Build With People, Not In Isolation

A lot of us want to help by building good, useful things. But retreating to make the elegant, principled "right thing" in isolation is a trap. Software matters because it gets used, and the surest way to get it used is to build it with the people who need it.

Things feel pretty intense right now, and a lot of us respond to that by wanting to help. We want to build good software, useful tools, things that matter. And when everything feels chaotic, there's a strong pull to retreat into making the right thing: something elegant, principled, technically solid, built exactly the way you think it should be.

I love that instinct. I also think it's a trap.

Because if you build in isolation, you're guessing. It doesn't matter how well-intentioned or well-engineered the thing is. Guessing is a poor substitute for working directly with the people who actually need the help.

The part that stuck with me

A while back I was watching an LTT video with Linus Torvalds on it, where he's watching the lesser Linus build him a computer. At one point he said something that stuck.

He said he doesn't really care how popular Linux is. He's not chasing adoption numbers or mindshare. But then he added the part that mattered: users matter.

Not because popularity is the goal. Because without users, the work stops making sense. If nobody were running Linux, there'd be no reason to keep building it. It wouldn't matter how correct or elegant or well-designed it was. It would just be time spent in a vacuum.

That framing feels especially relevant right now.

Perfect isn't the same as useful

It's tempting to believe that if you just build the best version of a thing, the cleanest architecture, the most principled approach, people will eventually find it and it'll all have been worth it.

Sometimes that happens. Mostly it doesn't.

A perfect tool that nobody uses doesn't help anyone. And it's worse than neutral, because it drains your energy while convincing you you're making progress. You can spend months polishing something that was never going to connect with anyone, and feel productive the entire time.

Value shows up when the work meets real people with real needs. Not before.

Community comes first, even though it's the least glamorous part

The most reliable way I've found to build something useful is also the least glamorous. You talk to people. You listen more than you explain. You work alongside them, and you build small things together instead of one big correct thing in the dark.

When you do that, the work changes underneath you. Your assumptions get challenged early, while they're still cheap to change. Features you were sure were essential fall away. Problems you never would have noticed become obvious the second someone hits one in front of you.

That's not the work going wrong. That's the work getting grounded.

Collaboration reshapes the product

When you ship gradually, in the open, with real users, the thing you end up with is almost never the thing you first imagined. It's usually smaller. More opinionated. More human. It carries the marks of real tradeoffs and constraints and lived experience instead of just your technical ideals.

That drift isn't a failure of vision. It's the clearest sign the work is connected to something outside your own head.

Build where the need already is

If you want to help, start by finding people who are already stuck on a problem. Work with them directly. Let what they need pull the direction of what you build, instead of deciding the direction up front and hoping they show up for it.

Software doesn't matter because it's clever. It matters because it's used. And the surest way to make sure something gets used is to build it with the people who need it, instead of handing it to them fully formed from a distance.

That's how the work stays meaningful. It's also the part I have to keep relearning.

This remind you of a problem you've been dealing with?

I'd be happy to discuss it with you and see if it's something we could turn into a solution.