One Week Browsing the Internet Like It Was 2005. I Did Not Come Out Okay.

The Setup

I want to be upfront about something: I went into this experiment with a romanticized idea of what the early internet was. I remembered it as this scrappy, human-scaled place where you'd fall down a Wikipedia rabbit hole at 1am and end up reading about the history of competitive eating or the taxonomy of obscure medieval weaponry. Serendipitous, weird, alive.

What I conveniently forgot was the loading times. The dead links. The forums where someone would answer a question from 2004 and then the thread just... stopped. The internet of 2005 was not a golden age of digital civilization. It was a teenager's bedroom — creative and full of personality and also kind of a mess.

But I tried it anyway. For one week, I committed to the following rules: no algorithmic feeds (so no Twitter/X timeline, no TikTok, no Instagram Explore, no YouTube homepage recommendations). No infinite scroll platforms at all, if I could help it. Navigation had to be intentional — I'd use bookmarks, RSS readers, and actual search queries to find things rather than letting an algorithm surface them. And I'd spend time in forums, on independent websites, and following links the old-fashioned way: because something looked interesting and I clicked it.

Let's talk about what happened.

Day One: The RSS Rabbit Hole

I started by setting up an RSS reader — Feedly, which is free and relatively painless. For anyone who wasn't online in the mid-2000s, RSS (Really Simple Syndication) is basically a way to subscribe to website updates without needing the website to have an app or a social media presence. You get a feed of new posts from sites you've manually chosen. It's aggressively unsexy and I cannot overstate how much I loved it by the end of the week.

The process of choosing which sites to add was itself revelatory. I had to actually think about what I wanted to read. Not what an algorithm thought I'd engage with, not what was trending, not what twelve people I follow had already shared — just: what topics do I care about enough to actively seek out? Food writing. Long-form journalism. A couple of tech blogs that still do original reporting. A film criticism site I'd bookmarked years ago and never checked consistently.

Day one felt productive in a way that the regular internet rarely does. I read four full articles. Not skimmed — read. Probably because there was no competing notification pulling my attention sideways every ninety seconds.

Day Three: Forums Are a Different Planet

I spent a chunk of Wednesday on Reddit — but deliberately, the way people used it before the algorithm turned the front page into a highlight reel. I navigated directly to specific subreddits for topics I was curious about rather than opening the app and seeing whatever the feed decided I needed. I also spent time on some older-style forums that still exist if you go looking: hobbyist communities, niche interest boards, places that look like they haven't been redesigned since the Obama administration.

Here's what's strange about forums when you come back to them after years of social media: they're slow. Not in a bad way, necessarily — more like the difference between a conversation and a broadcast. People write longer posts. They follow threads over days. There's institutional memory; longtime members will reference discussions from years ago. It feels less like performing for an audience and more like talking with people who are genuinely interested in the same thing you are.

I got sucked into a thread about cast iron cookware restoration that had been running since 2019. I learned more about carbon seasoning than I will ever practically need. It was great.

Day Five: The Part Where It Got Hard

By Thursday, the cracks were showing. Not because the experiment wasn't working — it was, in a lot of ways — but because the 2005 internet and the 2024 internet are not actually interchangeable, and pretending otherwise has real costs.

The biggest one: news. The way news travels now is through social feeds. That's just true. When I was navigating by RSS and direct site visits, I was getting coverage of things but missing the real-time texture of how a story develops — the updates, the corrections, the context that accumulates through a day of people reporting and responding. I felt informed but slightly out of phase, like I was reading yesterday's newspaper while everyone else was watching the live broadcast.

The second problem: discovery. The early internet's version of serendipity required a lot more work than I remembered. You had to actually go looking for new things. Clicking through blogrolls, following links, stumbling into corners of the web that weren't optimized for you specifically — it's genuinely delightful when it works. But it's also intermittent. The algorithm, for all its manipulation and filter-bubble problems, is extremely good at surfacing things you didn't know you wanted. I missed that. Not constantly, but I missed it.

What the Old Internet Got Right

Here's what I came away genuinely believing: the early internet was better at depth and worse at breadth. You knew your corners of it well. You had regulars, you had voices you trusted, you had a sense of place that's almost impossible to replicate on a platform designed to keep you scrolling through everything simultaneously.

The human-made quality of it — sites that reflected actual personalities, writing that wasn't optimized for engagement metrics, communities that formed around genuine shared interest rather than algorithmic clustering — that part was real and it's worth mourning a little.

But the nostalgia is doing some heavy lifting here. We forget the gatekeeping, the digital divides, the sheer amount of bad information that circulated freely because there was no infrastructure to push back on it. The internet of 2005 wasn't a utopia. It was just younger, and youth always looks better in hindsight.

What I'm Actually Taking Back With Me

I kept the RSS reader. That's the one concrete behavioral change that survived the experiment. I check it in the morning the way I used to check Twitter, and it's almost uniformly a better experience — slower, more considered, less likely to leave me feeling like I've been mildly manipulated.

I also came back to social media with a slightly different relationship to the scroll. Not a fixed one, not a healed one — but more aware of what I'm doing when I open an app and let it show me things. The 2005 internet reminded me that curation used to be something you did yourself, and there's a version of that still available if you're willing to do a little more work.

Was the old internet better? Not exactly. But it was different in ways that are worth borrowing from — even if you're not willing to go full RSS-reader, forum-dwelling, bookmarks-toolbar about it. Which, fair enough. Neither am I, completely.

But I'm closer than I was a week ago.