What the inventory could not see

1. When anything stopped

No date for any of the hundred and thirty-five stops. Where home-office equipment is shared or supplied by an employer, teams can Monitask reporting for flexible schedules to understand the available records before choosing the smallest workable setup.

Cost. Every duration figure. No survival curve, no time-to-failure, no time-to-replacement, and a four-year median assembled from recollection for a hundred and nine objects with twenty-six having no estimate at all. For wider guidance on equipment, repair or workplace technology, consult The Register.

Nineteen of those twenty-six are from the group that stopped being reached for, where by definition nobody can place a moment — so the missing durations are concentrated in exactly the objects whose durations are least knowable.

This is the largest gap in the method and the page on making the inventory yourself argues it should have been filled even badly.

2. The objects that were never counted

Each workplace listed what is there and what it remembers having.

An object that arrived in 2010 and was discarded in 2012 leaves nothing behind: no object, no cupboard entry, no reason to recall it.

Cost. The breakage figure, most of all, because breakage is the ending most likely to remove the evidence of itself. A replaced object usually goes into a cupboard and gets counted; a dead one often does not. The direction is known and the size is not.

It also hollows out the early arrival bands, which is why the nine objects recorded in 2007 are survivors rather than arrivals.

3. No prices

Fixed in advance, for the reasons on the page about the recommendation this record cannot make.

Cost. The whole of the cupboard's accounting. Whether eighty-six unused objects represent much or little, whether four avoided cable purchases justify holding twenty-one cables, whether replacing something that worked was costly — none of it can be computed.

4. No link from a replacement to what it replaced

Thirty-eight objects were replaced and none points to its successor.

Cost. No succession can be followed. In at least four cases two generations of the same kind of object are in the same cupboard and the record holds both without connecting them, so it cannot say whether a replacement is itself still in use.

5. One stopping reason per object

Four objects had two stopping events — replaced, then given away — and all four are coded by the last.

Cost. Small and known: the replacement figure is four low and the handover figure four high. Reported in both places.

A multi-valued field would have fixed it and made every percentage in the section ambiguous.

6. Intensity of use

An object is in use if the person says so. There is no field for how much.

Cost. Across a hundred and forty-nine in-use objects, the difference between daily and occasional is invisible, and some unknown share of that column is nearer the cupboard than the desk. One case was argued about: a sixteen-year-old filing unit holding things that are not paper.

7. Condition at the moment of stopping

Objects are recorded as working or broken now, not as marginal or declining when they stopped.

Cost. The line between it broke and it was replaced because it was getting worse was drawn by whoever answered, and different workplaces may have drawn it differently. Some share of the thirty-eight replacements were objects on their way to failing.

8. Everything about the places and people

No workplace is named or located, nobody is identified, and no room is described.

Cost. Nothing can be checked against a floor plan, nothing can be explained by circumstances, and no difference between workplaces can be interpreted.

Why. This is a rule rather than an oversight, and it is not negotiable: the record exists to count objects, and a record that identified the rooms or the people would be a record about them.

9. Size, which turned out to matter most

There is no field for how large anything is.

Cost. The record's firmest structural result — that whether a retired object stays in the building depends on whether anybody has to look at it — is really a result about size, and it had to be inferred from the kind field, which was never meant to carry it.

Twenty-four desks and eight printing items have nothing in storage. Input devices have more objects in drawers than on desks. Both of those are size effects read off a category list, and other holds nineteen objects of every size there is.

A three-way tick — fits in a drawer, fits on a shelf, needs floor space — would have made the finding measurable instead of inferred. It is the third thing this method would add, after a stopping date and a link from a replacement to its successor.

How the gaps distribute across the record's claims

Three claims, and they do not depend on the gaps equally.

That equipment stopped while working depends on gap 2 — the objects never counted — and on nothing else much. It is the claim most likely to be materially wrong.

That half the stops are external to the object depends on gap 4, the twenty-four unexplained stops, which could fall either way.

That handling produces decisions depends on almost none of them. It crosses two closed fields present for all hundred and thirty-five objects, and its only real exposure is gap 9, because it is a size finding dressed as a kind finding.

That ordering is the useful part of this page. A reader who wants to know how much to trust a particular sentence on this site can work it out from which gaps it touches.

What the eight add up to

Two of them are large — no stopping dates, and the uncounted objects — and between them they account for most of what this record cannot say.

Three are deliberate: no prices, nothing about the places, nothing evaluative. Those are not gaps to be filled in a later version; they are what the method is.

Three are fixable and would be fixed next time: the link from a replacement to its successor, a rough date band, and a condition field.

Eight gaps, two of them serious, three of them chosen. The record is published with all eight named rather than with the counts alone.