Somewhere on the internet right now, a journalist is citing a source that no longer exists. A Wikipedia editor is quietly flagging a reference as "dead link." A Reddit thread from 2019 is full of URLs that resolve to homepages, parking pages, or the digital equivalent of a shrug — a 404 error dressed up in whatever the current brand palette happens to be.
This is link rot. It is not a bug. It is not an accident. It is the slow, grinding, entirely predictable consequence of building a civilization on top of infrastructure nobody agreed to maintain.
What We Talk About When We Talk About Hyperlinks
The hyperlink was supposed to be the revolutionary act. Tim Berners-Lee didn't just invent a way to display documents — he invented a way to connect them. The whole premise of the World Wide Web was that knowledge could reference other knowledge, that context could travel with content, that you could follow a thread from one idea to the next without losing the plot.
For about fifteen minutes in the mid-nineties, that vision held together pretty well.
Then the money arrived.
Once websites became businesses, the link stopped being a civic gesture and became a liability. Linking to a competitor was marketing suicide. Linking to old content was an SEO problem waiting to happen. Maintaining a consistent URL structure across platform migrations, CMS upgrades, and executive-mandated rebrands was somebody else's job — specifically, nobody's job. And so the links started dying.
Studies have put the half-life of a URL somewhere between two and seven years depending on the domain. Academic research is particularly brutal: one widely cited analysis found that roughly half of all URLs referenced in Supreme Court opinions no longer function as intended. The highest court in the land is footnoting the void.
The SEO Paradox That's Eating the Archive
Here's where it gets genuinely darkly funny. The same industry obsessed with "content strategy" and "domain authority" is one of the primary engines of link rot.
Every time a media company decides to migrate from one CMS to another — say, from WordPress to some proprietary platform that a private equity firm bought and will definitely not sell in eighteen months — thousands of URLs change. Sometimes redirects are set up. Often they aren't, because redirects require engineering hours and engineering hours cost money and the whole point of the migration was to cut costs. So the old URLs die, and every article on the internet that pointed to them now points at nothing.
Then there's the content pruning trend. SEO consultants discovered that deleting old, "low-performing" articles can improve a site's overall search ranking. The logic is defensible in a narrow technical sense. The consequence is that someone's carefully cited source — a news story, a product announcement, a government press release — gets vaporized because it wasn't pulling enough organic traffic in Q3. The article citing it now references a ghost.
And paywalls. Lord, the paywalls. A publication runs a story in 2011. The story gets cited across the web. The publication gets acquired, pivots to a subscription model, and locks its entire archive behind a $14.99/month gate. Every link to that story now lands on a login screen. The citation still looks functional. It just doesn't work anymore — not for the 99% of readers who aren't subscribers.
The Wayback Machine Is Not a Solution
The Internet Archive is a genuine treasure and Brewster Kahle deserves a monument. The Wayback Machine has preserved snapshots of millions of websites that would otherwise be completely unrecoverable. It is heroic, underfunded, and structurally inadequate to the scale of the problem.
The Archive doesn't catch everything. It doesn't capture content behind logins. It doesn't always index dynamically generated pages. It can't preserve the context of a link — the fact that this specific URL, on this specific date, said this specific thing that justified the citation. And it recently survived a major cyberattack that temporarily took it offline, which should terrify anyone who treats it as a reliable backstop.
Relying on the Wayback Machine to solve link rot is like relying on a single smoke detector to solve arson. Technically present. Definitionally insufficient.
We're Citing Quicksand and Calling It a Foundation
The downstream effects of this are hard to overstate. Academic papers lose their evidentiary trails. News articles lose the context that made their claims verifiable. Wikipedia — which has strict sourcing requirements — maintains a running tally of dead citations that runs into the millions. The web's self-referential architecture, the thing that made it feel like a living, interconnected body of knowledge, is quietly becoming a collection of isolated islands.
This matters in ways that go beyond inconvenience. The internet is where Americans increasingly get their news, conduct their research, make their medical decisions, and form their political opinions. When the sources underpinning that content evaporate, what's left is assertion without evidence. Opinion without reference. Confidence without foundation.
And because most readers don't click the links anyway — studies suggest somewhere between 5 and 15 percent of readers actually follow citations — the rot is largely invisible. The article looks authoritative. The footnote looks like it goes somewhere. Nobody checks. The edifice stands because everyone agrees not to poke it.
The Industry Shrug
What's the fix? Nobody's particularly rushing to find one.
Permalinks are a start — URL structures designed to remain stable across platform changes. Some academic publishers use DOIs (Digital Object Identifiers) that remain consistent even when content moves. A handful of journalism organizations have made explicit commitments to URL stability. These are good practices. They are also wildly underadopted.
The economics don't incentivize preservation. A media company gets no revenue from maintaining a 2009 article that three people a month visit to follow a citation. It does get a cleaner analytics dashboard if that article gets deleted. The math is obvious, and it's killing the archive.
What would actually help? Mandatory redirect maintenance as part of any domain sale. Institutional investment in archiving infrastructure. CMS platforms that treat URL stability as a first-class feature rather than an afterthought. A cultural shift in how publishers think about their back catalogs — as infrastructure rather than dead weight.
None of these are coming soon. So in the meantime, the links keep dying.
Read This While It Lasts
Here's the uncomfortable punchline: this article will eventually rot too. The links in whatever research informed it will degrade. The platform hosting it will migrate or rebrand or get acquired or simply go dark one day without ceremony. Some future reader will find a reference to it in another article and click through to nothing.
That's not nihilism. That's just the web being honest about what it's always been — a temporary agreement between servers to pretend permanence. We built our information civilization on hyperlinks, and hyperlinks are held together with the digital equivalent of good intentions and crossed fingers.
The internet isn't broken in some dramatic, fixable way. It's broken in the slow, mundane, thoroughly documented way that nobody bothers to fix because the brokenness is spread thin enough to be tolerable.
Until it isn't.