Category: Ancient Web

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

Posted on by

Ancient Web: DF7CB Is a Ham Radio Site That Never Forgot the Web’s Personal Side

Christoph Berg’s DF7CB page is what happens when a technically minded personal website stays useful for decades instead of being rebuilt into a content platform. It is part ham-radio logbook, part technical notebook, part biography, and part evidence that the old personal Web never completely disappeared.

Visit DF7CB

Berg was first licensed in 1994 as DG8VD and received the DF7CB call sign after upgrading in 1995. His page records the long arc of the hobby unusually well: early CW work, packet radio on 70 centimeters, a long quiet period, and a return to amateur radio in 2018 with modern digital modes and renewed interest in contesting.

A Site Built Around Actual Activity

The useful thing here is that the page does not merely announce that somebody is a radio amateur. It links into an online logbook, contest activity, a collection of unusual QSL cards, and technical material including a QRP14 manual and schematics.

That mix matters. QSL cards are physical artifacts from radio culture, while the online logbook is a living record. The technical documents preserve the practical side of the hobby. Together they form the kind of small archive that search engines often treat as just another personal page even though it contains material a radio historian or operator may actually want.

Berg’s biography also captures a familiar crossover between amateur radio and computing. He drifted away from radio while becoming deeply involved with open source and Debian Linux, then eventually returned to the air. The website reflects both worlds: terse navigation, direct links, almost no decorative machinery, and content that assumes the visitor can click around without being herded through a funnel.

Still Updated, Still Personal

One of the best details is that this is not a frozen 1990s relic. The page was still being updated in 2026, including recent contest call-sign information. It survived because the underlying model was simple: own a domain, publish useful pages, keep the files intelligible, and update them when something changes.

Modern sites routinely bury less information behind more interface. DF7CB goes the other way. The page tells you who built it, what he does, what equipment and activities matter, and where the interesting documents live.

That is old-Web information architecture in its natural habitat: somebody knows something, puts it online, and leaves enough structure for other people to find it.

Explore the DF7CB amateur-radio pages

Posted on by

Ancient Web: DJ Riddler’s Rave Archive Preserves the Early 1990s From the Ground

DJ Riddler’s oldskool rave website does not approach early-1990s rave culture as a retrospective assembled from press releases and celebrity interviews. It is a participant’s archive: photographs, flyers, newspaper clippings, records, tapes and memories collected by somebody who was there.

Visit DJ Riddler’s oldskool rave archive

The site says its material is primarily from 1991-1993, though the collection and website continue beyond that period. As of May 2026, the homepage was still being updated.

Its navigation reads like the contents of several boxes pulled from an attic: rave photographs, newspaper clippings, flyers, a vinyl collection, mixtapes and MP3s, stories, pirate-radio tapes, clothing, slides and video material.

That is exactly why it is useful.

Primary Sources Instead of a Clean Retrospective

Rave history is easy to flatten into famous clubs, famous DJs and a handful of familiar photographs. Personal collections preserve the parts that broad histories tend to lose: local flyers, newspaper panic, clothing, anonymous crowds, tapes passed between people and the visual language of events that existed for one night and disappeared.

The Riddler archive includes material such as a 1993 “Close up North” rave reference, late-1980s newspaper coverage of acid-house panic, scanned flyers and recordings associated with pirate radio. The site’s owner is blunt about why he keeps it. He argues that later television histories often fail to represent what the scene actually felt like on the ground.

You do not have to accept every personal interpretation to recognize the archival value. Firsthand collections are useful precisely because they preserve a point of view alongside the artifacts.

The Mess Is Part of the Record

The current homepage still contains notes to the webmaster himself: photographs that need sorting, navigation that needs changing, scans that need redoing and material waiting to be added from an older version of the site.

On a corporate site those notes would be embarrassing. Here they make the project more legible. This is not a finished museum exhibit. It is a working personal archive that has been migrated, repaired and expanded over time.

For researchers interested in British rave culture, the strongest material is likely in the photographs, flyers, clippings and audio references. For web historians, the site is another example of why personal websites matter: people preserve things institutions do not know they should care about until decades later.

A box of flyers in somebody’s house is fragile. A scanned copy on a public webpage can suddenly become searchable by the entire world.

Explore DJ Riddler’s photographs, flyers and rave-history material

The archive is unfinished. Good. So was the scene it is trying to remember.

Posted on by

Ancient Web: Kenji Miyazawa’s Ihatov Still Lives in a Hand-Built Literary Website

Kenji Miyazawa created an imaginative version of his native Iwate called Ihatov. Decades later, a hand-built website still maps that literary world through climate, landscape, translations, essays and character guides.

Visit Kenji World’s Ihatov page

The English-language section introduces Ihatov through “The Four Seasons in Ihatov” and “Natural Features of Ihatov,” then branches outward into biography, characters, essays, a gallery and translations of Miyazawa’s work.

The structure feels less like a modern author landing page and more like a volunteer-built reference library, because that is effectively what it is.

Literature Anchored to Place

Miyazawa’s work is closely tied to rural northeastern Japan, and the site leans into that connection instead of treating the stories as isolated texts.

A quotation from “An Outline of the Essential Art of the Peasant” appears near the entrance to the Ihatov section, framing countryside and ordinary life as material for a larger creative world.

The site then gives readers several ways in: environmental context, characters, translated works and essays about Miyazawa. For English-speaking readers, that layered approach is useful because it supplies cultural and geographic context that a bare translation cannot.

The Old Web as a Translation Workshop

One of the quiet strengths of older literary websites is how much translation labor they preserve. The English section includes translated works for reading and downloading, alongside essays and explanatory material.

That sort of work often existed outside commercial publishing. Someone cared enough to make the material accessible, organize it, and keep the pages online.

The visual design now reads unmistakably as an older website, but the information architecture remains understandable. The links lead outward through a small literary ecosystem rather than funneling visitors toward an account signup or recommendation algorithm.

That makes Kenji World valuable as both a literary resource and a surviving example of what enthusiast scholarship looked like on the independent Web.

Explore the world of Kenji Miyazawa

Posted on by

Ancient Web: One Man’s Love Letter to a 1984-1988 Honda Civic 1500S

There are automobile reference sites, restoration forums and auction databases. Then there is this page: one man explaining, in considerable emotional detail, why his gray-and-silver 1984-1988 Honda Civic 1500S hatchback was basically automotive perfection.

Visit the Civic 1500S page

The owner had the car for 18 years and about 125,000 miles before selling it. The page records a rare gray/silver two-tone finish, Wolf OEM alloy wheels, air conditioning, trim, floor mats, a replacement Recaro driver’s seat and an aftermarket Sony radio-cassette system.

It also records something databases are terrible at preserving: attachment.

A Classified Ad Written by Somebody in Mourning

The page is divided into sections called Love, …and Loss (sold) and Beauty Incarnate. This is not neutral automotive journalism.

The owner describes the compact hatchback as exceptionally well proportioned and admits that seeing it parked under city lights could leave him speechless. When garage space became a problem, he prepared the car for a final photo shoot and sold it to somebody identified as “Dolphin Dave.”

The asking price was $1,000.

He also documented flaws rather than pretending the car was museum-perfect: old body damage, a poorly resprayed front panel, bubbling paint and ordinary wear. The page says there had been significant interest despite those issues.

At the bottom is a small call for information. The owner wanted to hear from people who worked on the third-generation Civic at Honda, knew specialists who restored them, or had an absolutely perfect gray Civic S for sale.

That combination makes the page more useful now than it probably seemed when it was written. It captures options, modifications, condition, ownership history and enthusiast language around a specific trim level before 1980s Japanese cars became widely collectible.

The Personal Web Preserved Sentiment Along With Specs

Modern automotive sites are excellent at specifications and market prices. Personal pages preserved something different: why an ordinary object mattered to one person before anybody decided it was historically interesting.

That is useful evidence. Enthusiast culture is not created retroactively when auction prices rise. It begins when somebody is taking medium-format photographs of an aging Civic because selling it hurts.

The page closes with a line telling readers to love their cars but live their lives so they can walk to work. That is a much stranger and more human ending than “subscribe for more Honda content.”

See the surviving Civic 1500S page and photographs

For once, calling a used Honda “Beauty Incarnate” does not feel like ad copy. The seller already sold it.

Posted on by

Ancient Web: The Kitchen Project Turned Sauerkraut Into a Tiny Food-History Archive

There are websites that explain how to make sauerkraut, and then there are websites that decide fermented cabbage deserves two thousand years of history, family recipes, etymology, folk medicine and side arguments about who invented what.

Visit The Kitchen Project’s sauerkraut history

The page belongs to The Kitchen Project, a long-running cooking site with a strong German-American family-history streak. Its sauerkraut material combines instructions for fermentation with stories about how preserved cabbage traveled through cultures and became associated with German cooking.

The site is not written like a modern recipe platform. It wanders. It argues with itself. It links sideways into family recipes and old food-history pages. That is part of the charm.

Food History Before Every Recipe Needed a Life Story

The page explains the basic lacto-fermentation process: shredded cabbage, salt, released water and an acidic environment that allows fermentation while discouraging spoilage.

From there it moves into the word “sauerkraut,” Chinese and European preservation traditions, Germany, Captain Cook, vitamin C and various historical claims. Some of those claims are presented more confidently than a modern academic treatment would probably allow, so the page is best read as a period food-history resource rather than a definitive scholarly source.

That distinction does not make it less interesting.

The wider Kitchen Project also preserves German family recipes and material connected to the author’s grandmother, Emma Block, whose cooking traditions were brought to the United States. Pages for sauerkraut, breads, soups and other dishes sit together as part cookbook and part family archive.

A Recipe Site That Became a Time Capsule

The page still carries the visual and structural habits of an older commercial Web: long scrolling sections, inline links, miscellaneous side material and a willingness to put several different ideas on one page.

Modern food sites are usually cleaner. They are not always richer.

The Kitchen Project shows how personal history, practical instruction and loose research often lived together online before publishing systems forced everything into standardized templates.

Explore The Kitchen Project

Posted on by

Ancient Web: The Website Obesity Crisis Was Warning Us in 2015

In 2015, Maciej Ceglowski stood up at Web Directions in Sydney and explained that ordinary text pages were becoming absurdly large. The surviving text of The Website Obesity Crisis is still online, and the punchline has aged badly for the Web because the problem did not exactly go away.

Read The Website Obesity Crisis

Ceglowski’s target was not rich web applications, video or sites where large media files actually served a purpose. He was talking about pages that mostly contained text but somehow required megabytes of assets, advertising and JavaScript to display it.

His measuring stick was deliberately ridiculous and therefore memorable: a text-based webpage should not weigh more than a major work of Russian literature.

When an Article Outweighs the Book It Mentions

The talk runs through examples of page bloat with the enthusiasm of somebody discovering increasingly large animals in his kitchen.

A 2012 article warning that the average webpage had passed one megabyte was itself around 1.8 MB. A later article warning about page growth had become about 3 MB. Ceglowski compared individual tweets and short articles with the full digital text of novels such as Crime and Punishment, War and Peace and The Master and Margarita.

The exact page sizes are historical snapshots, but the method remains useful: compare the amount of information a page delivers with the amount of machinery required to deliver it.

He also attacked the industry’s habit of responding to bloated websites with additional layers of technology. Faster formats, special mobile systems and optimization products can help, but they can also avoid the uncomfortable question of why a few paragraphs needed so much infrastructure in the first place.

The Page Demonstrates Its Own Argument

The surviving talk is itself mostly text, headings, links and images where needed. It loads as a straightforward document and remains readable more than a decade later.

That matters for the Ancient Web project because old pages frequently survive precisely because they were not elaborate. Their creators accidentally practiced long-term preservation by using less software.

For modern webmasters, this is not an argument that every site should look like 1996. It is an argument for remembering that every dependency, tracker, framework and asset has to earn its place.

A page can be visually excellent and still be technically restrained. Those ideas were compatible before the industry convinced itself otherwise.

Read the complete 2015 Website Obesity talk

The funniest part is that a decade later, comparing a webpage to War and Peace no longer sounds like satire. It sounds like a benchmark.

Posted on by

Ancient Web: The Great War Primary Documents Archive Has Been Serving World War I Sources Since 1995

Some old websites survived because somebody forgot about them. The Great War Primary Documents Archive survived because people kept using it.

Visit the Great War Primary Documents Archive

The archive says it launched in November 1995. Its visitor counter passed one million in 2001 and fifteen million by March 2008. More importantly, the site still organizes a remarkable amount of World War I material into treaties, official papers, documents by year, biographies, photographs, memoirs, medical material and maritime history.

The site was last updated as recently as 2024, which makes it a rare hybrid: a Web structure from the 1990s carrying a collection that never entirely stopped being maintained.

The Maritime War Is Its Own Rabbit Hole

The naval section is especially deep. Rather than treating the war at sea as a footnote, the archive gives it dedicated space under “The Maritime War,” alongside the larger document collection.

That matters for researchers because World War I naval history is scattered across fleet orders, reports, memoirs, ship histories and diplomatic records. A hand-built index can still be more useful than searching blindly for isolated documents.

The larger archive includes material such as prewar diplomatic documents, wartime statements, Armenian records, chronological event lists and biographical references. It is the sort of collection where one useful page tends to lead to twelve more.

A 1995 Information Architecture That Still Works

There is nothing mysterious about the design. Large categories. Text links. Sub-indexes. A keyword search box. The value comes from editorial organization rather than interface novelty.

That is one of the recurring lessons of the Ancient Web project: durable sites often separate content from fashion by accident. A page designed around hyperlinks and headings can remain understandable decades later with almost no technological drama.

The Great War Primary Documents Archive is not merely nostalgic old HTML. It remains a serious research starting point and a reminder that much of the useful early Web was built by people manually organizing knowledge.

Explore the World War I archive

Posted on by

Ancient Web: Futureware 2001 and the Open-Source Workshop From Vienna

Futureware 2001 looks like somebody opened a cupboard from the turn-of-the-millennium European software scene and forgot to close it. The Austrian group described itself simply as people working together to produce software and funware, and much of that old project structure is still sitting online.

Visit Futureware 2001

The page preserves news from the period when SourceForge was becoming central infrastructure for open-source projects. Futureware announced that its projects and mailing lists had moved there, advertised for programmers, graphics people, musicians, designers, writers and even crypto specialists, and listed the people and organizations helping host its site.

One surviving note dates the Linux release of a project called Würstelstand to a Linux Party in Vienna on August 27, 1999.

That sentence alone could power several hours of Internet archaeology.

*Quick Tried to Be a Universal Lookup Box

One of the better documented Futureware projects is *Quick, an open lookup and translation system. It could translate words, expand acronyms, explain error numbers, identify postal codes and search additional databases.

The project offered separate software for Linux/Unix and Windows, source code for other systems, and downloadable data sets. Website owners could even paste a small HTML form into their own pages to query the service remotely.

The surviving documentation mentions KDE/Linux software, Windows software, localization files and third-party databases. Available interface translations included German, Dutch and Croatian. This is very recognizably pre-cloud software: download the program, download the data, put the pieces on your own machine and make it work.

Distributed Rendering Before Everybody Called Everything a Cloud

Another surviving Futureware branch, RenderWorld, attempted to pool unused computers into a global Blender render farm. Users could contribute idle CPU cycles, while artists submitted models for distributed rendering. Finished output could be returned as PNG frames, ZIP archives or video.

The basic idea now sounds familiar because distributed computing and hosted rendering are ordinary concepts. The implementation belongs to an earlier Internet, when volunteer CPU time, screensavers and downloadable clients were common ways to imagine global computing projects.

Futureware is valuable because the site preserves the messy middle layer of software history. Not the famous operating systems or billion-dollar companies, but small groups experimenting with Linux, translation databases, distributed processing and open collaboration while the tools around them were still young.

Explore the surviving Futureware project pages

It is difficult to ask for a more period-correct artifact than a software collective recruiting programmers, musicians and crypto specialists while moving its mailing list to SourceForge.

Posted on by

Ancient Web: The NeXT Computer Historical Site Preserves the Black Cube Before It Became a Legend

The NeXT Computer is remembered now as the black machine tied to Steve Jobs, object-oriented software and the ancestry of modern Apple operating systems. Simson Garfinkel’s NeXT Computer Historical Site preserves something better than hindsight: the original material.

Visit the NeXT Computer Historical Site

The site collects NeXT brochures, hardware specifications, articles, screenshots, software information, icons and magazine material. It even includes hand-drawn notes from the NeXT unveiling in San Francisco on October 12, 1988.

That one item alone makes the page worth keeping.

The Machine as NeXT Wanted You to See It

One preserved brochure introduces the NeXT system as “the first computer of the 1990’s,” the sort of claim that sounds wonderfully period-specific because it is. The marketing emphasizes higher education, graphical software development, multimedia and an integrated workstation environment.

The technical pitch was not empty. The original NeXT Computer used a Motorola 68030 processor, built-in Ethernet, a digital signal processor, high-resolution graphics and a magneto-optical drive. Its software combined UNIX foundations with a graphical development environment that included tools such as Interface Builder.

Other archived material is less flattering, which makes the collection better. A Macworld article discusses NeXT’s high price, software shortage and performance problems while also recognizing the influence of its object-oriented development tools and interface ideas.

Why Original Brochures Beat Modern Summaries

A modern history can explain what NeXT eventually meant. Contemporary brochures reveal what people thought it meant before the ending was known.

That distinction matters.

The brochures show which features NeXT considered revolutionary. The magazine coverage shows which problems reviewers noticed. Screenshots and icons show how the system actually presented itself. Garfinkel’s handwritten launch notes preserve the event at human scale rather than as corporate mythology.

For researchers, this is the kind of small historical collection that becomes increasingly valuable as original computing material scatters across dead sites, auction listings and private collections.

Explore the NeXT historical archive

Posted on by

Ancient Web: A Garfield Postcard Collection Frozen in Simple HTML

Not every old website needs a mission statement. Sometimes somebody owns a pile of Garfield postcards, scans them, makes a page, and calls the job finished. The Garfield postcard page at garfield.dk is exactly that kind of artifact.

Visit the Garfield postcard page

The page is almost aggressively simple. A heading says GARFIELD POSTCARDS, followed by a grid of postcard thumbnails. Each image can be clicked for a larger view. There is very little commentary, no collection-management interface and no attempt to turn the material into an article.

That minimalism tells us something about the period of the Web this kind of page came from.

The Collector Page Was Its Own Genre

Before image-heavy social platforms, collectors routinely built small pages around whatever they happened to own: cereal boxes, ticket stubs, toys, postcards, trading cards, autographs, model cars, matchbooks. The point was not necessarily to become an authority or monetize an audience. The page itself was the display case.

The Garfield page survives in that form. Roughly twenty postcard images sit in rows with the repeated instruction to click to enlarge. The page identifies Garfield and links outward toward the official Garfield presence, but otherwise stays out of the way.

It is a useful reminder that the old Web was full of small enough websites. A page did not need a publishing schedule. It did not need enough material to justify a brand. It could simply answer one question: what does this person’s Garfield postcard collection look like?

The Domain Gets Stranger if You Wander

The broader garfield.dk domain also contains unrelated old technical material, including pages on mobile and distributed computing. That sort of collision is another old-Web signature. Domains often accumulated whatever their owners were working on over the years instead of being kept within a single content niche.

Modern publishing systems tend to encourage separation: one site for professional work, another for hobbies, a social account for collections, cloud storage for personal documents. Older personal domains frequently behaved more like hard drives exposed to the world.

For researchers, pages like this preserve ordinary material culture that formal archives may never bother collecting. For webmasters, the lesson is even simpler: if the content is the point, a heading, thumbnails and links can be enough.

Browse the Garfield postcards

No login. No carousel framework. Garfield remains visible despite these hardships.