What this site is not

Not a survey

The eleven workplaces were inventoried because somebody agreed to it. There was no population, no sampling frame and no weighting.

So no percentage here is a rate for anywhere else. Fourteen per cent of stops were breakages in these eleven places; the figure for any other set of places is unknown and this record offers no estimate of it. There are no confidence intervals anywhere on this site, because an interval on a convenience set of eleven is a number that looks like statistics and is not.

The eleven are not even independent of each other. Most of the thirty-one objects retired by a change of work are the same move away from paper, appearing once or twice in each workplace, so that figure is nearer to three or four events than to thirty-one.

Not a guide to what to buy

No brand is named anywhere. No model, no specification, no price. No object was examined, tested, measured or assessed by anybody.

There is therefore no page here that says what to buy, and no figure on this site that could be turned into one. Where a reader might expect a recommendation to follow from a finding, the penultimate section sets out why it cannot.

Not a comparison

Objects are recorded by kind — a desk, a second screen, a task lamp — and two objects of the same kind cannot be told apart in this record.

The per-kind figures that do appear are counts of one person's classification, since kind was the one field assigned rather than asked. They are used for description and never as an order of merit. The duration figures in particular look like an order of merit and are not: they sort kinds by how many ways a thing can stop being used, which is why desks come first and audio last.

Not a benchmark

The record has no stopping dates, so it contains no failure rate, no survival curve and no time-to-replacement. Its four-year median for stopped objects is built from recollection and is flagged wherever it appears.

Anybody wanting a figure for how long desk equipment lasts will not find a defensible one here, and the section on duration explains at length why the six-year and four-year medians cannot be made to supply it.

Not a judgement on anybody

No workplace is named, located or described. Nobody is identified. No arrangement is called good or bad.

Twenty-seven per cent of the equipment counted here is kept and unused, and the record declines to read that as waste or as prudence. It shows instead that no decision was ever made: the share of objects ending up stored runs between twenty-five and thirty-four per cent across all seven origins, which is what a default looks like rather than a preference.

Not finished

Five questions would be settled by asking the same eleven workplaces a second time, and none of them needs a price or a brand: whether the twenty-four unexplained stops acquire explanations, whether the cupboard grows, whether five pending repairs ever happen, how long things actually last, and whether the two empty rows stay empty.

The record is one reading of eleven rooms. It is published with nine named gaps, three claims and four refusals, and it does not ask anybody for anything.