The Man Who Invented Infinite Scroll Wants It Gone

In 2006, a young interface designer named Aza Raskin was annoyed at Google. Search results came ten to a page, and getting to result eleven meant a full click, a full page load, a full wait, just to keep looking at more of the same list. It seemed like an obviously fixable inefficiency. His fix was simple: when the user nears the bottom of the results, quietly load more underneath, before they even ask. No click, no page load, no interruption. He built it, showed it around to companies including Twitter and Google, and it worked exactly as designed. Getting more content in front of a user got easier. That was the entire goal.

Nearly two decades later, that same person has testified in court, under oath, describing his own invention as a mistake - and has put a number on the damage that's hard to shake once you've heard it.

An Engineer's Problem With an Engineer's Solution

It's worth sitting with how narrow the original problem actually was. Raskin wasn't trying to build a persuasion technology or study human psychology. He was trying to remove a click. Pagination - the "next page" button - was, by any reasonable interface-design standard of 2006, a small tax on the user: an interruption with no informational value, existing purely because of how web pages were built. Removing pointless friction is close to the founding principle of the entire discipline of user experience design. By that standard, infinite scroll wasn't just acceptable - it was correct.

What Raskin has since said he didn't anticipate wasn't a design flaw in the mechanism. It's what happens once a frictionless mechanism gets attached to content that's engineered to be endlessly appealing, inside a business model that profits from every additional second of attention it captures. "What I was blind to," he's said, describing the years after infinite scroll spread across the web, "was that despite my good intentions, in technology, incentives eat intentions." The interface fix was good faith. The context it landed in wasn't neutral.

Why Nobody Flagged It at the Time

It's easy, with two decades of hindsight, to wonder how a change this consequential got built, shipped, and copied without anyone raising an alarm. The honest answer is that there was nothing to raise an alarm about, by the design standards of the period. Interface design in the mid-2000s was still fighting a fairly recent memory of clunky, multi-step, form-heavy web software - the entire cultural momentum of the field was pointed at removing steps, not adding them. A designer proposing to keep pagination on purpose, specifically to preserve a stopping point, would have sounded like they were arguing for worse software. The concept of a deliberately-preserved inefficiency being good for the user hadn't really entered the field's vocabulary yet. It would take another decade of infinite feeds, engagement metrics, and platforms explicitly built around attention capture before "maybe some friction is the feature" became a legible idea at all - which is part of why Raskin's regret reads less like an admission of an obvious error and more like a genuine, hard-won update to how an entire discipline thinks about its own defaults.

The Scale He Now Attaches to It

Raskin has publicly estimated that infinite scroll, at current usage levels, burns through roughly half a million human lifetimes every month - a back-of-envelope figure meant to convey scale, not a peer-reviewed statistic, but a deliberately large number from someone who understands the mechanism from the inside. He's described the pattern as "one of the first products designed to not simply help a user, but to deliberately keep them online for as long as possible" - a description that applies just as much to the feeds built on top of his original idea as to the idea itself.

In February 2026, Raskin testified as a witness in a New Mexico state trial against Meta, speaking directly to how a piece of interface design he built to solve a minor annoyance became, in his own account, a load-bearing part of an attention economy he no longer feels good about having helped construct. It's a rare thing: a named, identifiable person willing to stand up and say, on the record, that a piece of technology he personally invented did real harm at scale. Most critiques of attention-capturing design come from outside - researchers, journalists, former users. Raskin's carries a different weight because it comes from the inside, from someone with nothing to gain and a specific technical understanding of exactly how the mechanism he's describing actually functions.

Why "Remove the Friction" Isn't Always the Right Rule

There's a design maxim that's mostly correct: friction is the enemy of good software. Every extra click, every unnecessary wait, every confusing step is a chance for a user to give up, get frustrated, or go somewhere else. Two decades of interface design has largely been a project of finding and removing exactly this kind of friction, and for the vast majority of software, that project has made things better.

The blind spot infinite scroll exposed is that this maxim silently assumes every stopping point in a piece of software is accidental - a leftover inefficiency rather than a feature doing real work. Pagination on a search results page genuinely was an accident of how web pages get built; there was no reason a person should have to click to see result eleven. But the "end" of a page in a printed newspaper, the credits at the end of a TV episode, the back cover of a book - these aren't bugs to be optimized away. They're the moments that used to force a small, low-stakes decision: do I keep going, or was that enough? Infinite scroll's insight - remove the click - turned out to work exactly as well on those meaningful stopping points as it did on Google's pagination, because from a pure interface-design standpoint they're structurally identical. The system can't tell the difference between friction that's protecting you and friction that's just badly designed. Only the content on either side of it can.

What Actually Changed Once the Pattern Spread

Twitter adopted a version of it. Facebook's News Feed did. Instagram, TikTok, YouTube Shorts all built entire products around a variant of the same principle: never let the user arrive at a natural stopping point, because a stopping point is a moment where they might close the app. Each individual implementation is defensible on its own terms - nobody sat down and said "let's design something addictive" so much as each team, independently, noticed that removing the stop made the numbers go up, and kept it. Raskin's account of "incentives eat intentions" describes exactly this: no single villain, just a pattern that survives and spreads because it works, regardless of whether working is good for the person on the other end of the screen.

That spread also explains why blaming any single company for infinite scroll misses the actual shape of the problem. By the time short-form video platforms were being designed, infinite scroll wasn't a novel choice requiring justification - it was simply the default assumption of what a feed looks like, inherited whole from a decade of prior products that had already normalized it. Nobody at a short-form video company needed to independently rediscover Raskin's insight. It arrived pre-built, already proven, already the water everyone in the industry was swimming in.

The Uncomfortable Position of the Original Designer

There's a particular kind of discomfort in Raskin's position that's worth naming directly, because it's rarer than it might seem. Most public reckonings with harmful technology come from people who can distance themselves from the original decision - a journalist, a researcher, a regulator, someone whose job was never to build the thing being criticized. Raskin doesn't have that distance. Every time he describes the half-million-lifetimes figure, he's describing the downstream consequence of his own hands, his own code, his own reasonable-seeming decision made at a specific desk in 2006. That's a much harder position to speak from honestly than criticism from the outside, and it's part of why his account carries a kind of weight that a think-piece from someone who never built anything doesn't: he's not speculating about how the mechanism might work. He watched it work, from the inside, on something he made.

Putting a Version of the Old Stopping Point Back

None of this is really an argument for bringing back pagination generally - Google's ten-results-per-page click was a genuine annoyance, and Raskin was right to remove it there. The argument is narrower: for the specific feeds where the missing stop is doing the damage he's since described, restoring some version of it addresses the actual mechanism, not a symptom of it. That's the specific niche a handful of tools occupy now - Opal and One Sec apply that restoration at the whole-app level; Dam It applies it narrower still, putting a stop specifically back inside YouTube Shorts, Instagram Reels and TikTok on Android, in the exact spot Raskin's original design removed one twenty years ago.

The man who took the stop away is on record wishing someone would put it back. That's a strange kind of endorsement for the idea, but it's a real one.