The First Time My Family Played the Real Thing

There’s a particular kind of nervousness that comes with showing someone you love something you’ve built.

It’s different from showing a colleague, or posting online, or demoing to strangers. With strangers you can handle indifference. With colleagues you can handle critique. But with your family — the people who will tell you the truth whether you want it or not — there’s no hiding.

I’d been working on the first real playable version of Three Word Tale for a few months. Long enough that I’d stopped being able to see it clearly. The interface made sense to me, but I’d built it — I’d watched every button come into existence. I had no idea if it would make sense to anyone else.

On a Sunday evening in June, I opened my laptop at the kitchen table and said, more or less: “Do you want to try the thing I’ve been building?”

It was immediately confusing in exactly the ways I hadn’t anticipated.

The join screen worked fine. Everyone got in. But the first word phase — I’d designed it to go in order, each player waiting their turn — wasn’t obvious from the interface. My daughter thought it was her turn when it wasn’t. My partner typed something and wasn’t sure if it submitted. There was a moment where we genuinely weren’t sure if the game was broken or if we were just doing it wrong.

We were doing it wrong. But “doing it wrong” shouldn’t be possible in a game this simple. That was on me.

I resisted the urge to explain while they played. That’s the rule I’d given myself: if I have to explain it, it’s not finished yet. So I watched, and I took mental notes, and I made myself sit on my hands.

Then something shifted.

The story started to take shape. Somebody added something unexpected — I think it was something about a detective who was also a sandwich — and everyone laughed at the same time. And suddenly they weren’t thinking about the interface at all. They were thinking about the sandwich detective. They were leaning forward. They were arguing (gently, happily) about where the story should go next.

That’s the thing about a game that actually works: it disappears. You stop seeing the buttons and you start seeing the story.

By the time we hit the drawing round, I’d forgotten to be nervous.

My daughter pulled up the canvas and started sketching. It wasn’t elaborate — the canvas was basic, this early version — but she was in it. She was drawing the sandwich detective with complete commitment. She held up her phone to show us the result and laughed at her own drawing.

When the AI assembled the story into the comic strip at the end, there was a moment of quiet while we all looked at it.

“That’s actually really good,” my partner said.

They didn’t mean the app. They meant the story. Ours. The one we’d made together in twenty minutes at the kitchen table, which now had a beginning and a middle and an end and a drawing of a detective made of rye bread.

That was the moment I knew I had something.

Not finished. Not polished. The interface still needed work. The joining flow was confusing. I had a list of things to fix that filled half a notebook. But the core of it — three words, then you pass, then you draw, then you have something — that worked.

It worked in the way that mattered: my family wanted to play it again.

That’s the only test that counts.