Category: Ancient Web

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

Posted on by

Ancient Web: Cowboy Showcase Preserved Real Dutch Oven Camp Cooking

Cowboy Showcase’s Dutch oven page feels less like a recipe site and more like somebody pulled a chair up next to a camp cook and started taking notes.

Visit the Cowboy Showcase Dutch oven page

The featured cook is Floyd Crandall, described as a rancher, mule man and Dutch oven cook from Fairfield, Idaho. His instructions are practical rather than precious. He begins with the thing beginners actually need to know: how to control heat.

His rule of thumb is to put roughly twice as much heat on top of the oven as underneath. With charcoal, that can be counted. With wood coals, the answer depends on the fuel, and the page simply says to experiment until you learn what works.

That tone continues through the recipes. A large roast gets onion soup mix, water, charcoal, then celery, onions, carrots and potatoes. Biscuits are crowded into a twelve-inch oven and browned by changing the coal placement as they rise. A dump cake is dry cake mix over fruit with a little lemon-lime soda.

Nobody is plating microgreens.

The best part is the cast-iron folklore

The page also preserves a seasoning method attributed to the late Harold Nelson, a ranch cook from Estes Park, Colorado. Instead of immediately reaching for oil, the method calls for boiling dry hay in the new cast iron, rinsing and drying the vessel, then applying cooking oil afterward.

The page suggests tea leaves as a substitute and notes that tannic acid reacts with rust, tying the folk method to the chemistry behind it.

That is exactly the kind of detail that disappears when traditional knowledge is rewritten into generic lifestyle content. The page is valuable not because Dutch ovens are obscure—they are not—but because it connects technique to people who actually used the equipment in ranch and camp settings.

Cowboy Showcase itself sprawls well beyond cooking. There are pages on saddles, bits, spurs, horse handling, cattle, brands, ranch stories, cowboy poetry, oral histories, chuckwagons and regional traditions. It has the structure of an old subject-matter portal rather than a modern publication chasing whatever keyword is hot this week.

The Dutch oven page is one small part of that larger archive, but it captures the site’s strength perfectly: practical knowledge, named sources, odd details, and enough context to show where the information came from.

Read Floyd Crandall’s Dutch oven methods on Cowboy Showcase

Posted on by

Ancient Web: Syd Lexia Ranked 100 Virtual Boy Games When There Were Barely 20

The Nintendo Virtual Boy did not have enough commercially released games to support a normal top-100 list. Syd Lexia did one anyway. That is the whole joke, except the site commits to it with the kind of completely unnecessary effort that made the older enthusiast web so much fun.

https://www.sydlexia.com/top100vb.htm

Published around April Fools’ Day 2010, the feature presents itself with the self-important machinery of a serious gaming ranking: an introduction, voting rules, contributors, commentary colors, rankings, and a dramatic countdown from 100 to number one.

The problem is that the Virtual Boy’s real software library was tiny.

Nintendo’s red-and-black tabletop system lasted only briefly in the mid-1990s, and North America saw only a small fraction of the number of games required for a hundred-entry ranking. Contemporary forum reactions immediately understood the absurdity. Some readers initially asked how a top 100 was possible for a machine with roughly a couple dozen releases. Others pointed out that this was precisely the joke.

Syd Lexia stretches that joke until it becomes structure.

Rather than merely posting “the Virtual Boy didn’t have 100 games,” the page mimics the familiar format of earnest gaming journalism. It creates the appearance of methodology, debate, expert voting, ranking anxiety, and editorial seriousness around a list with an impossible premise. The joke works because the presentation refuses to break character.

The site’s own introduction says the rules were modeled after earlier NES and SNES lists, with voters supposedly ranking one hundred games and only North American commercial releases being eligible. That requirement makes the impossible arithmetic even funnier. You can practically hear the spreadsheet screaming.

The page also became a small Internet reference point of its own. Gaming forums linked it, argued about it, repeated the joke, and used it as an example of a good parody list years later. That afterlife is part of the artifact: a one-off joke page survived because links and forum conversations kept pointing back to the original instead of replacing it with screenshots.

There is a broader lesson here about old gaming sites. They were often built by people with enough technical ability to publish and absolutely no editorial department telling them that spending this much time on a Virtual Boy gag had poor return on investment.

https://www.sydlexia.com/top100vb.htm

That is exactly the return. Sixteen years later, the page is still online, still understandable, and still funnier than the phrase “Top 100 Virtual Boy Games” has any right to be.

Posted on by

Ancient Web: Living Image Turned Vintage Cameras Into a Browsable Timeline

The Living Image Vintage Camera Museum does something simple and useful: it lets you browse camera history by decade and see the machines change shape in front of you.

Visit the Living Image 1930s camera room

The 1930s page is especially good because the decade was a hinge point. Wooden construction had mostly faded from serious cameras. Chrome plating was replacing nickel. Folding bellows cameras were still common, but 35mm cameras using daylight-loading 135 cartridges were beginning to spread quickly.

Bakelite appeared in some bodies. Die-cast metal became more common. Twin-lens reflex cameras matured. The Exakta VP helped establish the single-lens-reflex form, while Leica’s chrome-and-black styling pointed toward a visual language that would dominate camera design for decades.

The page then turns that history into a catalog you can actually browse: Agfa, Argus, Ensign, Ihagee, Kodak, Leica, Rolleiflex, Voigtländer, Zeiss and others. Many entries lead to individual pages and rotating views of the physical cameras.

The museum works because it is organized around objects

This is a different experience from reading a general history of photography. The chronology is visible in the hardware.

A 1930 camera still looks mechanically close to the previous era. A later model suddenly has the proportions, controls and materials that feel recognizably modern. Looking at dozens of examples side by side makes those changes easier to understand than a paragraph saying “camera design evolved.”

The site also separates its collection into decade rooms from the 1900s through the 1970s and later, with additional material on camera anatomy, workshops, collections and using old cameras. It is not trying to be a database of every model ever made. It is a curated museum built around real objects and the author’s own explanatory structure.

That is an important distinction. Old personal museums often have a point of view. They tell you what the collector noticed rather than pretending a neutral algorithm assembled the page.

The Living Image site has been online since around 2000 and still feels like the web was designed for exactly this kind of thing: shelves of knowledge without the glass cases.

For researchers, collectors and anyone trying to identify an old camera, it remains practical. For web archaeology, it is also evidence that a single subject can stay useful for decades if the links still work and the owner keeps the structure intact.

Browse the Living Image Vintage Camera Museum

Posted on by

Ancient Web: asdf.com Made Four Home-Row Letters Into a Web Philosophy

There are websites built to sell software, publish research, host communities, or explain hobbies. Then there is asdf.com, a site built around four letters your left hand learns before it knows what it is doing.

https://asdf.com/

The homepage is almost aggressively small. It presents the word asdf, links to an explanation of what asdf is, an about page, and a forum. That is roughly the whole proposition. The domain itself was registered in 1998, and the site’s own explanation treats the four-letter keyboard reflex as both joke and raw material.

The “What is asdf?” page is the real artifact.

It offers a scattered collection of definitions: asdf is what you type in a subject line when you have no subject; asdf is among the first things learned in typing; asdf is nothing; asdf “just is”; asdf likes jkl;. It also wanders into deliberately pointless facts, Scrabble scoring, cousins such as “aoeu,” and jokes about capitalization.

This is not useful in the normal sense.

That is why it is useful here.

The early web had room for pages whose entire purpose was an idea, a gag, a domain name, or an experiment that did not need to justify itself with traffic projections. asdf.com still carries that attitude. Its about page says the hosting project was formed to promote the creative expansion of the web through simplicity and experimentation. That is a much cleaner mission statement than most venture-backed platforms have managed with fifty times the words.

The page design also demonstrates how little is required for a site to have identity. No framework is visible. No recommendation rail appears. Nothing asks for notifications. There is no carousel explaining the brand. The site has a name, a joke, a few links, and enough stubbornness to survive for decades.

It is easy to dismiss tiny pages like this as Internet debris. But debris tells you what an environment allowed people to make. A web full of nothing but optimized commercial pages would never produce asdf.com because there is nothing obvious to optimize.

https://asdf.com/

That is the preservation value. asdf.com represents a period—and an attitude—where getting a domain and making something odd with it was a sufficient reason to publish. Four keys. One site. No growth strategy required.

Posted on by

Ancient Web: The Lenna Page Preserved the Origin Story of Computer Vision’s Most Famous Test Image

For decades, one cropped portrait showed up everywhere in image-processing research. The Lenna image became a standard test picture for compression, filtering and reconstruction algorithms, and this old City University of Hong Kong page preserves the story people once passed around with the file.

Read the Lenna 97 page

The page was first published in 1997 by Lai-Man Po and updated through 2001. It explains that the familiar test image was cropped from a November 1972 Playboy centerfold featuring Lena Söderberg, whose name was often rendered “Lenna” in technical circles.

The page also records how the image entered engineering culture. According to material collected there, researchers at the University of Southern California’s image-processing lab digitized the photograph in the 1970s for work on image processing. A later recollection from former USC programmer Chuck McManis identifies William K. Pratt as the researcher responsible for the scan used in that environment and notes that image data was distributed on 1600 BPI nine-track tape, with the color planes stored separately.

That detail alone makes the page worth saving.

A test image became infrastructure

Lenna endured because it was technically useful. The crop contains smooth areas, texture, edges, shading and fine detail—exactly the mixture needed to expose strengths and weaknesses in image-processing algorithms.

The page links the image to wavelet compression research, SPIHT, JPEG comparisons and the USC-SIPI image database. It also documents Lena Söderberg’s appearance at an imaging conference in Boston in 1997, by which point she had become an accidental celebrity within a technical community she had not originally known was using her photograph.

There is a modern complication the old page mostly predates: the image came from an adult magazine, and later researchers have questioned whether a sexualized source image should remain a default scientific benchmark. That debate matters. But historically, the image’s ubiquity is undeniable, and understanding why it spread is part of understanding digital-imaging culture.

The page is useful because it comes from the era when the story was still being assembled. It includes corrections, emails, links to research labs and references to physical distribution media that later retellings often flatten into a paragraph.

This is what personal academic pages did well: preserve the weird connective tissue around technical history before anyone knew which details would matter later.

Visit the original Lenna 97 history page

Posted on by

Ancient Web: A Confused Dragon Built a Pixel-Art Web Desktop on Neocities

Not every Ancient Web site has to be old enough to rent a car. Some matter because they deliberately keep the older personal-web idea alive. A Confused Dragon is a Neocities site built by a game programmer and hobby pixel artist who treats the browser less like a content feed and more like a place.

https://aconfuseddragon.neocities.org/

The site is often described as H2O-OS, and it sits in the loose modern tradition of web desktops: pages that borrow the visual language of operating systems, windows, icons, folders, and old computer interfaces to turn navigation itself into part of the artwork.

That is a very old-web instinct.

The creator’s work elsewhere identifies them as a programmer working in games and a hobby pixel artist. Their downloadable projects include Windows 95-inspired icon packs, nostalgic interface experiments, browser games, and large animated pixel-art pieces. An older itch.io page notes that the artist eventually moved high-resolution GIF hosting to the Neocities gallery because maintaining the work on the personal site was easier.

That decision says a lot.

A social platform can host an image. A personal site can decide what surrounds it.

The difference is visible in projects like this. The page architecture, interface references, colors, menus, backgrounds, iconography, and tiny jokes all belong to the same authored space. There is no need to squeeze the work into somebody else’s card layout or infinite scroll. The site itself becomes part of the portfolio.

This is also why Neocities keeps appearing in old-web preservation conversations even though the service is modern. It gives people inexpensive space to rediscover a behavior the commercial web mostly trained out of users: building a homepage because you want a homepage, not because a platform needs another account in its engagement funnel.

A Confused Dragon’s art is especially suited to that format. Animated pixel scenes, Windows 95 references, faux desktop elements, and nostalgic icon packs make more sense inside a deliberately constructed computer-like environment than they would as isolated posts in a feed.

https://aconfuseddragon.neocities.org/

The site is useful as a living example of continuity. The technologies changed. The instinct did not. Someone still wanted a corner of the Internet where the layout could be weird, the navigation could be personal, and the page could feel like opening somebody else’s computer in 1998 without actually having to install BonziBuddy.

Posted on by

Ancient Web: Niagara Railfan Documented the Railways Around Niagara Falls

Niagara Falls has been photographed to death. The railways around it, less so. Niagara Railfan is an old personal site focused not on the postcard waterfall but on the tracks, locomotives, yards, industrial infrastructure, and railway activity surrounding Ontario’s Niagara region.

http://niagara.railfan.net/index.html

Old directory descriptions summarize the site as a guide to railways in the Niagara region and surrounding area, with photographs, train graphics, trackside information, and links. Contemporary mailing-list posts show that the site was already active around the end of the 1990s, when its operator was announcing batches of new photographs to other rail enthusiasts.

One such update said the collection had reached roughly 285 photographs and highlighted aerial images from in and around Niagara Falls. Other updates mixed locomotive sightings, maintenance scenes, unusual rolling stock, and ordinary railway work that would never have made a tourist brochure.

That ordinary material is where old railfan sites become historical records.

Rail infrastructure changes constantly but rarely dramatically enough to make the news. A siding disappears. A signal system is replaced. A locomotive class gets retired. A yard gets rationalized. An industrial customer closes. A bridge remains but the trains using it change. Twenty years later, somebody researching the region discovers that the only easily accessible photograph of the old arrangement was taken by a hobbyist who happened to be standing there.

The site’s late-1990s context matters too. Digital photography was not yet frictionless, broadband was not assumed, and publishing hundreds of railway images online required considerably more deliberate work than dumping a phone gallery into a social feed. Pages were organized by hand because discoverability depended on structure, not an algorithm guessing what users might want next.

The surrounding railfan Internet was equally manual. Webmasters exchanged links. Mailing lists announced site updates. People posted locomotive numbers and yard reports in plain text. The web page became the stable archive while email handled the conversation.

Niagara Railfan preserves that arrangement: a subject-specific website functioning as both personal collection and public reference.

http://niagara.railfan.net/index.html

For railway researchers, local historians, or anyone trying to identify what used to run through the Niagara region before today’s equipment and track patterns, sites like this are disproportionately useful. The photographer was not trying to preserve history. He was documenting what interested him. History showed up later.

Posted on by

Ancient Web: MIT Still Hosts a Complete Technical Writing Handbook

The Mayfield Handbook of Technical and Scientific Writing is what happens when a serious reference book gets poured directly into the web without being turned into a subscription product.

Open the Mayfield Handbook at MIT

Written by Leslie C. Perelman, Edward Barrett and James Paradis, the handbook survives on MIT’s web servers as a large hypertext reference for technical communication. Its table of contents alone is enormous.

The opening sections cover the fundamentals: accuracy, clarity, conciseness, coherence, audience, document purpose, outlining, drafting, revision, review, collaborative writing, ethics and document design.

Then it keeps going.

There are sections for memoranda, meeting documents, literature reviews, laboratory reports, research reports, consulting reports, proposals, manuals, procedures, theses, oral presentations, email, websites, hypertext, résumés and notebooks. Other chapters deal with figures, tables, flowcharts, paragraphs, sentence problems, word choice, punctuation, mechanics and citation styles.

In other words, it is not a blog post titled “17 Tips for Better Technical Writing.” It is an actual handbook.

Hypertext used exactly as intended

The format is especially interesting now because it demonstrates a very straightforward use of the web: take a large reference work and break it into directly linked topics.

Need advice on writing for technicians instead of managers? There is a link. Need a section on overloaded sentences, dangling modifiers, schematic diagrams or research-report structure? There is a link. Need to remember how a technical document’s front matter differs from its body and end matter? Also a link.

Nothing forces the reader through a predetermined path. The table of contents is the interface.

That feels obvious, but modern publishing has spent years trying to make reference material behave like feeds. The Mayfield Handbook does the opposite. It assumes the reader has a problem and wants to get to the relevant page quickly.

It is also a useful artifact of early academic web publishing. Universities put substantial instructional material online because distributing knowledge was part of the point, not because every page needed to be monetized separately.

The design has aged. The information architecture has not.

For students, engineers, developers and anyone who writes documentation, the site remains surprisingly practical. More importantly for web historians, it shows how well plain linked documents can age when the structure is good and nobody decides the whole thing needs to become an app.

Browse the complete Mayfield Handbook

Posted on by

Ancient Web: One Man Documented Building His Own Microhydro Power Plant

Some DIY pages explain a project after it is finished. The Ludens microhydro page is better: it records the project while reality is still punching holes in the plan.

Visit the original microhydro build log

The author had a stream with enough flow and elevation drop to generate electricity for his home, so he set out to build a small hydroelectric plant. The page follows the job from the first trenches and pipe purchases through a forebay, penstock, turbine house, generator, Turgo runner, custom nozzles and control electronics.

Nothing is presented as magically easy.

The long penstock uses large PVC pipe, with stronger pipe reserved for the high-pressure lower section. When commercial bends were unavailable, the author built an infrared PVC pipe heater from quartz elements so the pipe could be softened and bent in the field without attacking it with a blowtorch.

A road had to be crossed. A creek needed a bridge. Air-relief valves had to be installed. A forebay was poured from concrete mixed and hauled largely by hand. At one point the concrete came out looking, in the author’s words, roughly like Swiss cheese because the aggregate mixture was poor.

He used it anyway and kept going.

Engineering as a public notebook

The page becomes valuable because it contains all the awkward parts that polished project summaries remove.

There are questions about water depth, vortex formation, pipe pressure, trash racks, sediment, nozzle size, generator mounting and jet alignment. The author repeatedly asks readers for suggestions. He documents bad cast fittings, missing seals, improvised repairs and welding done with less-than-ideal equipment.

The system eventually becomes a complete little power plant: channel, forebay, buried penstock, powerhouse, turbine and generator.

This is exactly the kind of technical page that search engines ought to love and often bury. It is long because the project was long. It has dozens of photographs because dozens of things happened. The writer does not pretend to know everything in advance, which makes the page much more useful to anyone attempting a similar build.

It also captures a style of engineering culture that thrived on personal sites: publish the process, show your failures, invite corrections, and leave the information online for the next obsessive person.

There is no funnel. No course. No lead magnet. Just several hundred lines of a person figuring out how to make falling water turn into household electricity.

Read the complete Ludens microhydro project

Posted on by

Ancient Web: BlueTongueSkinks.NET Became a Long-Running Reptile Reference

The specialist web used to be full of sites where one person picked one animal, one machine, or one hobby and then went completely overboard. BlueTongueSkinks.NET is a surviving reptile version of that model: a site built around blue-tongued skinks that became a reference point far beyond its original little corner of the hobby.

http://bluetongueskinks.net/blue.htm

References to the site’s care material show up across reptile forums going back at least to the mid-2000s. Keepers repeatedly linked newcomers to the same central guide rather than retyping the basics in every thread. That is one of the clearest signs that a personal site had become infrastructure for a community: people stopped saying “here is my opinion” and started saying “read this first.”

The surviving material goes well beyond a single care sheet.

There are species and subspecies pages, distribution notes, photographs, genus explanations, older buying and breeding pages, book recommendations, archived care discussions, and a forum ecosystem that later continued at BlueTongueSkinks.org. Some pages are obvious time capsules. A 2004 sales page talks about babies born that July, PayPal, overnight UPS shipping, and a small breeding setup run as a hobby rather than a business. A 2007 page still advertises that year’s babies.

Other pages became reference material in their own right. The genus section attempts to sort the various forms of Tiliqua and explain the naming system to hobbyists who do not have a taxonomy textbook open beside the terrarium. Individual species pages preserve descriptions, range information, and older hobby knowledge about forms such as Northern, Indonesian, Merauke, and Irian Jaya blue tongues.

That last category also illustrates why old animal-care sites need to be read historically rather than treated as frozen authority. Husbandry standards change. Taxonomy changes. Import knowledge changes. Even the site’s own community has noted that some older information remains useful while other pieces are dated.

That does not reduce the archival value. It increases it.

You can watch a reptile-keeping community build a shared body of knowledge before Facebook groups and short-form video took over much of hobby discussion. The old pages show what keepers believed, what animals were entering the trade, which books mattered, what terminology was common, and which questions kept being asked year after year.

http://bluetongueskinks.net/blue.htm

For current husbandry, use current veterinary and herpetological guidance. For understanding how a niche Internet community taught itself and each other for more than twenty years, BlueTongueSkinks.NET is the artifact.