Tracing the same complaint across 30 years of forums, feeds, and false golden ages
There is a comment you can find in almost any online space, in almost any year since 1993: it wasn't like this before. Before the newbies. Before the ads. Before the algorithm. Before "everyone" showed up and ruined it.
The remarkable thing isn't that people say this. It's that they've been saying it, in nearly identical language, for three decades — about completely different internets.
The pattern has a name, and it's older than you think
In September 1993, AOL gave its subscribers access to Usenet, the sprawling network of discussion groups that had, until then, been populated mostly by university students and researchers. Every autumn before that, a new class of undergrads had arrived, asked the same beginner questions, and been absorbed within a few weeks as they learned the norms. Old-timers called it "September."
AOL's rollout broke the pattern. Now newcomers arrived constantly, in volumes the old culture couldn't absorb or educate fast enough. Longtime users started calling it the Eternal September — the month that never ended, the moment the internet stopped belonging to the people who'd built its culture.
This is worth sitting with, because it means the founding complaint about the internet getting worse wasn't about smartphones, or social media, or algorithms. It was about too many people arriving too fast for a culture to metabolize them. Everything that came after is a variation on that same anxiety.
Every decade gets its own "before"
The 1990s golden age, as remembered by the people who were there, was Usenet and early web forums — text-heavy, slow, populated by a self-selecting group of the technically curious. Getting online required enough effort that it filtered out casual users. The nostalgia here is really nostalgia for exclusivity, though it's rarely described that way.
The early-2000s golden age was blogs and forums — LiveJournal, phpBB communities, the first wave of Wikipedia. People who came up in this era remember a web that felt handmade, where a "community" was a few hundred people who all knew each other's usernames. The complaint that killed this era, in real time, was the arrival of easy publishing tools and broadband: suddenly everyone could have a blog, and the people who'd been writing since 2001 felt outnumbered.
The Web 2.0 golden age was the social web before it consolidated — MySpace profiles you coded yourself, the early feed of Facebook when it was chronological and only for people you actually knew, the first few years of Twitter when it felt like a clubhouse of writers and journalists. This era's death, as its mourners tell it, came from algorithmic feeds and the arrival of brands, influencers, and eventually bots.
The mid-2010s golden age — the one a lot of people in their twenties and thirties are nostalgic for right now — was peak Tumblr, early Instagram, Vine, the specific texture of a web that hadn't yet been fully optimized for engagement metrics. Its death is blamed on the same mechanism as the one before it: algorithmic ranking, ad load, and platforms tuning themselves for time-on-site rather than user satisfaction.
Notice the shape. Every single "golden age" is remembered as smaller, slower, and more sincere than what replaced it. And every single ending is blamed on the same two forces: more people arriving, and platforms optimizing for something other than the users' experience.
Three things are actually happening at once — and they get conflated
The reason this complaint feels true every single time, to every single generation, is that it's actually three distinct phenomena bundled into one feeling.
First, there's a real economic and design shift. Ad-supported platforms do change over time, usually for the worse from the user's perspective. The writer Cory Doctorow coined the term "enshittification" for this specific pattern: a platform starts by being good to users to attract them, then shifts value toward business customers (advertisers, sellers) once users are locked in, then finally extracts value for shareholders at the expense of everyone else. This is a real, documented business dynamic — not nostalgia. Early Amazon really did subsidize prices to build a user base before raising them. Early Facebook really did have a chronological feed before algorithmic ranking maximized ad exposure. This part of the complaint is often correct.
Second, there's a scale effect that has nothing to do with corporate greed. Any space that grows from a thousand people to a hundred million people will feel different, even with zero changes to its business model, because culture and norms are load-bearing structures that only work at certain sizes. A dinner party and a stadium concert are different experiences not because the stadium is "worse," but because the mechanics of a thousand-person crowd are categorically different from an eight-person conversation. Early internet communities felt intimate because they were, structurally, small. That intimacy was never going to survive success — not because of any villain, but because of arithmetic.
Third — and this is the one people like least to hear — there's you. The person who joined a platform at 19 and left at 29 experienced the platform's actual evolution, but they also experienced their own evolution: from someone with unlimited free time and low stakes to someone with a job, relationships, and a shrinking tolerance for nonsense. Some of what feels like "the internet got worse" is "I got less available to be delighted by novelty." Psychologists have long noted that the intensity of nostalgia tends to peak for the cultural artifacts of adolescence and early adulthood — the phenomenon shows up for music, food, and slang, not just platforms. The internet you fell in love with probably corresponded with a specific window in your own life that was never coming back regardless of what any platform did.
The Eternal September never actually ended
Here's the uncomfortable twist: the Eternal September framing implies a one-time event — a before and after. But the actual pattern is that it happens again, to a smaller population, every single time a space grows. Usenet old-timers felt it about AOL users. Early bloggers felt it about MySpace teenagers. Early Twitter felt it about the "normies" who arrived after Ellen's celebrity selfie. TikTok's earliest adopters felt it the moment it stopped being a lip-sync app for teenagers and became the default video platform for everyone, including their parents.
Each of these groups was right that their experience of the platform changed. Where they tend to be wrong is in assuming this was a unique historical rupture rather than the same structural event that has happened, on schedule, to every online space that ever grew.
What this means if you're feeling it right now
If you're someone who thinks "the internet used to be better" — and if you're old enough to be reading a Medium essay, you probably are — the honest version of that feeling has three separate parts worth untangling:
Some of what you're mourning is a real, documented shift in incentives as platforms matured from growth mode to extraction mode. That part is a legitimate critique, and it's worth directing at regulation, competition, and interoperability rather than nostalgia alone.
Some of what you're mourning is simply what happens to any small culture that succeeds. It's not corruptible or fixable; it's the cost of the thing you loved being good enough that other people wanted it too.
And some of what you're mourning isn't the internet at all. It's a version of yourself who had the time, the attention span, and the low stakes required to fall in love with a website. That version of you is not coming back, no matter which platform you open.
The internet isn't in perpetual decline. It's in perpetual turnover — a machine that manufactures golden ages and then, with total reliability, produces the generation that mourns their passing. The only unusual thing about your version of this complaint is that you think it's the first time it's happened.
Top comments (0)