The commute is the only scroll of the day that ends on time. The train arrives and it is over. That external edge is doing all the work, and it is why nobody feels bad about their commute scroll while feeling terrible about an evening that was shorter. Take the edge away, as remote work did for millions of people, and the same behaviour stops being harmless.
Why this one does not hurt
Every other scroll in the day is bounded by your own decision to stop, and that decision is exactly what the feed is built to prevent. There is no seam between items, no episode ending, no natural moment where continuing becomes a choice. That architecture is set out in the infinite scroll mechanism.
The commute solves that from outside. The doors open. You stand up. The session ends at a moment you did not have to choose, and because you did not have to choose it, no capacity was spent.
Three other things make it unusually clean:
The time was already spent. Twenty-eight minutes on a train is twenty-eight minutes whatever is in it. The scroll is not displacing anything, because the alternative was looking at a window.
The state is right. You are between two things, carrying neither. Nothing is being avoided, which is the difference between this and the stretch described in the phone before work.
There is a container. A beginning, a middle, an end, imposed by a timetable. Almost nothing else in a day has that shape any more.
What happened when the commute went away
For a lot of people the container disappeared in 2020 and never came back, and what disappeared with it was not travel time. It was the only reliably bounded phone session of the day.
The behaviour did not stop. It moved. It reappeared in the morning before starting work, where there is no edge at all, and in the evening in front of the television, where there is no edge either, and the same twenty-eight minutes became forty and then ninety. The evening version has its own account in where the evening went.
This is the part people misread about themselves. The commute scroll was not a habit they successfully controlled and then lost control of. It was a habit that was controlled by a timetable, and when the timetable went, nothing about the habit had to change for the outcome to get much worse.
The commute that stopped working
There is a version of the commute scroll that does go wrong, and it has a specific tell: it does not end at the station.
You get off the train still holding the phone. You walk to the office reading, cross a road reading, arrive at your desk and finish the video before opening anything. The container was there and the session ignored it.
When that starts happening, the edge is no longer external, because you have learned to step over it. This is the same learned dismissal that makes iOS limits stop working after a fortnight, described in you tap Ignore Limit before you read it, and it produces the state where the reach and the arrival have no decision anywhere between them, examined in picking up the phone without deciding to.
The other tell is what you feel when the train is delayed. If a twelve minute delay reads as twelve free minutes rather than an annoyance, the scroll has stopped being a way to pass the commute and has become the reason the commute is tolerable.
What to do about it, which is mostly nothing
The correct answer for most people is to leave the commute alone.
It is the wrong target. It is bounded, it is not displacing anything, and it is the one place where the phone is doing exactly the job it is good at. Tools that treat all screen time as identical will flag it, and the flag is wrong, which is a large part of why people abandon those tools inside a month.
Two narrow exceptions.
If the session survives the station, then what is needed is an edge at the end rather than a limit on the length. Something at the arrival, not something at minute fifteen. The distinction between interrupting on a schedule and interrupting on behaviour is worked through in interrupting only while you scroll.
If the commute is gone, then the thing to replace is not the scrolling, it is the container. The behaviour was never the problem. The timetable was the solution, and it left. What stands in for it has to be small enough to survive a person who is not being watched by a train, which is the sizing argument in what should happen when your phone interrupts you.
reCalm agrees with the do-nothing answer, which is why shields are per app, never per phone. Leave the commute apps alone and put the breath only where the evenings go. And the exception is covered: an arrival edge is just a shield that asks one question, so a session that survives the station meets one breath instead of a flag.
Frequently Asked Questions
Is scrolling on my commute bad?
Usually not. It has a hard external ending, it displaces nothing, and it does not require you to spend any capacity deciding to stop. It is the healthiest phone habit most people have.
Why does a short evening scroll feel worse than a long commute scroll?
Because the commute ended on time without you choosing, and the evening did not. The feeling tracks whether you had to stop yourself, not how many minutes passed.
My commute scroll follows me into the office. Is that different?
Yes, and it is the one warning sign worth acting on. It means the external edge has stopped registering, which is the same failure that makes app limits stop working.
I work from home now and my phone use went up. Why?
The commute was a container, not a habit. The habit was unchanged when the container disappeared, and habits without containers expand into the nearest gap, usually before work and after it.