The gap

The Hour That Costs Almost Nothing

A gap of one hour leaves seven of the eight, and the hour it takes is the one nobody was using. That is why it misleads.

Seven hours out of eight survive a gap of one. In proportional terms that is a loss of an eighth, which is less than a day a week, and it is the only point on the curve where the arithmetic and the intuition agree that the damage is small.

What is worth noticing is which hour goes. It is not shared out across the day. It comes off the front.

Teams can look through the available reporting options, state the question the data is meant to answer and set a date for removing any setting that adds no value. A useful external reference for the surrounding topic is ILO.

The first three steps down

outlined: the day where the person is
hatched: the other end
solid: the hours they share
a reversed cell counts none

Four gaps drawn.

The morning share drains first. The afternoon share holds at four until the gap reaches four hours, then begins to drain in its turn.

Computed from the gap. The working day is a stated convention, not a measurement; everything else here is arithmetic.

The morning drains first

At no gap the shared stretch runs the whole day, four hours before the middle and four after. At a gap of one it runs from ten to five: three hours before the middle, four after. At two, from eleven: two and four. At three, one and four.

The afternoon share does not move at all until the gap reaches four, at which point the morning share has been exhausted and the afternoon begins to go instead. For the person whose clock is behind, the pattern is mirrored — the afternoon drains first and the morning holds.

Gap	Before the middle	After
0	4	4
1	3	4
2	2	4
3	1	4
4	0	4

That asymmetry is invisible in the single total, and it is usually the thing that actually determines whether the arrangement is bearable. A pair who need a morning hour together lose it at a gap of four, not at a gap of eight.

The total tells you how much time exists. The split tells you when it exists, and work is not indifferent to when.

Why one hour feels free and is not

A single hour of gap is routinely treated as equivalent to no gap. Two things make it feel that way: the loss is small enough to absorb without rearranging anything, and the remaining stretch still covers both the late morning and the whole afternoon, which is where most of the day's deliberate contact happens.

Both of those are true. The reason to resist the conclusion is that a gap of one is rarely a stable state. It arises because somebody has moved slightly, and the next move is the one that takes it to three or four, at which point the morning is gone entirely and the loss has stopped being absorbable.

Seven hours, and where they are

At a gap of one hour the shared stretch runs from ten to five on the clock of the person who stayed, which is nine to four for the person who moved. Seven hours, unbroken, covering the late morning, the whole of the middle of the day and the whole of the afternoon.

What has been lost is the first hour: the hour in which one person is at their desk and the other is not yet. For most arrangements that hour was never heavily used anyway — it is the hour of arriving, reading what came in overnight and deciding what the day is — and its loss is therefore close to invisible.

This is exactly why the gap of one is dangerous to reason from. The hour that goes is the hour with the least obvious value, so the loss is absorbed without comment, and the conclusion drawn is that gaps of this size do not matter. The conclusion is right about the first hour and wrong about the second, third and fourth, each of which costs the same and takes something that was being used.

What it looks like after a few months

An arrangement at a gap of one tends to settle rather than to be managed. Nobody schedules around it because nothing needs scheduling around; the overlap covers every hour anybody would have wanted.

The effect of that is that no habit forms. The pair never establishes a protected hour, never agrees where the fixed commitments go, and never develops the discipline that arrangements with thinner overlap are forced into early.

Which is fine, until the gap changes. If one end moves, or if a seasonal shift at one end only widens the gap for part of the year, the pair has no practice at working within a constraint and no agreed structure to fall back on. The arrangement that was comfortable at one hour frequently handles four worse than an arrangement that was at three from the start.

An arrangement that never had to be organised is not robust; it is untested. The first real constraint finds out whether anything was ever agreed.

What a single hour buys back

Read in reverse, the same line says that closing a gap from two to one returns a whole hour of shared time — the same hour that closing a gap from seven to six returns. There are no cheap hours and no expensive ones.

This is the strongest practical argument in the first part of the record: if any gap can be reduced at all, the reduction is worth exactly as much as it would be worth anywhere else on the scale. The question is never whether it is worth closing a gap. It is only whether the gap can be closed.