A sitting
What a sitting is
A stretch of work at one place, ended by something or by nothing. The unit the whole record rests on, fixed before the logs began, and what it cannot see.
One day of the forty-five, sitting by sitting
Six sittings across one day. Median of the days drawn here 44 minutes. Three of them end bare, because the work ended rather than the place. Where shared homes fragment schedules, teams can read the vendor documentation to understand the available records before deciding whether those records answer a real planning question.
Six sittings in one day, which is the median for this record. For broader research beyond this small household record, consult work published by Harvard University.
A sitting is a stretch of work spent at one place, from arriving to leaving. It ends when the person gets up and does not return to that place for the next stretch.
That is the whole definition. It was written down before any of the nine logs began, and it did not change afterwards.
Why the definition had to come first
A unit settled after the data arrives will always be settled in a way that flatters the data. If the rule about what counts had been decided in week six, it would have been decided by somebody who already knew which decision produced a tidier table.
So the rule was fixed, published to the nine households in the same words, and then left alone — including on the occasions when leaving it alone was inconvenient.
The five fields
Each sitting was recorded as five things and nothing else:
| field | what went in it |
|---|---|
| place | whatever the household already called it |
| start | clock time, to the nearest five minutes |
| length | minutes |
| ending | one of nine named kinds, or a dash |
| day | the date |
The dash is the important one and it has a page of its own. It means the sitting ended because the work ended, not because the place stopped working.
What a place is
A place is wherever somebody sat or stood to work, named by the people who live there. Not a room: a room can hold two places, and in four of the nine homes it did.
The rule for "the same place" is that the household says so. If somebody moved from one end of a table to the other and called it the same place, it was one place. If they called it two, it was two.
The names in this record are the names people already used. Nothing was renamed to make the columns line up.
That makes the place count — thirty-one across nine homes — a count of distinctions the households themselves drew, which is both its weakness and the only thing that kept it honest.
Minutes, not hours
A sitting is measured in minutes because the thing being measured is how long somewhere held, and the shortest sitting in the record ran six minutes. An hour-based log would have rounded that to nothing, or to one, and either way it would have disappeared.
The longest ran a hundred and ninety-four. Between those two the median is forty-eight.
What the unit cannot see
Three things, stated here rather than buried, because a unit's blind spots are part of its definition:
- Standing up without leaving. Somebody who stood for ten minutes and sat back down at the same table logged one sitting. The record cannot distinguish that from somebody who never moved.
- How anybody felt at any point. There is no field for it, no question about it, and nothing in this record can be turned into a statement about comfort in the ordinary sense of the word.
- Work with no place. A call taken while walking, a note written on a phone in a corridor, thinking done in the shower. None of it is here.
The third of those is the largest omission by far, and nobody knows how large, because measuring it would have required a different log.
Why not count hours
Hours tell you how much work happened. They do not tell you whether anywhere held.
Two people each work five hours. One works them at a single table. The other works them across four places, pushed out of three of them. The hour columns are identical and the days are not remotely alike, and only the second person was being moved by anything.
The unit had to be the thing that can end. Hours do not end; they accumulate.
The median day
Five sittings, two places. That is what a logged working day looked like in the middle of the distribution, and the day drawn above is one of them — six sittings, three places, a little busier than the middle and chosen for that reason.
Eleven days out of forty-five were spent entirely at one place. On the other thirty-four, people moved, and mostly between two points rather than around the whole dwelling.
What follows from the unit
Everything else on this site is a count of these objects or a comparison between them. There is no second dataset, no survey, no reading of anybody else's work. If the unit is wrong, the entire record is wrong in the same direction, and that is worth saying plainly at the start rather than at the end.
It may well be wrong. A unit that cannot see standing, cannot see feeling and cannot see placeless work is a narrow instrument. What can be said for it is that it is a consistent one, and that it was defined by somebody who did not yet know what the answer would be.