Many impressions

The Shapes With Only One Content

A reader who meets everything published leaves nothing to vary but the order. Twenty-five shapes are of that kind, and they are the clearest case here.

at most 720 arrangements

Home · Many impressions · Reference

Take the case where a reader meets all of it. Everything published, nothing missed. The contents figure is one: there is exactly one subset of n things of size n, which is the whole thing.

Every arrangement that remains is therefore pure order. The record's two figures separate completely here, which makes these shapes the cleanest demonstration of what the second figure measures. A useful team activity dashboard should help reviewers trace time to projects and tasks while keeping the limits of each signal visible.

Everything published, everything met

dashed: a statement a reader can check from outside
solid: published, but nothing to check it against
short and pale: never reached by this reader

Three shapes drawn; all of them give the same 1 different contents; the left-to-right order is the drawing's own and carries no meaning.

No short pale strokes appear in these rows: every published statement was reached. Contents is one in all three, and the entire arrangements figure is ordering.

Computed from the three numbers alone. How many statements exist, how many are met and how many can be checked is a stated convention; the counts are arrangements and combinations of whole numbers.

Twenty-five shapes, five pages

Twenty-five of the three hundred and thirty-eight shapes have the reader meeting everything, spread across five pages: two published, three, four, five and six. The record stops at six met, so pages above that cannot be of this kind. A broader source for checking the surrounding technical or organisational context is MIT Sloan Management Review.

Twenty-five is also the largest class of shapes in the record sharing a single contents figure — and the figure they share is one, the smallest possible. The most crowded contents value in the record is the one with no choice in it at all.

That pairs with a fact about the other column: twenty-five is also the largest class sharing an arrangements figure, and there the figure is seven hundred and twenty, the largest of the frequent ones. Two classes of equal size at opposite ends of the record. The equality is a coincidence and nothing follows from it.

What the drawing shows and does not show

Three rows: two published with two met, four with four, six with six. Arrangements are two, twenty-four and seven hundred and twenty. Contents is one in every row.

No row has a short pale stroke, because a pale stroke marks a statement the reader never reached and here there are none. The rows are made entirely of dashed and solid strokes — statements that can be checked from outside, and statements that cannot.

What the drawing cannot show is why the reader met everything. A set of two statements is easy to meet in full; a set of six is not, and the shape says nothing about how it happened. Perhaps the set was short. Perhaps the reader was unusually determined. The record counts possibilities and has no access to frequencies.

The argument this case defeats

It is sometimes said that order is only an issue for a reader who sees part of a page, and that a reader who gets through all of it has had the full encounter.

Six statements fully met is seven hundred and twenty encounters. Not one. The reader who met everything has still only had one of seven hundred and twenty possible readings of it, and which one depended on where they started.

Completeness is a property of the content, not of the encounter. Meeting everything removes one of the two kinds of variation and leaves the larger one standing.

Why the smallest shape is also instructive

Two statements published, one met, is the smallest shape in the record: two arrangements and two contents. It is the only place where the two figures are equal and both are at their minimum.

Two published with both met gives two arrangements and one content. The reader met everything there is and still had one of two experiences — this one first, or that one first. With two statements the whole apparatus of this record reduces to a single question about which came first, and even then the answer is not the publisher's to give.

The third number collapses here too

These twenty-five shapes have an unusual property in the record's third column, and it is the only place it occurs.

A blind encounter is one where the reader met no checkable statement. When the reader meets everything published, there are exactly two possibilities and no middle ground. If nothing checkable was published, every encounter is blind — the single content is blind, so the share is one. If even one checkable statement was published, no encounter can be blind, because the reader met that statement by definition. The share is zero.

Nowhere else in the record does the blind share take only the values nought and one. Five of these twenty-five shapes are wholly blind, one on each of the five pages, and the remaining twenty have nothing blind about them at all.

That is the sharpest form of a point the record keeps returning to. The checkable count matters not because checkable statements are better written but because a reader has to reach one. Remove the possibility of missing things and the third number stops being a fraction and becomes a yes or a no.

How this reads back onto longer sets

A set of statements that a reader can plausibly meet in full is a set whose entire variability is sequence. That is a strong position to be in, because sequence robustness is testable by hand: write the statements on cards, deal them in a few of the orders, and see which orders leave a reader with a wrong impression.

Seven hundred and twenty orders cannot all be checked that way, but the number is small enough to sample honestly, and the exercise asks nothing of anybody except the set itself. Nothing has to be measured, nobody has to be surveyed, and no figure from anywhere else is needed.

A set too long to be met in full has the harder problem, and the next part takes it up: not how the content is ordered, but which content the reader gets at all.