The Quiet Weeks

It's been a while since I last posted. And honestly, that's exactly what this entry is about.

The app was actually starting to feel good. Solo Quest was there, the duel mode was still around as a side dish, and we had built a small library of activities that players could come back to. It wasn't huge, but it was alive.

Achievements

So we kept layering. The next big addition was an achievement system — milestones, little rewards, those small dopamine hits that tell players "hey, you're making progress." Designing it took longer than I expected. Not the code itself, but the logic around it: what counts as an achievement? When does it trigger? How do we make sure it feels earned and not handed out? We went back and forth a lot.

In the end, what we landed on works pretty well. Players notice it, and that's the whole point.

The auth swamp

Then came the part I really struggled with: authentication.

On paper, a login system sounds like one of the simplest things in an app. In practice, it's a swamp. We went through three different authentication systems before landing on something we were happy with. The first one was too rigid. The second one felt clunky for the player. The third — the one we kept — is frictionless: you basically open the app and you're in. No forms, no passwords to remember, no friction between you and Foxtrot.

It's the right approach for a learning app. But even now, we still get the occasional crash that I haven't fully tracked down yet.

150 scripts

After all those changes, the code was starting to look like a patchwork. Three auth systems leave traces. New features stacked on top of older ones leave traces. So I took about three weeks to refactor everything. And by everything, I mean it: around 150 scripts to go through, one by one. Cleaning up, renaming, restructuring, fixing bugs, addressing the dozens of small warnings and hints I had been ignoring for weeks.

Boring work. Necessary work.

The quiet

And honestly? Those weeks were hard.

Not technically — the refactor itself was fine. But mentally. When you spend three weeks improving code that already worked, you don't see anything new on screen. No new feature to show Quentin. No new screen to test. Nothing satisfying to share. And meanwhile, the to-do list of things we still want to build keeps growing in the background.

It's frustrating. You feel like you're not progressing, even though you objectively are. The foundation gets stronger, the bugs go down, the code becomes something you can actually keep building on. But none of that feels like progress in the moment.

That's why I went quiet here too. Writing a dev blog post when you have nothing visible to show is hard. The work was real, but it didn't translate into anything I could screenshot.

I think every dev knows this period. The quiet one. The weeks that don't make it into the highlight reel but without which everything else collapses six months later.

So I'll just put it here, in the journal: it happened, it mattered, and we're better for it.