Temperature

Cold, mostly

Thirteen of the eighteen are cold. Five are warm. A record of rooms and interruptions becomes more useful when it is paired with actual project hours, so a team can privacy-aware productivity reporting before defining a limited, transparent pilot.

The imbalance is less interesting than it looks, because the logging months were not evenly spread across the year. Six of the nine weeks fall in the colder half, and the two homes contributing most of the cold endings logged in November and February. The wider design and work context can also be checked through UCLA.

What can be said without reference to season is that temperature endings cluster hard. Three dwellings account for twelve of the eighteen; four dwellings have none at all.

What the three have in common

Home F is a bungalow. Home I is a ground-floor flat. Home C is a terrace with a through-room.

Every one of them has a place that the household described in terms of being away from the rest of the dwelling — a back room, a far end, a return. Those are exactly the spots that a heating system sized for a household's evening habits will reach last.

The three homes with none are a one-bed flat, a room in a shared house and a four-bed semi with a study. In the first two the working place is in the same space as everything else; in the third it is a room the household heats deliberately.

Temperature endings happen where somebody works in a part of the dwelling that the dwelling was not being warmed for.

That is a plausible story and it rests on nine cases. It is included because the alternative — reporting the eighteen and declining to look at them — is not neutrality, it is just less information.

The warm five

All five are in two homes, both logging in the warmer months, and four of the five are in the afternoon.

Three of those four are at places that also produce glare endings. Sun on a surface is both a light problem and a heat problem, and the instrument forced households to pick one word.

This is a genuine ambiguity in the record. If somebody at a sunlit table at two in the afternoon writes temperature and somebody else at a comparable table writes light, the two endings are the same event under different names.

How often that happened is unknown. The overlap is at most a handful of entries, because only nine endings in the whole record occur at a place that produced both kinds, but a handful of eighteen is not nothing.

The hours

Cold endings are a morning phenomenon: nine of the thirteen fall before eleven. Warm endings are an afternoon one: four of the five fall after one.

Both make straightforward physical sense, and both mean that a place assessed at a single hour is being assessed in a way the record says is inadequate. A back room at half past eight in February and the same back room at three are not the same proposition.

What a household can do

More than with light, and still not much.

Heating a room that is otherwise unheated during the day is possible and costs money, and the record contains no information about whether anybody weighed that trade, because the sheet did not ask and the question would have been intrusive.

Moving to a different place is free, and it is what these households mostly did. Of the eighteen temperature endings, fourteen were followed by a sitting elsewhere within the hour, and eleven of those were at a place closer to the middle of the dwelling.

The remaining four ended the working day.

Two readings this will not bear

The first is that anybody was cold in any sense that matters beyond moving rooms. The log recorded a decision to go somewhere else, and a decision to go somewhere else is all it recorded.

Nor is it a claim about anybody's heating, which the record knows nothing about: not the system, not the schedule, not the cost, not whether the household would have described any of this as a problem if they had not been keeping a log that asked.

The four that ended the day

Fourteen of the eighteen temperature endings were followed by a sitting somewhere else within the hour. Four were not: the working day stopped.

All four are cold rather than warm, all four are after three in the afternoon, and three of the four are in the two weeks logged in November and February.

That is the same pattern as the three light endings that ended a day, and it points the same way: in winter, late in the afternoon, the alternative to a cold or dark room is not another room. It is stopping.

Seven entries across the whole record, which is too few for anything except noticing that both of the envelope causes behave this way and none of the other seven causes does.

What is not claimed

That anybody worked less because of it. A working day that ends at four may have ended at four anyway, and the log records only that the last sitting of the day had a named ending rather than a dash.