Category: Ancient Web

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

Posted on by

Ancient Web: The Primary Water Institute Kept a Disputed Hydrology Theory Alive Online

The old Web did not merely preserve consensus knowledge. It preserved arguments people refused to let die.

Visit the Primary Water Institute

The Primary Water Institute promotes the idea of primary water: water generated or stored deep within Earth and pushed upward through geological structures, potentially providing large usable supplies independent of ordinary rainfall-fed aquifers.

The site goes further than simply discussing deep-Earth water. It argues that this resource may be vastly abundant, that springs and wells can tap it through faults and fractures, and that the idea has historically been suppressed or ignored.

That is the site’s claim. It is not the mainstream hydrological position.

The U.S. Geological Survey describes most groundwater as precipitation that infiltrates soil and rock, recharging aquifers over time. Deep-Earth or “juvenile” water is a real geological concept, but that does not establish the much stronger claim that enormous quantities of easily accessible, potable primary water are waiting to solve ordinary water-supply problems.

That distinction is exactly why this site is worth preserving as a historical artifact.

The Web as an argument archive

The Institute’s homepage offers a downloadable Primary Water Magazine and frames primary water as a potential answer to global water scarcity. It also invokes Biblical and ancient references, geological pressure, mountain springs and desert oases as part of the case.

The modern temptation is to reduce a page like this to a binary label—TRUE or FALSE—and move on. That misses what an archival website can show.

The interesting question is not only whether the theory won. It is how the theory was presented, which evidence its advocates emphasized, what historical narrative they constructed, and how they used the Web to route around institutions they believed had dismissed them.

That makes the Primary Water Institute a useful specimen of scientific dissent on the independent Web.

You can see the rhetorical pattern directly: an under-recognized resource, a forgotten researcher, establishment resistance, practical demonstrations, and a global problem waiting for the suppressed solution. That pattern appears in legitimate scientific revolutions, failed theories, pseudoscience and everything in between. The pattern itself proves nothing. Evidence still has to do the work.

The page is therefore most valuable when read with a second tab open.

Read the Institute’s case. Then read conventional hydrology. Compare the definitions. Compare the evidence. Ask what would actually distinguish precipitation-derived groundwater from a deep juvenile source.

That is a much healthier use of an old controversial website than either blindly believing it or deleting it from cultural memory.

CacheRat’s 1,967 Ancient Web Domains research list is meant for exactly this kind of material: sites that preserve not just facts, but abandoned arguments, minority theories and the strange intellectual side roads of the public Internet.

The Web used to make room for people to publish the whole case themselves.

Sometimes they were right. Sometimes they were wrong. Either way, the record matters.

Read the Primary Water Institute’s case

Posted on by

Ancient Web: Perfect Sound Forever Was Calling Out Bad Songs in 1997

Before reaction videos, ranked-list factories, and social feeds turned musical disagreement into an industrial process, an online music magazine could simply ask people to explain which songs they thought were terrible. Perfect Sound Forever did exactly that in February 1997.

Read “Bad Songs” at Perfect Sound Forever

Perfect Sound Forever describes itself as an online music magazine that has been around since 1993, making it an unusually early example of independent Web publishing devoted to music. The “Bad Songs” page, credited to editor Jason Gross, is a good specimen of what that publishing culture looked like when participation still meant sending in words rather than tapping an angry face.

The concept is simple: collect opinions about songs people cannot stand and let the contributors make their cases. What survives is not merely a list of titles. It is argument—subjective, occasionally ridiculous, sometimes persuasive, and very human.

That distinction matters because old Web writing often preserved individual voices more clearly than modern engagement systems do. Today a disliked song may accumulate thousands of scores, comments, reaction clips, and algorithmically sorted hot takes. A 1997 webzine page had limited space and no engagement dashboard. Somebody had to care enough to compose a complaint and somebody else had to publish it.

Perfect Sound Forever itself is the larger artifact. Independent Web magazines were one of the first demonstrations that publishing no longer required owning a press, buying broadcast time, or persuading a traditional magazine to give you column inches. A site could develop its own editorial personality, contributors, recurring subjects, and readership with comparatively little infrastructure.

The design also leaves the writing in charge. There is no infinite scroll fighting the article, no autoplaying recommendation rail, and no demand that every opinion become a video. The page is text because the text is what somebody came to read.

For researchers studying early online music criticism, fan culture, or Web publishing, that kind of survival is useful. It captures not only what people thought about particular songs but how an independent online magazine structured participation in the mid-1990s.

And, naturally, it preserves one of humanity’s oldest cultural traditions: telling other people that the music they like is garbage.

Browse the original “Bad Songs” page

The technology aged. Musical disagreement appears to be doing fine.

Posted on by

Ancient Web: The 1998 Linux Tea Party Put Silicon Valley Linux in One Room

In November 1998, Linux in Silicon Valley was still small enough that a gathering page could list people, employers, food, and the network setup without sounding absurd. Marc Merlin’s surviving page for the Silicon Valley Linux Users Group Linux Tea Party is a tiny snapshot of that moment.

Visit Marc Merlin’s Linux Tea Party page

The event took place on November 25, 1998. The page records attendees connected with names that now read like a compact history of commercial Linux: Red Hat, VA Research, Caldera, SGI, Linuxcare, and others. It also records the mundane infrastructure that made a late-1990s computer gathering a computer gathering: people brought machines, a 10BaseT network was arranged, and food and drinks were part of the plan.

That combination is what makes the page interesting. A corporate history can tell you which companies existed. A product timeline can tell you which distributions shipped. This page tells you that the people involved were in the same room, plugging hardware into Ethernet and eating together.

Old event pages are often better cultural evidence than polished retrospectives because nobody wrote them expecting historians to care. They were logistics. Names, affiliations, directions, equipment, and social details were included because attendees needed them. Decades later those details are precisely what make the page valuable.

The date matters too. In 1998, Linux had already developed a serious technical community, but it had not yet become invisible infrastructure under cloud services, phones, routers, televisions, and half the machinery people use without thinking about an operating system. The Tea Party catches Linux at the stage where participation still felt like joining a scene.

For Web archaeologists, the page is also a reminder to preserve event material. Conference schedules, user-group pages, release parties, mailing-list meetups, and photo indexes often look disposable once the event is over. They are not. They document who knew whom, which organizations overlapped, and what a technical community considered normal at a particular time.

The page survives without needing a documentary voice-over or retrospective redesign. It simply remains what it was: notes around a Linux gathering in 1998.

Read the Linux Tea Party page

Sometimes Internet history is not hidden in a manifesto. Sometimes it is sitting in an attendee list next to a note about Ethernet.

Posted on by

Ancient Web: Hugh Smith Made a Photo Catalog of Fungi That Grow on Dung

The modern Web is full of sites trying desperately to be about everything. Hugh Erle Smith built a page about mushrooms growing on poop.

That is better.

Visit Hugh Smith’s Dung Fungi page

The page is a simple photographic catalog titled Dung Fungi. Smith identifies the fungus, where and when he photographed it, and—critically for this particular branch of natural history—what kind of dung it was growing on.

Horse manure. Cow manure. Moose dung. Deer dung. Elk dung. Dog feces.

There is no algorithm attempting to determine whether you are emotionally prepared for Phycomyces blakesleeanus on dog shit. You clicked Dung Fungi. You knew what business you were getting into.

Field notes disguised as a web page

The value here is not visual sophistication. It is accumulated observation.

Smith’s examples span years and locations. Entries on the page include Panaeolus semiovatus photographed at Fort Bragg in 2001 and 2011 on horse manure; Panaeolus papilionaceus on cow manure; Protostropharia alcis on female moose dung in Alaska in 2011; Coprinopsis nivea on horse and moose dung; and multiple examples of Phycomyces blakesleeanus growing on dog feces.

That structure turns a gallery into a compact field record. Species, substrate, place, year. Four pieces of metadata, repeated consistently, become more useful than a page full of SEO paragraphs explaining that fungi are “fascinating organisms.”

It also illustrates a recurring strength of personal specialist sites: weird subjects do not need permission.

A magazine editor might reasonably ask whether a multi-year photographic archive of coprophilous fungi could support an issue. A social platform would probably bury it unless somebody found a way to make a mushroom on elk dung controversial. A personal website only has to answer one question: does the person maintaining it care enough to put it online?

Smith clearly did.

The page also preserves the visual culture of the scientific hobbyist Web. Thumbnails lead to larger images. The browser is expected to behave like a browser. The visitor is trusted to click around. No carousel hijacks the mouse. No modal asks for an email address before revealing the moose shit.

That matters because natural-history knowledge is often built from many small observations, not a single grand publication. A photograph with a date, location and substrate can remain useful long after the software around it looks ancient.

CacheRat’s 1,967 Ancient Web Domains research list is full of pages like this: aggressively narrow, deeply personal, and far more interesting than their traffic numbers would ever suggest.

The old Web had room for somebody to spend years documenting mushrooms on animal droppings and publish the evidence for anyone who cared.

That is what a healthy information ecosystem looks like.

Browse Hugh Smith’s Dung Fungi photographs

Posted on by

Ancient Web: Leffingwell Built a Technical Library Around the Sense of Smell

The sense of smell gets treated casually until somebody loses it, studies it, formulates fragrances around it, or tries to explain how several hundred receptor types turn airborne molecules into perception. Leffingwell’s olfaction pages are built for the people who kept asking the next question.

Visit Leffingwell’s olfaction reference

The page opens into a specialist collection covering the sense of smell, olfactory receptors, receptor genes, anosmia and hyposmia, research links, and related organizations. It is less an article than a hand-built doorway into a technical subject.

That structure is worth noticing. Older specialist websites often behaved like personal reference desks. The creator did not need to flatten the subject into a beginner-friendly content funnel. A page could simply say, in effect: here are the important concepts, here are the tables, here are the researchers, and here are the next twenty doors.

Olfaction is particularly suited to that treatment because it crosses several fields at once. Biology explains receptors and neural pathways. Chemistry deals with volatile compounds. Medicine deals with smell disorders. Flavor and fragrance work adds its own vocabulary and practical concerns. One useful page can therefore become a junction between subjects that would otherwise be scattered across different libraries.

Leffingwell’s site preserves that junction. Its tables and linked pages retain terminology that helps researchers move beyond broad searches such as “how smell works.” A reader can encounter receptor families, smell dysfunction terminology, treatment and research organizations, and older scientific resources from within the same small Web neighborhood.

In 2026, that is useful for reasons beyond nostalgia. Search engines are excellent at answering common questions, but specialist research often begins when the common question stops being good enough. Old reference pages can expose vocabulary and connections that tell you what to search for next.

The site is also a good reminder that a useful technical website does not have to impersonate a journal, magazine, social network, or software application. Sometimes an organized collection of links, tables, terminology, and domain knowledge is exactly the product.

There are thousands of pages on the Web about sensory science. Very few feel like somebody’s actual working bookshelf.

Explore the olfaction material at Leffingwell

That is what makes this one worth keeping around.

Posted on by

Ancient Web: Guardian’s Egypt Built a Great Pyramid Primer Before Everything Became Video

There is a certain type of old website that feels less like a page and more like somebody quietly built a textbook in HTML.

Visit Guardian’s Great Pyramid page

Guardian’s Egypt is one of those sites. Its Great Pyramid of Khufu page does not open with a cinematic drone shot, a newsletter popup, or a three-minute preamble begging you to subscribe. It opens with measurements.

Original height: 146.6 meters. Current height: 138.75 meters. Side length: 230.37 meters. Estimated volume: 2,521,000 cubic meters.

Then it keeps going.

The page walks through the pyramid’s major passageways and chambers, including the descending passage, subterranean chamber, ascending passage, Grand Gallery, so-called Queen’s Chamber, antechamber, King’s Chamber, relieving chambers, well shaft, grotto, mortuary temple, boat pits, and satellite pyramids.

This is not just a tourist summary. It is organized like a research primer.

A web page with a bibliography

The page credits Andrew Bayuk and identifies the material as an excerpt from Guardian’s Ancient Egyptian Pyramid Primer, with copyright dates spanning 2000–2005. It also includes a bibliography pointing readers toward works by I.E.S. Edwards, Ahmed Fakhry, Zahi Hawass, Mark Lehner, J.P. Lepre, W.M.F. Petrie and others.

That matters because old specialist sites often expose their scaffolding. You can see where claims came from. You can see which diagrams were borrowed from which books. You can see where the author is describing accepted archaeological interpretation and where a possibility is being discussed.

The page also pushes back on mystery-first pyramid storytelling. It explicitly notes that speculation about the Great Pyramid is abundant while archaeological evidence still provides a great deal to work with. It discusses workmen’s markings, structural features and construction evidence instead of treating every empty space as proof of aliens, Atlantis or whatever YouTube discovered this week.

Some details have naturally aged. Archaeology keeps moving, and a page assembled mainly in the early 2000s should not be mistaken for the final word on Giza. But that is part of what makes the site interesting now: it preserves how a serious enthusiast assembled and presented pyramid research on the public Web two decades ago.

The design helps. Sections are addressable. Images sit next to explanations. Sources are named. The page can be read top to bottom or used like a reference shelf.

CacheRat’s 1,967 Ancient Web Domains research list exists to find exactly this kind of surviving specialist work: not famous enough to dominate search, but detailed enough that losing it would make the Web dumber.

The Great Pyramid has survived roughly 4,500 years. Its early-Web fan scholarship probably deserves a backup plan too.

Explore Guardian’s Great Pyramid page

Posted on by

Ancient Web: Alps Roads Has Been Photographing Highways Since 1997

Some people photograph mountains. Some people photograph architecture. Alex Nitzman photographs the infrastructure that gets everybody there. Alps Roads has been online since 1997, documenting roads, streets, highways, signs, bridges, and the miscellaneous concrete machinery of getting from one place to another.

Visit Alps Roads

The site is immediately recognizable as a project that grew over time rather than being rebuilt every few years around whatever a design agency thought a homepage should look like. It offers road material directly, including mirrors and accommodations for older browsers. Its footer runs from 1997 through 2026. That is not simulated retro styling. It is continuity.

The photographs are the important part. Roads change constantly in ways that are easy to overlook while the work is happening. Signs are replaced. Interchanges are rebuilt. Lane markings change. Old route shields disappear. Bridges are widened or demolished. A temporary traffic pattern becomes permanent, and ten years later it can be difficult to remember what was there before.

A road enthusiast with a camera therefore becomes an accidental historian.

Alps Roads contains the kinds of images that official transportation agencies often have little reason to preserve publicly: ordinary approaches to interchanges, sign assemblies, local streets, unusual route configurations, and roadside details that seemed unremarkable on the day they were photographed. The older those images become, the less ordinary they are.

There is an interesting Web-design lesson here too. Photography sites today are frequently treated as galleries first and references second. Alps Roads organizes images around places and road systems. The point is not an algorithmic stream of attractive pictures. The point is that somebody looking for a particular highway can find the material associated with it.

That makes the collection useful to several groups at once: road enthusiasts, local historians, sign researchers, transportation nerds, and people trying to answer the wonderfully specific question, “What did this interchange look like before they rebuilt it?”

Projects like this are why independent websites remain worth finding. The 1,967 Ancient Web Domains research list is full of similar cases where one person’s narrow interest quietly became a historical resource after enough years passed.

Explore the road photography at Alps Roads

A highway page from 1997 does not need to have predicted the future. It only needed to keep the photographs.

Posted on by

Ancient Web: Chocolate Wrappers Turned Candy Packaging Into a Global Database

Most chocolate wrappers are engineered for a very short career: attract somebody in a shop, keep the chocolate clean, then go in the trash. ChocolateWrappers.info looked at that arrangement and apparently decided the trash part was unacceptable.

Visit ChocolateWrappers.info

The result is a collector’s database organized across countries, manufacturers, brands, and individual pieces of packaging. The site’s navigation immediately gives away the scale of the project. Instead of a handful of favorite bars, there are country indexes and long manufacturer lists leading into wrapper after wrapper.

This is one of those collections whose value becomes clearer with age. Packaging is commercial ephemera. Companies redesign logos, change weights, merge brands, alter ingredients, enter new countries, leave old ones, and quietly retire products. A wrapper that was completely ordinary when somebody bought it can become evidence of what a company sold, how it marketed itself, what language it used, and what consumers saw on shelves at a particular moment.

The Web was unusually good at enabling this kind of collecting. Before online publishing, a private collection could be enormous and still remain effectively invisible unless you knew the collector personally or encountered an exhibition. A basic website could turn boxes of fragile paper and foil into a reference available to another collector on the other side of the planet.

ChocolateWrappers.info feels built around that older model of the Web. The information architecture follows the collection rather than a feed. You browse geographically, then narrow toward makers and individual items. English and Czech navigation sit alongside the material, emphasizing that this was not assembled for some generalized global-content strategy. It grew from the collection itself.

For researchers, the useful part is not merely that old wrappers are pretty. Packaging can help date products, trace brand ownership, compare regional releases, identify discontinued candy, and document changes in graphic design. Collectors already understand that an object becomes more useful once somebody bothers to classify it. The website extends that classification beyond the physical collection.

There is also a lesson here for modern websites: boring structure can be a feature. A large specialist archive benefits from predictable indexes, descriptive labels, and direct paths through the material. It does not need to manufacture a reason for visitors to keep scrolling. The wrappers are the reason.

The site is still worth exploring precisely because almost nobody expects somebody to have built a global database of disposable chocolate packaging—and yet somebody did.

Browse the Chocolate Wrappers collection

The old Web contains a surprising number of useful databases created from things everybody else threw away.

Posted on by

Ancient Web: Dan Hates Spam Turned Junk Email Into a Legal Project

The early commercial internet produced spam almost immediately. It also produced people angry enough to turn the spam into evidence.

Visit Dan Hates Spam

DanHatesSpam.com is Daniel Balsam’s public-service site about unsolicited commercial email and the legal fights that followed it.

Balsam says he has been storing and tracking spam since the spring of 2002. Rather than limiting the project to filters and blocklists, he pursued senders and advertisers through small-claims, superior and federal court cases, particularly under California anti-spam law.

The site says he accumulated more than fifty court judgments against spammers, along with settlements and additional cases. It also preserves media coverage stretching back through the 2000s and early 2010s, when spam was one of the internet’s loudest everyday irritations and lawmakers were still arguing over how to regulate it.

A personal grievance became a web archive

The structure is wonderfully specific.

There are pages for small-claims cases, federal cases, spam law, research tools and press coverage. Old links point to WHOIS services, secretary-of-state databases, IP lookups and other tools used to identify the entities behind unwanted mail.

This is not a generic “spam is bad” essay. It is a record of one person’s method for turning unwanted messages into names, companies, complaints and court filings.

The site is also a snapshot of a particular legal moment. Balsam criticizes the federal CAN-SPAM Act of 2003 from a consumer perspective and explains why he continued relying on California provisions dealing with falsity and deception. Anyone using the site for current legal guidance should verify the law independently; the valuable thing here is the preserved history of how an individual internet user fought the problem at the time.

There is even a deliberately aggressive no-spam policy claiming a $25,000 fee for reading and responding to unsolicited commercial email sent to addresses on his domain. Whether one loves the tactic or rolls their eyes at it, nobody can accuse the site of hiding its position.

CacheRat’s 1,967 Ancient Web Domains research list keeps uncovering pages where old internet problems are documented by the people who were actually fighting them. This is a good example.

Spam history is usually told through legislation, botnets and statistics.

DanHatesSpam.com preserves the smaller scale: one inbox, one angry recipient, a stack of evidence, and eventually a surprising number of court cases.

Read Dan Hates Spam

Posted on by

Ancient Web: StonePages Put Europe’s Megaliths on the Web in 1995

StonePages is what happens when two people start putting photographs of ancient stones on the Web in 1995 and then simply keep going. Long before location apps, social photo feeds, and polished heritage portals, Diego Meozzi and Paola Arosio were building an online guide to European megalithic sites one visit at a time.

Visit StonePages

The site’s own history is unusually well documented. It says the project began in 1995 and was originally intended as a guide to the major megalithic monuments of Europe. Meozzi and Arosio did not just compile place names from books. They traveled, photographed sites themselves, wrote descriptions, and distinguished firsthand material from information gathered from other sources.

That matters. StonePages is not merely an old directory whose links survived. It contains fieldwork.

By 2001 the project reported more than 5,000 photographs and information on over 500 archaeological sites. The subject matter ranges across stone circles, standing stones, dolmens, chambered tombs, alignments, and other prehistoric monuments. Famous places sit alongside structures that are much easier to miss unless somebody has already done the legwork.

The website also preserved its own technical evolution. It was renamed Stone Pages in 1998 and went through several redesigns as the Web changed. Its data moved from a flat-file system into an SQL-backed database. That little detail makes the site useful to web historians as well as archaeology readers: this was a serious information project adapting to the Web while both the tools and expectations were still being invented.

There is also Stone Pages Archaeo News, which expanded the project beyond a photographic field guide into an archaeology-news resource. The result is less like a single homepage and more like a small independent publishing operation that happened to grow out of two people visiting old stones.

For a modern webmaster, StonePages is a reminder that useful sites can begin with a brutally simple advantage: the creator has gone somewhere, seen something, measured something, photographed something, or otherwise done work that cannot be generated by rearranging everybody else’s summaries.

That is also why the site still matters in 2026. Search results can tell you that a monument exists. StonePages preserves the accumulated decisions of people who spent years deciding which monuments were worth finding, how to describe them, and what they actually looked like on the ground.

The larger 1,967 Ancient Web Domains research list contains many projects built on the same old-Web principle: somebody cared enough about one subject to make themselves useful.

Explore StonePages and its megalithic site guides

Nearly three decades after its launch, StonePages is still a good argument for fieldwork plus HTML.