper can quickly distinguish commonly recommended staples from occasional foods and items the site says should not be fed.

That sounds simple because it is simple.

The interesting part is what happened around it.

Bearded-dragon forums contain references to the chart going back at least to 2006. Threads from 2008, 2009, 2010, 2011, 2012, and later repeatedly point newcomers toward Beautiful Dragons when the same practical questions appear: Which greens? Which vegetables? Which feeder insects? What should be avoided?

People did not need another social-media post explaining the same list. They needed one stable page they could point at.

For a while, that page became community infrastructure.

A bookmark can become a standard reference

Old hobby forums were full of this pattern.

One person built a useful chart, calculator, FAQ, identification page, or compatibility list. Other users tested it, argued about it, remembered it, and linked it. Eventually the URL became shorthand. A new question appeared and somebody replied, essentially, “Use this.”

That is a form of web authority that search rankings do not describe very well.

Beautiful Dragons is especially interesting because the surrounding discussions preserve how people used the page. Forum users specifically mention the color coding and routinely recommend printing or bookmarking the chart. The reference escaped the boundaries of its own site and became part of the shared working memory of a hobby community.

Pet-care information also comes with an obvious warning: husbandry guidance changes, and an old web chart should not replace current veterinary advice or newer evidence. The historical value here is not that every recommendation must remain eternally correct. It is that the page shows how a community organized and shared practical knowledge before short-form platforms became the default place to ask everything again from scratch.

For webmasters, there is a wonderfully boring lesson in that. A page does not need feeds, accounts, notifications, or an engagement loop to matter. It needs to solve a recurring problem well enough that people remember the URL.

CacheRat’s 1,967 Ancient Web Domains research list contains a lot of pages whose real importance shows up in old forum links rather than press coverage. That trail is worth following.

Thousands of forum posts come and go.

The little chart they kept linking is still the interesting artifact.

Open the Beautiful Dragons nutrition chart

Posted on by

Ancient Web: Thompsonian Became Bill Thompson’s Personal Historical Warehouse

Some personal websites choose a subject. Bill Thompson’s site chose Bill Thompson’s interests, which is much closer to how the old Web was supposed to work. Over the years Thompsonian.info accumulated music research, Texas history, baseball, old broadcasts, photographs, documents, and whatever else seemed worth preserving.

Visit Thompsonian.info

One of its deepest veins is Frederick Delius. Thompson assembled recordings, discographical details, radio material, photographs, concert information, and odd connections that would have been easy to lose individually. The Delius Society did not merely notice the work; its journals repeatedly directed readers to Thompson’s pages.

In one case he compiled Delius selections from the BBC’s Desert Island Discs and added the results to his site. In another, the site made a rare 1929 recording of Sea Drift easier to investigate. Society publications described Thompson’s pages as containing a wealth of Delius-related information.

Then the site wanders elsewhere.

There is Houston Colt .45s history serious enough that the Handbook of Texas cited Thompsonian as a source. There is material on the old Southwest Conference, Houston-area music, vintage radio and recording history, photographs, personal compositions, and collections of links that still resemble the human-edited Web more than a search-engine results page.

That breadth is the point.

A homepage can become a source without planning to

A modern publishing system tends to demand a category, audience, content strategy, and clean information architecture before anybody writes the first paragraph. Personal websites often worked backwards. Somebody put up what they knew, added another page when they learned something else, and kept doing that for twenty years.

Eventually the strange pile acquired value through accumulation.

Thompsonian is especially interesting because later researchers cite it. A personal site crossed the invisible line from “here are my interests” to “this is where that obscure piece of information lives.”

That is a recurring pattern in CacheRat’s 1,967 Ancient Web Domains research list. The useful old Web is full of people who never founded an institution but slowly built something institution-like anyway.

Bill Thompson kept collecting, writing, scanning, linking, and organizing. Other people eventually needed what he saved.

That may be one of the better definitions of an Internet archive.

Explore Bill Thompson’s site

Posted on by

Ancient Web: GT-Four Pages Preserved the Details the Manuals Left Out

Old enthusiast websites are often most valuable where factory documentation becomes annoyingly general. The Toyota Celica GT-Four pages at supras.org.nz went the other direction: model differences, transmission variations, suspension hardware, swap details, and the stuff owners discover only after the car is already apart.

Visit the GT-Four technical site

The site is associated with New Zealand enthusiast Richard Doig, known online as RunningRich, and it circulated through Toyota communities for years as a technical reference. Contemporary forum posts routinely sent owners there rather than trying to reproduce the information from memory.

One good example is the GT-Four gearbox material. The site documented multiple transmission variants rather than treating every all-wheel-drive Celica transmission as interchangeable. That matters when the question is not academic but sitting on a garage floor covered in gear oil.

The Super Strut suspension pages are another useful survivor. Toyota’s Super Strut system is mechanically more complicated than an ordinary MacPherson-strut layout, and the old site broke down the components and differences in a form enthusiasts could actually use. Modern GT-Four communities still point back to the material when explaining how the system works.

There are also pages on 3S-GTE engine conversions, differentials, final-drive arrangements, and other pieces of the Toyota all-wheel-drive puzzle. The result feels less like a polished publication than a technical notebook that escaped onto the Web—which is often exactly what makes these sites useful.

The enthusiast Web was distributed documentation

Before every repair question migrated to a handful of giant platforms, knowledge accumulated across personal pages, club sites, mailing lists, and forums. One owner photographed a gearbox. Another measured something. Someone else figured out which model-year part actually fit. A webmaster organized it before the information vanished into a thread with a broken image host.

That system was messy, but it produced unusually specific documentation.

The GT-Four pages also demonstrate why old technical sites deserve preservation even when newer forums exist. Forum discussions repeatedly reference these pages as a standing source. Remove the old page and later discussions become less useful because their evidence disappears underneath them.

This is the kind of dependency that becomes obvious while working through CacheRat’s 1,967 Ancient Web Domains research list: the Web is full of newer pages quietly leaning on older pages that nobody thinks about until they are gone.

The design here is not the attraction. The accumulated mechanical detail is.

Browse the surviving GT-Four reference pages

Posted on by

Ancient Web: The MechWarrior II FAQ Captured a Game While It Was Still Being Built

Before game wikis, Discord servers, searchable patch notes, and official knowledge bases, a useful game reference could be one person collecting answers from Usenet while the developers themselves were still posting there. Will Day’s MechWarrior II pages are a surviving example of exactly that.

Visit Will Day’s MechWarrior II archive

Day compiled the Unofficial MechWarrior II FAQ in 1995 around Activision’s MechWarrior 2: 31st Century Combat. Version 2.0 is dated August 9, 1995, only weeks after the game reached stores. This was not retrospective fan archaeology. It was documentation happening while everyone was still figuring the game out.

The best part is the provenance.

Day explains that several members of the MechWarrior II programming team were posting to the relevant newsgroups and answering questions. The FAQ therefore preserves answers attributed directly to people working on game logic, enemy logic, level design, movies, and 3D models. Tim Morten appears repeatedly. So do John Keating, Sean Vesce, J.J. Franzen, and others from Activision.

That makes the FAQ more than a cheat sheet. It is a little slice of software-development communication from 1995.

The hardware section alone timestamps the era beautifully: a 486DX2/66, 8 MB of RAM, double-speed CD-ROM, VESA Local Bus or PCI video, DOS 6.0, and a Sound Blaster-compatible card. Higher resolutions were possible if your CPU could survive the experience.

Elsewhere the archive covers mission counts, graphics modes, bugs, custom Mechs, multiplayer, music hardware, and tiny utilities. Day even offered a DOS command-line program called mekname.zip for inspecting or changing custom Mech variant names. FTP addresses, HTTP URLs, email addresses, and news: references all coexist naturally because that was simply how online PC gaming information moved.

Why it still matters

Modern game documentation is often cleaner but less personal. A support article gets rewritten. A Discord answer sinks. A studio shuts down a forum. Search results increasingly point to copies of copies.

This FAQ preserves the discussion close to its source. You can see which questions players were asking in 1995, what the developers told them, what hardware mattered, and which problems were still actively being patched.

For web historians it is also a reminder that early fan pages did not merely talk about games. They served as practical infrastructure around them.

CacheRat’s larger 1,967 Ancient Web Domains research list contains many more sites from this same period where documentation survived because an individual bothered to keep it online.

Will Day’s MechWarrior II material is still worth opening because it catches a major PC game before history had hardened around it.

Explore the surviving MechWarrior II pages

Posted on by

Ancient Web: iiichan’s Shutdown Page Became Its Own History

Most dead communities leave behind a broken domain, a parking page, or a login screen connected to nothing. iiichan did something more useful: it explained what happened.

Visit iiichan

The surviving page gives the site’s history in unusually plain terms. The community launched in 2005 as foolchan, moved to the iiichan address in 2007, and shut down its imageboards in January 2012. Instead of erasing the old identity, the remaining domain became a small directory for redirects, subdomains, and projects that survived the boards themselves.

That makes the page valuable beyond the community it served. It is a first-party timeline written by the people close enough to know the sequence.

Imageboards are especially prone to disappearing badly. Their content is fast-moving, frequently anonymous, image-heavy, and dependent on software that was rarely designed as a historical archive. When a board closes, the normal outcome is not a tidy institutional collection. It is a hole.

iiichan’s history is also tied to a specific corner of Internet culture. Outside histories describe foolchan and iiichan as associated with the chiptune musician known as agargara or humblefool, and the community developed strong music, tracker, and demoscene connections. The surviving page points toward remnants including music-related archives, radio projects, and other subdomains.

That matters because “imageboard history” can easily collapse into a story about a handful of giant sites. Smaller boards often had distinct cultures built around art, games, music, technology, or tightly connected groups of regulars. They were not interchangeable copies simply because the software looked similar.

The iiichan page preserves that distinction by naming what remained after the main boards closed. A shutdown becomes a map of the projects that had grown around the community.

For webmasters, there is a practical lesson hiding here: when a site ends, leave an explanation. A few paragraphs containing launch dates, previous names, migration dates, closure dates, and surviving projects can become enormously valuable later. It costs almost nothing compared with the years of material that disappeared.

The old Web is full of mysteries created by silent shutdowns. Domains change hands. Redirect chains break. Search engines remember titles without pages. Researchers then have to reconstruct basic facts from cached fragments and forum posts.

iiichan spared us some of that work.

The page is no longer primarily an active imageboard. In 2026, its most interesting function is documentary. It tells us that foolchan existed, that it became iiichan, that the boards ended in 2012, and that not everything around them vanished at the same moment.

Sometimes a website’s final useful feature is a good obituary for itself.

Read iiichan’s surviving history