Broken on Purpose: Why the Worst UX Online Is Secretly the Best Thing Happening

There's a Discord server out there — you won't find it through Google, and nobody's going to post the invite link on Twitter — where a few hundred people have been quietly building one of the most coherent bodies of music criticism you've never read. The channel structure makes no immediate sense. The pinned messages are three years old. The bots are half-broken. And somehow, that's exactly why it works.

We've spent the better part of a decade being told that the goal is frictionless. Seamless onboarding. Intuitive navigation. One-click everything. The big platforms have poured billions into making themselves easier to enter, easier to scroll, easier to never actually leave. And what did all that smoothness produce? An internet where the loudest stuff floats to the top and anything requiring more than thirty seconds of effort quietly disappears.

Maybe the friction was the point all along.

The Filter Nobody Designed

Here's the thing about a platform that's genuinely annoying to use: it self-selects. Hard. If you have to dig through a wiki with inconsistent formatting, learn a custom tagging system, or sit in a read-only channel for two weeks before you can post anything, the people who stick around are the ones who actually want to be there.

That sounds obvious when you say it out loud. But the implications are pretty significant. Most of what makes mainstream online spaces unbearable isn't the content itself — it's the ratio. The signal-to-noise problem is fundamentally a density problem. When a platform is easy to join and costs nothing to participate, you get everyone. And everyone, collectively, tends to drag conversations toward the least interesting version of themselves.

Niche forums built on software that hasn't been updated since 2011 don't have that problem. Neither do the IRC channels that still somehow have active regulars. Or the wikis where the community has developed such a specific internal vocabulary that a newcomer would need a glossary just to understand what's being argued about. These places are, by almost any UX standard, failures. They're also frequently the most intellectually alive corners of the internet.

Accessibility as a Double-Edged Thing

This isn't an argument against accessibility in the broader sense — not even close. There are real and serious reasons why inclusive design matters, and none of that is up for debate here. But there's a difference between designing something that excludes people based on ability, and designing something that excludes people based on commitment.

The second one is actually a feature. Commitment filters are how any community with standards maintains them. It's why certain subreddits with strict posting rules still produce better discussion than ones with millions of subscribers and no moderation philosophy. It's why the best fan wikis — the ones that go genuinely deep on a subject — tend to live on platforms that look like they were designed by someone who really loved databases and really hated drop shadows.

The mainstream internet optimized for reach. These places optimized for depth. And depth, it turns out, doesn't scale well.

Where the Actual Work Gets Done

If you pay attention to where creative and cultural movements actually originate before they surface into mainstream awareness, the pattern is pretty consistent. It's not on the platforms with the best analytics dashboards. It's in the spaces that are almost deliberately hard to find — where the interface communicates, in no uncertain terms, that this place wasn't built for you unless you're willing to figure it out.

Game modding communities have operated this way for decades. Certain corners of fan fiction culture. Experimental music circles. Speedrunning communities before they became a mainstream esports curiosity. Political organizing at the local level. Language learning communities that go way past what Duolingo can offer. In almost every case, the culture that eventually bleeds outward started somewhere that most people couldn't be bothered to navigate.

There's also something worth noting about the economics here. A platform with 300 highly invested users and no VC funding has zero pressure to grow, monetize, or pivot to video. The absence of a growth mandate is genuinely liberating. Nobody's optimizing the experience to keep you scrolling. Nobody's tweaking the algorithm to surface content that makes you feel bad but keeps you engaged. The platform just... sits there, doing its weird thing, for the people who want to do that weird thing.

The Paradox of Making Things Easier

Every time a niche platform gets "improved" — cleaner interface, mobile app, streamlined onboarding — something shifts. Sometimes it's subtle. Sometimes it's immediate and catastrophic. The community that built the thing watches a wave of new users arrive who have no context for the norms, no patience for the culture that developed, and no particular reason to develop either. The friction that kept things coherent is gone, and what replaces it is usually just... more internet.

This is the part that nobody in product development wants to hear: some of the things that look like bugs are load-bearing walls. The annoying registration process. The lack of a search function that actually works. The fact that you have to read fifty posts before you understand what anyone is talking about. These aren't failures of design — they're the mechanism by which a community stays itself.

Making something easier to use is, in a very real sense, making it easier to dilute.

What This Means for Anyone Actually Looking

If you're trying to find where the interesting stuff lives online — the conversations that haven't been focus-grouped, the communities that are building something without anyone watching — the smoothest platforms are probably the last place to look. Follow the friction. Find the things that require a little patience, a little learning, a little tolerance for interfaces that feel like they were built by someone who cared more about the content than the container.

It won't always be worth it. Some difficult platforms are just difficult, with nothing particularly valuable on the other side of the learning curve. But the ones that are — the ones where you put in the work and suddenly find yourself in a room full of people who have been thinking seriously about something for years — those are increasingly rare online.

And they're rare precisely because they never made it easy to get in.

That's not an accident. That's the whole thing.