Boring Technology Is Sometimes the Brave Choice

Every new tool looks cheap on the day you install it. It's fast, it's clever, the demo works on the first try, and for an afternoon you feel like you're building the future. The bill comes later — usually months on, when the excitement is gone and something important is broken at the worst possible time.

I've written before about choosing boring tools to protect my energy on low-capacity days. That's the personal case. There's a broader one that has nothing to do with me and everything to do with the software outliving the day it was written.

Novelty is a loan

Reaching for the newest framework, the clever database, the abstraction three people on the internet understand — every one of those is borrowed against the future. You get speed now. You pay it back with interest, and the interest is charged in things you can't spare when the bill arrives.

Ask what a tool costs across its whole life, not just at install:

When it breaks, is the answer findable? A mainstream tool has years of answered questions, real documentation, and a fix sitting on the first page of search results. An exotic one has a chat server and a shrug.
Can someone else pick it up? Every choice is a future handover — to a teammate, a contractor, or yourself in a year with no memory of it. Boring tools come with a shared vocabulary and people who already know them. Clever ones have a bus factor of one.
How stable is the ground under it? A dependency with a large community and a slow, careful release cadence is ground you can build on. A fast-moving package with three maintainers is a renovation you'll be doing forever, whether you wanted to or not.
What does it cost to hold in your head? Familiar tools are nearly free to think about. Novel ones tax your attention every time you touch them, and that tax compounds across every future session.

None of these show up in the demo. All of them show up in the maintenance.

What I actually build on

My site and my product run on deliberately unremarkable choices: React and TypeScript on the front end, Supabase for data and auth, Vercel for deploys, Kotlin for the Android app. None of it is going to impress anyone at a conference. That's the point. It's the documented, default path — the one where, when something fails, the answer is already written down by someone who hit it before me.

I didn't pick those because I couldn't learn something newer. I picked them because I want to spend my limited, hard-won capacity on the problem I'm actually trying to solve, not on being the one person who can debug a tool nobody else uses.

The honest counterargument

Boring can rot into an excuse. "We've always used this" is how teams end up stranded on dead technology, refusing to learn anything new, mistaking stagnation for discipline. That's not what I'm defending.

New is worth the risk when the payoff is real and specific: the tool solves a problem your boring stack genuinely can't, or solves it so much better that the difference changes what you're able to build. It's worth it when you have the capacity to absorb the learning curve without starving the actual work. And it's worth it when you can contain the bet — try the risky thing in one corner where a failure is survivable, not wired through the foundation where ripping it out means rewriting everything.

The test isn't old versus new. It's whether you're choosing the tool for what it does, or for how it makes you feel while you install it. Excitement is not a technical requirement.

The brave part

Choosing boring looks like a lack of ambition from the outside. It's usually the opposite. It means you've thought past launch day to the long, unglamorous middle where the software actually lives — the part where it has to keep working while you're tired, or busy, or gone, and someone who isn't you has to keep it running.

Anyone can pick the exciting tool. Picking the one that will still make sense in two years, when the excitement is somebody else's problem, takes more nerve. That's the choice I keep coming back to.

Novelty is a loan

Ask what a tool costs across its whole life, not just at install:

When it breaks, is the answer findable? A mainstream tool has years of answered questions, real documentation, and a fix sitting on the first page of search results. An exotic one has a chat server and a shrug.

Can someone else pick it up? Every choice is a future handover — to a teammate, a contractor, or yourself in a year with no memory of it. Boring tools come with a shared vocabulary and people who already know them. Clever ones have a bus factor of one.

How stable is the ground under it? A dependency with a large community and a slow, careful release cadence is ground you can build on. A fast-moving package with three maintainers is a renovation you'll be doing forever, whether you wanted to or not.

What does it cost to hold in your head? Familiar tools are nearly free to think about. Novel ones tax your attention every time you touch them, and that tax compounds across every future session.