Limits

What Lies Outside Eight And Eight

Three kinds of filter this record does not describe, listed so that its silence about them is not mistaken for a claim.

Home · Limits · Reference

The record covers lists of up to eight compulsory items and up to eight optional ones with a whole-number threshold. Three things lie outside it, and they are different in kind from each other.

One: longer lists

The easiest to dispose of. A ninth compulsory item halves again; a ninth optional item extends the staircase by a tread following the binomial coefficients.

To keep the comparison fair, teams can explore the management toolkit and hold the project, reporting period and correction case constant. For an independent technical reference, see Databricks.

Nothing new happens, the law is on the page about the record's ceiling, and anybody can extend the table themselves. The record stops because the rows stop being informative, not because the arithmetic stops.

Two: rules of a different shape

Harder, and genuinely outside.

Real requirement lists sometimes contain rules this record cannot express:

Items that only count if another is present. "Experience with X, or with Y provided you also have Z."
Items that substitute for each other. "A degree, or five years in the field."
Weighted items. "Three points for this, one for that, eight points needed."
Rules that vary by which items were met, rather than by how many.

Each of those is a different filter with its own arithmetic, and none of the figures on this site applies to them. The record describes a conjunction over one set of items and a count over another, and that is all.

The bound at eight is a decision about where to stop printing. The bound on the shape of the rule is a decision about what this record is, and it cannot be relaxed by adding rows.

Three: everything about people

The largest absence and the one covered at length elsewhere.

The record counts profiles and knows nothing about who holds them, how common any of them is, whether items travel together, or what happens when a document meets an actual population. Those pages are the first three in this section.

Why the absences are listed

Because a record that does not state its limits invites the assumption that it has none.

A reader who finds no nine-item row might conclude that nine-item lists are impossible. A reader who finds no substitution rule might conclude that substitution is somehow covered by the thresholds. Neither is true, and neither would occur to somebody who had been told where the edges are.

Listing the absences costs a page. Leaving them implicit costs the reader's ability to tell what the record is claiming.

Why eight, and why twice

The two caps are a choice and not a discovery, and the reasons are worth stating so that nothing here looks inevitable.

Eight is where a list stops being something a person holds in their head. It is also the point at which the spread between the widest and narrowest shape at a given length stops growing, so extending the cap would add shapes without adding range. And it keeps every denominator in the record at sixty-five thousand five hundred and thirty-six or below, which keeps every figure checkable by hand.

Capping the two parts separately rather than the total is what produces the structure the later sections describe — the peak at one in two hundred and fifty-six, the ladder of classes, the flattening of the ratio at eight. A record built on a single cap of sixteen would have different structure, and that structure would be an artefact of its convention exactly as this one is of its own.

What a reader outside the bounds can still use

The method, which does not depend on the caps at all.

A compulsory item halves the intake whether there are three of them or thirty. A threshold over optional items admits the running sum of binomial coefficients from the threshold upward, for any number of items. The product of those two is the intake, and nothing in that calculation knows about a limit of eight.

So a reader with a list of twenty-five items can compute its intake in the same minute, against a denominator of thirty-three million and change. What they lose is the comparison tables on this site, which stop at sixteen. What they keep is every figure they can derive themselves, which is the part that mattered.

A fourth absence: items that are not yes or no

Every item here is met or not met, and a great many real items are neither.

Experience measured in years, a language held to some degree, a qualification held in a related field — these admit degrees, and a rule over them is not a threshold over a set but a comparison against a level. That is a different kind of filter with different arithmetic, and none of the figures on this site apply to it.

The record treats such an item as a coin because the convention demands a coin. Where a list is mostly made of graded items, the right response is not to adjust the figures here but to recognise that they are measuring a different object.

The one thing inside the bounds worth restating

Within them, the record is complete and exact. Every shape of list with up to eight compulsory and eight optional items and a whole-number threshold has been computed, and the fraction given is the fraction — no rounding, no sampling, no approximation.

That combination — exact inside a stated boundary, silent outside it — is the most a record of this kind can honestly offer, and the final page sets out what it still will not tell you.