The Cost of Explaining

There's a specific kind of stuck that only shows up when you build alone. You're the person who most needs a second opinion, because nobody is checking your reasoning as you go. You're also the only person who has enough context to give that opinion any weight. Six months of small decisions live in your head and nowhere else. Whoever you'd ask has to somehow get all of it before their input means anything.

That's the part people miss when they tell solo builders to just ask more. It's not that asking feels beneath you. It's that asking has a real cost, and the cost is proportional to how deep the system goes. A quick question about a library function is cheap. A question about why the whole thing is shaped the way it is requires you to reconstruct months of tradeoffs for someone who wasn't there for any of them. Somewhere in the middle of that reconstruction you usually notice the fix yourself. By the time the explanation is good enough for someone else to use, you've often already used it.

On a team this gets absorbed without anyone noticing. Someone touched that file last sprint. Someone sat in the meeting where the decision got made. There's always a person who's already halfway inside your problem, so bringing them the rest of the way is a short walk. Solo, that shortcut doesn't exist. Every time you want a second brain, you're building one from nothing, in real time, usually at the exact moment you have the least patience for it.

A different kind of favor

What surprised me is how differently it feels to ask a friend or someone from an online community versus a colleague. A colleague owes you attention in some diffuse, structural way; that's part of what the working relationship is for. A friend or a person from a forum owes you nothing. Their evening, their weekend, their patience while you fumble through an explanation of a system they'll never touch again, all of that is pure goodwill. There's no invoice for it, no shared incentive to point to. It sits closer to asking someone to help you move furniture than to asking a teammate for a code review, and I've noticed I ration that kind of favor much harder than I ration my own time.

Some of that caution has softened. I've gotten better at asking for narrow, well-bounded things: a specific error message, one design choice with two options and a clear question, something that doesn't require anyone to load my whole project into their head. Those questions travel well. People can answer them in five minutes without owing me an afternoon, and I've stopped feeling like each one is a withdrawal from some finite account of other people's goodwill.

What I still push through alone is anything with fuzzy tradeoffs baked into the constraints, the kind of decision where explaining the situation honestly would take as long as just deciding. I tell myself it's because nobody else has enough context to help, and that's often true. But some of it is just habit. When you've spent years being the only person who understands the whole system, defaulting to solving it yourself stops feeling like a choice and starts feeling like the shape of the job.

I don't think that's fully fixable when you're building alone. The goal I'm working toward isn't asking for help more often across the board. It's getting sharper about which questions are actually worth the cost of explaining, so the ones that matter don't quietly stay unasked.