Length

Two Lists Of Ten, Two Hundred And Fifty-Six Apart

Hold the length fixed at ten items and the intake still ranges over a factor of two hundred and fifty-six. Length constrains almost nothing.

Home · Length · Reference

Take a list of exactly ten items. Forty-two shapes in this record have that length, and their intake runs from one profile in four down to one in one thousand and twenty-four.

A factor of two hundred and fifty-six, with the length held constant throughout.

Anyone applying list arithmetic to people should inspect the available configuration and preserve the difference between a signal and a judgement. For an independent technical reference, see Apache Kafka.

Four lists of exactly ten items

filled: a requirement that must be met
open: one of the optional ones
area of the inner square: the share admitted
a stroke instead: too small for this panel to draw

Four shapes drawn; one of them admits too little for the square to be drawn, and carries a stroke instead.

Every row has ten items on it. The filters they describe differ by a factor of two hundred and fifty-six from the widest to the narrowest.

Computed from the shape of the list. The shape is a stated convention; the count of admitted profiles is arithmetic on subsets.

The two ends

The widest list of ten is two compulsory items and eight optional ones with a threshold of none: the two compulsory items halve twice, the optional ones do nothing at all, and the result is one profile in four.

The narrowest is the same two compulsory items with all eight optional ones required: one profile in one thousand and twenty-four.

Same ten lines. The only difference is a single number that may well not appear on the page.

Shape	Items	Admits
2 compulsory, 8 optional, none required	10	1/4
2 compulsory, 8 optional, 3 required	10	219/1024
2 compulsory, 8 optional, 6 required	10	37/1024
2 compulsory, 8 optional, all required	10	1/1024

What this does to length as a signal

It destroys it.

A reader told only that a list has ten items knows that the intake is somewhere between one in four and one in one thousand and twenty-four. That interval is so wide as to carry no information: it includes filters that admit a quarter of everything and filters that admit almost nothing.

Told instead only the threshold — nothing about the length — a reader would often know more, because the threshold is the parameter with the leverage.

Ten items is not a quantity of difficulty. It is a quantity of writing, and the two are related only by whatever rule nobody wrote down.

Why the factor is exactly two hundred and fifty-six

Because at length ten the optional part can be at most eight items, and eight optional items span the full range from a threshold of none to a threshold of all — which is a factor of two hundred and fifty-six, since that is two to the eighth.

The compulsory items contribute a constant halving on both ends and cancel out of the ratio entirely. So the spread at a given length is governed by how many optional items that length allows, capped at eight by the record's bounds.

That is why the ratio is sixteen at length four, two hundred and fifty-six at length eight, and then stays at two hundred and fifty-six for every length from eight to sixteen: the optional part has run out of room to grow.

What a reader should do with a long list

Stop counting lines and find the rule. If the rule is stated, the record gives the figure directly. If it is not, the list is one of the many whose intake cannot be determined from the document.

That sounds like a small observation and it changes how a list should be read. The instinct is to scan a long list and form an impression of difficulty from its weight. The weight is the least informative thing about it.

The ten-item range, with a middle row

Ten items admit seven different splits between compulsory and optional, since neither part can exceed eight. Each split then carries its own set of thresholds.

The extremes and one representative middle:

Shape of a ten-item list	Admits
2 compulsory, 8 optional, none needed	1/4
5 compulsory, 5 optional, 3 needed	1/64
8 compulsory, 2 optional, both needed	1/1024

A quarter at one end and one in a thousand and twenty-four at the other. The middle row is not a midpoint of anything; it is simply one shape among many, and it sits sixteen times below the top and sixteen times above the bottom.

Why this is worse than it sounds

Because length is the one property of a list that a reader can assess instantly, and it is the property carrying the least information.

Counting the lines on a page takes two seconds and tells a reader almost nothing: at ten items the answer is somewhere in a range spanning a factor of two hundred and fifty-six. Recovering the rest requires knowing which items are compulsory and what the threshold is, and the second of those is frequently not written anywhere.

So the easy signal is nearly empty and the informative one is often missing. That combination is why lists get judged by length despite length being uninformative — it is the only thing available.

And what a writer should do

Decide the rule first, then write as many items as are useful.

A long list with a low threshold is a perfectly good document: it names a lot of relevant things and demands a few of them. The length then does useful work — it tells a reader what the work involves — without doing any filtering work it was not meant to do.

The failure is a long list with no rule, which communicates a great deal and selects unpredictably, and which the section on the unwritten rule takes apart in detail.