← Desk

Timer

An interval you set, and two marks
Work on one thing for an interval of a length you set yourself. When an interruption comes from inside you, make one mark; when it comes from outside, make the other; and let the interval run on. The sheet of marks is the part worth keeping.

The interval

Pick a length you can win. The founder's first interval was ten minutes, chosen because it felt like a dare. His paper's preferred unit was thirty, and its range ran from twenty to thirty-five with forty as the ceiling. There is no sacred number.
20:00
One thing until the bell.
After a handful of intervals, take a longer rest.

Today's sheet

One mark per interruption, against the interval it came in. After a week you have a record of where your attention leaked and whose doing it was. No ratio, no score.

Earlier sheets

What the record says. Keep a fixed interval, of a length you set yourself. Plenty of people find that helps. On the small evidence there is, it helps how you feel, a little, and you should not expect it to change how much you get done. In 2023 a study of eighty-seven students across one day of real self-study found "no difference between groups in invested mental effort or task completion." In 2025 a study that randomly assigned ninety-four students to one two-hour session found the fixed twenty-five and five "led to a faster increase in fatigue," with "no differences" in productivity, task completion or flow. The sheet with its two marks is the part the season would save from the fire: it costs a pencil, nobody has tested it, and it would still be kept before anything else in the method. The method paper's one scientific sentence says it has been "proven" that twenty to forty-five minute intervals maximize attention, and the ISBN on its one citation belongs to a study guide to Animal Farm. The season has not read the guide: the claim may be in it; it may not. Either way, one study guide is not a proof.
Why this works, in one picture →
Card 7 of the Desk Playbook: the interval lengths the method's own paper gave, what two small student studies found, and the citation behind the word "proven".