The first fortnight

Here's what I'd expect in the two weeks after the window goes in, so that the things that happen don't look like the method failing.

The first evening, the wonder is loud. The phone is in the kitchen and you know there's something in it, because there's always something in it, and part of you is in the kitchen with it. Pielot and Rello's volunteers, in chapter 11, felt this. Less distracted, more productive, and some of them anxious about not being as responsive as people expected. The anxiety is the class in your own head, the version of you that learned the lesson too. It doesn't mean the window's wrong. It means it's new.

By the fourth or fifth evening, the wonder is quieter, and here's why. You've had four mornings of opening the inbox and finding that nothing in it needed you last night. Four is enough to start believing it. Chapter 5 asked you to keep a week's list, and this fortnight is the same list, running in real time.

Somewhere in the fortnight, one thing happens. A message arrives at seven in the evening and you don't see it until eight the next morning, and there's a small consequence. Someone had to wait. Someone used the old numbers. It feels like proof that the window was a mistake. Look at it closely. It was a small consequence, and the person it happened to had a fallback, and used it, and the world went on. That's the evidence you needed. The consequence you were spending every evening to avoid turned out to be this size.

Somewhere else in the fortnight, the other door doesn't ring. It doesn't ring all fortnight. Sit with that too.

The class adjusts faster than you'd think. By the second week, Priya sends the slide 8 question at half past eight in the morning, because that's when she's learned it gets answered. Your manager, if you had the conversation, has stopped copying you in on the ten o'clock ones, or hasn't and doesn't expect anything. The thread that used to have four exchanges at night has one in the morning. Nobody says anything about it, because from the class's side nothing dramatic happened. A shop changed its hours and put up a sign.

What I'd not expect is for the wonder to vanish. It gets quieter and it stays quiet as long as the window holds. Pielot and Rello followed up their volunteers two years later and found that half of those who'd meant to change how they handled notifications were still doing it, which is a better rate than most changes get and a reminder that the other half had drifted. Chapter 23 is about drift.

The practical point is to treat the fortnight as an experiment with a known result. You already know what the inbox looks like in the morning. The fortnight is just where you find out that the evening version was the same inbox, watched.