Limits

Why Nothing Is Named Anywhere Here

No organisation, product, service or person appears in any sentence here. Three reasons, and the third is about what happens to pages over time.

Home · Limits · Reference

Nothing in this record is named. No organisation, no product, no service, no tool, no place and no person appears in any sentence here, including sentences that would deny something about one.

That is not squeamishness and it is not a limitation of the subject. It follows from three things, and they are worth setting out because the absence is conspicuous.

Two checkable statements, three page sizes

the whole bar: every content a reader might come away with
the dashed part: contents holding nothing checkable

Three shapes drawn.

Two verifiable statements in each set, with the reader meeting half. The blind share rises from one fifth to three fourteenths to two ninths as the page grows.

Computed from the three numbers. A blind content is one holding no checkable statement: that is a property of the page, not a prediction about any reader.

A named example becomes a judgement

The moment a counting problem is illustrated with a real organisation, the figures stop being arithmetic and become a finding about that organisation.

A blind share of two ninths is a property of a shape. Attached to a name, it reads as an accusation — that this particular set of pages leaves its readers unable to check things — and the accusation would rest on a classification somebody made by hand and on a convention about reader behaviour that the record itself says is wrong.

The figures are not strong enough to carry a name. They are counts under declared assumptions, and a count under assumptions is fine as a description of a shape and unfit as a verdict on anybody.

A name would not be checkable

The record's own claim to be auditable rests on every figure following from three printed numbers. A reader can recompute all of it.

A named example breaks that immediately. The reader would have to take on trust that the organisation publishes ten statements, that a reader meets six, that three are verifiable — three judgements, none recomputable, all of them doing the work the arithmetic was supposed to do.

So a record that named things would be exactly the kind of set of pages this record is about: assertions a reader has to accept because the author says so.

A name dates, and the arithmetic does not

This is the reason that settles it.

Facts about organisations expire. They are reorganised, renamed, absorbed, closed. A set of pages built around named examples starts decaying on the day it is published and nothing on it signals when it stopped being true — which is the failure that empties an address of its readers and eventually of its owner.

The counting problem does not move. P(8, 4) was one thousand six hundred and eighty before anybody wrote this down and will be afterwards. Every figure here can be recomputed from the three numbers beside it, which means there is no date at which any of it quietly stops holding.

A record with no names in it has nothing to go out of date. That is not a modest property; it is the only one that survives a change of ownership.

What the absence costs

Concreteness, and the loss is real.

A reader who wants to know whether their own set of pages has this problem gets no worked example from the world here — only shapes, and the instruction to count their own three numbers. That is more work and it is the only version that stays true.

The record also cannot show the one thing a named case would show best: whether the statements that matter are the verifiable ones. That question needs content, the record has none, and it is listed among the four things no figure here can answer.

What is named here, and why that is different

Two things on this site do get named: the arithmetic and the conventions.

P(n, r), C(n, r), the factorial, the binomial coefficient — these have names and the names are used, because they refer to operations a reader can perform rather than to anybody's property. Naming an operation commits nothing and dates nothing. The three conventions are likewise named and numbered so that they can be cited and rejected individually.

That is the whole of the record's vocabulary of proper nouns. There is no organisation in it, no product, no person, and no previous occupant of this address — not because any of them would be unflattering to mention, but because a name is the one thing on a page that cannot be recomputed.

The drawing's three rows

Two verifiable statements in each row, the reader meeting half the set.

Six published with three met: four blind contents of twenty, one fifth. Eight with four met: fifteen of seventy, three fourteenths. Ten with five met: fifty-six of two hundred and fifty-two, two ninths.

The share rises as the page grows, because two verifiable statements go less far in a longer set. One fifth, three fourteenths, two ninths — 0.200, 0.214, 0.222.

Three anonymous shapes, three exact figures, and nothing to date. A reader can check every one of them in a minute and does not have to believe anything about anybody.