You remember the old web. Admit it.
You remember sites that looked like someone actually built them — not generated them. Tables stacked inside tables, floats doing things they were never designed to do, absolute positioning holding everything together with digital duct tape. It was messy. It was hacky. And it had something almost no website built in 2024 has: a pulse.
Then Flexbox arrived. Then Grid. We celebrated. We said finally, we can lay things out properly. No more hacks. No more clearfix. No more three-column layouts that collapsed into chaos the moment you resized the browser.
We got exactly what we wanted. And the web died a little.
The tools that gave us total freedom also gave us permission to stop thinking.
Think about it. Before Flexbox, if you wanted a sidebar next to your main content, you had to fight for it. You had to understand the box model. You had to calculate widths. You had to make real decisions about how things related to each other. Every layout was an act of problem-solving, and problem-solving leaves fingerprints. Fingerprints are personality.
Now? You type display: flex. You add gap: 16px. You move on. The browser handles the rest. And so does everyone else — because they’re typing the exact same thing.
I’m not romanticizing broken layouts. Floats were a nightmare. Vertical centering was a war crime. But here’s the uncomfortable truth nobody wants to say out loud:
The constraint wasn’t the problem. The constraint was the personality.
When you have to work around limitations, you make choices. Choices are visible. They’re the difference between a site that feels like it was made by a person and one that feels like it was extruded from a content management system.
Look at the web today. Every SaaS landing page is the same hero section, the same three-column feature grid, the same testimonial carousel, the same pricing table with the middle plan highlighted. Every developer portfolio is the same dark mode, the same monospace font, the same terminal aesthetic. Every documentation site is the same sidebar-plus-content layout that looks like it came from the same template — because it probably did.
We didn’t lose the ability to design. We lost the reason to.
Here’s what actually happened: Flexbox and Grid didn’t just solve a technical problem. They solved a cultural one — the problem of having to decide. When layout was hard, you had to think about every element’s relationship to every other element. You had to have an opinion. Now the default is good enough. And good enough is the enemy of memorable.
Efficiency optimized away the friction that made the web feel handmade — and friction was the last thing giving it character.
Someone in the comments on that original article said it perfectly: we got our wish to lay out stuff however we want, and it turns out our imagination is indistinguishable from the average LLM output. All beige, all gray, all devoid of soul.
That’s not a coincidence. It’s a direct line. When the path of least resistance becomes a 12-column grid and a utility-first CSS framework, the path of least resistance becomes the entire internet. The bottleneck didn’t disappear — it shifted. It used to be technical: you couldn’t build it. Now it’s cultural: you can build anything, so you build what everyone else builds.
I’m not saying go back to tables. I’m not saying floats were better. I’m saying we need to reckon with what we traded away.
Because the early web had something we can’t recreate with better tools. It had people making weird, specific, sometimes ugly choices because they had to. A GeoCities page with animated GIFs and a visitor counter and a background MIDI file was objectively terrible by today’s standards. But you knew a human made it. You could feel the fingerprints.
Can you say the same about the last ten websites you visited?
The web didn’t lose its soul because we ran out of creativity. It lost its soul because we stopped needing it.
So here’s the challenge, if you build things for the web: next time you reach for the same layout pattern you’ve used a hundred times, ask yourself whether you’re making a choice or following a groove. Ask whether the default is serving your content or whether your content has been quietly reshaped to fit the default.
The tools are not the enemy. The complacency is. Flexbox didn’t kill web design. We did — one justify-content: center at a time.
The best websites of the next decade won’t be the ones with the cleanest code. They’ll be the ones where someone decided that good enough wasn’t good enough. Where someone chose friction over efficiency. Where someone left fingerprints on purpose.
Because in a world where every site can look the same, the ones that don’t will be impossible to ignore.
FAQ
Q: Aren't you just romanticizing a worse era of web development?
A: Partially, yes. Floats were genuinely awful and nobody misses clearfix hacks. But the point isn't that old tools were better — it's that the friction they created forced decisions, and decisions create personality. You can have modern tools AND make interesting choices. Most people just don't.
Q: So should developers stop using Flexbox and Grid?
A: No. That's insane. Use them. The practical implication is about awareness: when layout is trivial, the creative burden shifts entirely to you. The tool won't save you from being boring. You have to actively choose to be interesting.
Q: Isn't consistency actually good for users?
A: Usability and sameness are not the same thing. Users benefit from predictable navigation and readable layouts. But the web has overshot — we've optimized for 'not confusing' so hard that we've arrived at 'not memorable.' A site can be intuitive AND have a soul. We just stopped trying to do both.