404: Memorial — The Internet Archaeologists Digging Up Dead Websites Before They're Gone for Good

There's a website out there — or rather, there was — that sold handmade ceramic frog figurines out of a garage in Tulsa, Oklahoma. It had a teal background, a hit counter frozen at 4,872, and a MIDI rendition of "Für Elise" that started playing the second you loaded the page. It was somebody's whole thing. Their little corner of the internet, built with love and probably a pirated copy of FrontPage 98. And now it's gone.

Not archived. Not memorialized. Just gone.

That kind of quiet digital death is what keeps people like Marcus Delray up at night. Marcus runs a personal blog out of Columbus, Ohio, where he documents dead websites the way some people document abandoned malls or shuttered diners. He calls it "web archaeology," and he's not alone.

The Impulse to Preserve

The old internet was a mess. Frames within frames. Animated GIFs of dancing babies. Guest books full of spam and genuine heartfelt messages living side by side. It was chaotic and personal and deeply, almost embarrassingly human. And because nobody thought of it as culture worth saving, most of it simply evaporated.

That realization hit a certain generation of Americans hard around the mid-2010s, when platforms like GeoCities had already been bulldozed and the web had started consolidating around a handful of major apps. People began to mourn something they hadn't even realized was precious while they had it.

"I grew up visiting these random little websites," Marcus says. "Fan pages for TV shows. A guy who just cataloged every type of cloud he'd ever photographed. A teenager's poetry page that probably got like six visitors total. That was the internet I fell in love with. And I didn't save any of it."

That regret became motivation. Marcus now spends hours every week crawling through old forums, following dead links, and using tools like the Wayback Machine to piece together what sites used to look like. When he finds something that hasn't been captured yet, he screenshots everything he can and uploads it to his own archive.

The Wayback Machine and Its Limits

The Internet Archive's Wayback Machine is the closest thing the web has to an official memory. Founded in 1996 by Brewster Kahle, it has crawled and stored billions of pages over the decades, and for a lot of preservationists, it's the first stop when chasing a dead URL.

But it has real gaps. Not every page was crawled. Not every crawl captured images or working stylesheets. And some sites actively blocked archiving bots, meaning entire communities, storefronts, and creative projects exist now only in the memories of people who happened to visit them.

"The Wayback Machine is incredible, but it's not a guarantee," says Priya Nambiar, a librarian in Portland who volunteers with a small preservation collective. "A lot of small, personal websites just slipped through. And once the domain expires and the hosting lapses, that's it. There's no recovery."

Priya's group focuses specifically on early 2000s fan communities — message boards, shrine pages, and fanfiction archives that existed before platforms like Tumblr or Archive of Our Own centralized that culture. They use a combination of automated scraping tools and good old-fashioned manual clicking to capture as much as they can before links go cold.

More Than Nostalgia

It would be easy to write all of this off as pure nostalgia — grown adults pining for the internet of their childhoods. And sure, there's definitely some of that going on. But the preservationists themselves push back on that framing pretty hard.

What they're really doing, they argue, is saving primary sources. The web of the 1990s and early 2000s is a direct record of how ordinary Americans communicated, expressed themselves, built communities, and processed the world around them — often without any institutional filter. It's the kind of raw cultural document that historians usually only dream about.

"Imagine if we had this level of personal documentation from the 1950s," Priya says. "Diaries, letters, home movies — those are treasures. The personal web was the same thing, just digital. And we're letting it rot."

Marcus makes a similar point when talking about the ceramic frog website. "That woman — I assume it was a woman, based on the vibe — she put real time and care into that page. It was her creative output. It deserves to exist somewhere."

The Tools of the Trade

For anyone curious about joining the effort, the toolkit is more accessible than you might think. The Wayback Machine lets you submit URLs for archiving directly through its "Save Page Now" feature, which is genuinely free and takes about ten seconds. Tools like HTTrack let you download entire websites locally. Browser extensions like SingleFile can capture a full rendered page as a single HTML file, images and all.

For more advanced work, preservationists use tools like Wget with specific flags to crawl entire domains, though that gets complicated fast when sites have dynamic content or require logins.

There are also organized community efforts worth plugging into. Groups on Reddit, Discord, and even some old-school mailing lists coordinate around specific preservation targets — particular hosting platforms that are shutting down, fan communities that are migrating, or genres of sites that seem especially at risk.

What Gets Lost When We Don't Look

Here's the thing that haunts this community more than anything else: they know they're losing the race. Every day, domains expire. Hosting plans lapse. Servers get wiped. And unlike a physical building that crumbles slowly enough for someone to photograph it on the way down, a website can vanish in an instant — one missed renewal payment and it's just gone, replaced by a domain squatter's blank page or a parking lot of ads.

Some of what's disappearing is trivial, sure. But some of it isn't. Early LGBTQ+ community spaces that predate mainstream social media. Regional music scenes documented only on now-dead MySpace pages. Small business histories. Local journalism archives. Recipes. Rants. Art.

All of it lived on the web, and all of it is at risk.

"People think the internet is permanent because it's digital," Marcus says. "But digital doesn't mean permanent. It just means it can disappear faster and more completely than almost anything else."

The Last Video Store Feeling

There's a reason the last Blockbuster in Bend, Oregon became a cultural landmark — it represented something people didn't know they'd miss until it was almost gone. The web preservationist community is chasing a version of that feeling, but at a scale that's almost impossible to comprehend.

The early web wasn't just a delivery mechanism for content. It was a place. It had texture and personality and weirdness that's genuinely hard to replicate in the algorithmic, monetized internet of today. And the people working to archive it aren't just being sentimental — they're trying to make sure that when someone fifty years from now wants to understand what American culture looked like in 1998 or 2003, there's actually something left to look at.

The ceramic frog lady's website is gone. But maybe the next one doesn't have to be.