June 30, 2026
404: Our memories not found
Gone but Not Forgotten: Recovering the Dead Web
The internet is vanishing, and commenters say our digital memory is hanging by a thread
TLDR: A major study found huge parts of the web disappear over time, while the Wayback Machine saves only some of what’s lost. Commenters turned that into a bigger warning: if archives fail, history itself can be edited, erased, or quietly vanish.
The big number making people clutch their bookmarks: 38% of webpages from 2013 are gone, according to Pew, and roughly a quarter of pages from the last decade have simply disappeared. Enter the Wayback Machine, the internet’s unofficial attic, which the article says has managed to save about 15% of pages that would otherwise be lost. In plain English: a huge chunk of the web is evaporating, and a nonprofit archive is one of the only things stopping our online past from turning into dust.
But the real heat is in the reaction. The loudest mood from the community is part gratitude, part panic. One commenter flat-out called Archive.org an “international treasure,” then swerved into full political alarm, warning that powerful interests and even governments can wipe out publicly funded knowledge with shocking ease. That’s the drama bomb here: this isn’t just about broken links or old blog posts disappearing. People are reading it as a story about who gets to control memory itself.
And yes, there’s a dark joke running underneath it all: the web was supposed to remember everything, but commenters are basically saying it has the lifespan of a mayfly unless someone actively saves it. The vibe is equal parts museum fundraiser, digital doomsday, and “maybe the weird old internet forums were more valuable than we thought.” In other words, the community isn’t treating this like boring data loss—they’re treating it like a slow-motion cultural heist.
Key Points
- •The article says a 2024 Pew Research Center study found 38% of webpages from 2013 were inaccessible a decade later, and about 25% of webpages sampled across 2013–2023 were no longer accessible.
- •It compares Pew’s findings with other link-rot studies from Ahrefs, Jonathan Zittrain/The Atlantic, and Old Dominion University, all showing substantial web content loss over time.
- •The article argues that some studies overlook web archives and therefore do not fully account for content that can still be recovered after disappearing from the live web.
- •Its analysis focuses on the Wayback Machine and states that it has rescued roughly 15% of pages that would otherwise be considered dead.
- •The authors used Pew’s shared dataset of 5.4 million URLs, transformed from Parquet files, to check whether dead or live pages had archived copies in the Wayback Machine.