Stretching

Where Stretching Stops Paying

Some gaps cannot be reached by any amount of lengthening. Beyond a point, the hours bought are hours nobody can use.

Lengthening a working day buys one additional workable gap per hour, in each direction, with no curve. The returns do not diminish in the arithmetic.

They diminish sharply in every other sense, and it is worth being precise about where and why.

When evidence is scattered across tools, it may help to see the current service description and see whether one report can replace several manual summaries. A useful external reference for the surrounding topic is Basecamp.

What the far gaps have to offer

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

Eight gaps drawn; four of them leave nothing at all.

Four reversed cells in a row. Stretching moves the boundary outwards one gap at a time and each gap it reaches delivers the row above the reversal, which is one hour.

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

The hours bought are always the worst hours

Each hour of lengthening brings the next gap out into range, and the shared time at that newly reachable gap is always one hour.

A four-hour lengthening takes the furthest workable gap from seven to eleven. At a gap of eight, nine, ten or eleven, the shared time is one hour. Four hours of additional working day have purchased four arrangements that each deliver a single shared hour.

Lengthening	Furthest gap reached	Shared time there
1 hour	8	1
2 hours	9	1
3 hours	10	1
4 hours	11	1

That column is the whole argument. The lever does not stop working; it works perfectly and delivers something of very little value.

The gaps that cannot be reached at all

A gap of twelve hours is the widest in this record, and nothing here reaches it. Reaching it would require one end's working day to be twelve hours long, at which point the two days touch at a single instant and share nothing measurable.

Past twelve the gap begins closing again from the other side, because a day has only twenty-four hours in it. That symmetry is why this record stops at twelve in each direction: every offset beyond that is an offset already counted, approached from the other way round.

The unreachable region is not a gap in the record. It is the part of the problem where the arrangement is between people who are, in working terms, on opposite sides of the day.

Why there is no clever construction

Two working days of ordinary length cannot be made to meet across a half-day gap by any rearrangement that leaves them recognisable as working days.

The split shape reaches furthest, to a gap of nine, by having two blocks that can catch two different parts of the other day. A shape with three blocks would reach further still and would also be a working day in name only — three two-and-a-half-hour blocks spread across fifteen hours of clock is not a schedule anybody keeps.

The record declines to compute such shapes, because the arithmetic would be easy and the result would be a number attached to something nobody would do.

What to do instead

Stop trying to reach. An arrangement across a gap of ten hours or more has no shared working time, cannot be given any by lengthening within reason, and should be built as an arrangement that does not need any.

That is not a failure of the arrangement. The section on three ends makes the same point and it holds here: work organised around handovers, written artefacts and decisions that can wait a day is a coherent way to operate, and it is strictly better than a working day stretched to twelve hours in pursuit of one shared hour.

What the hours bought are actually like

A single shared hour at a gap of eleven is not an hour in the middle of anybody's day. It sits at the extreme edge of both.

The person who lengthened their day to twelve hours is in their twelfth hour. The person at the other end is in their first. One of them has been working since the morning and the other has just arrived, and the hour they share is the worst hour either of them has to offer.

That is a consequence of the geometry rather than an incidental detail. The hour bought by stretching is always the hour at the boundary, because the boundary is where the stretching reached. There is no arrangement in which lengthening a day buys a shared hour in the middle of it.

What was done	Where the shared hour falls
nothing, gap of 7	end of one day, start of the other
lengthened to 9, gap of 8	the ninth hour against the first
lengthened to 12, gap of 11	the twelfth hour against the first

The comparison nobody makes

Set the two options side by side honestly.

Option one. A twelve-hour working day, indefinitely, producing one shared hour in which one party is exhausted and the other has not started.

Option two. An eight-hour working day, no shared hour, and work organised around written exchanges with a day of latency.

Stated that way, the second option wins for almost any work and almost any person. It is also the option that feels like giving up, which is why the first one keeps getting chosen.

The record cannot say which is right in a particular case. It can say that the first option's benefit is one hour of poor-quality overlap and its cost is four hours of daily work, and that this ratio does not improve at any point on the scale.

Stretching does not fail gradually either. It works exactly as advertised and what it delivers is not worth having, which is a different kind of failure and a harder one to see.

The boundary, stated plainly

Up to a gap of seven, do nothing special. Between seven and nine, the levers in this section are worth considering. At ten or beyond, there is nothing here that helps, and the right response is to stop looking for it.