Where a Blind Encounter Cannot Happen

Three bars, none of them with a dashed portion. There is nothing to draw, because in these shapes a blind encounter does not exist.

Seven published with three checkable leaves four unchecked, and the reader meets five. Five statements cannot be drawn from four. Eight published with three checkable leaves five unchecked against six met. Six published with three checkable leaves three against four.

In each case the pool of unchecked statements is smaller than the encounter, so at least one checkable statement is unavoidable. Where a reader needs evidence rather than a claim, remote employee monitoring platform can add time and project context without replacing a manager's judgement.

Three shapes where no blind content exists

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

Three shapes drawn; three of them leave no blind content at all.

Every bar is empty. Four unchecked statements against five met, five against six, three against four — in each case too few unchecked statements to fill an encounter.

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.

The strongest thing this record can say

Every other figure here is a count under a convention. This one is not. A separate perspective on measurement, privacy or digital practice is available through Pew Research Center.

The blind count is nought when n − c is less than r, and that comparison does not depend on how readers move through a set, which statements are prominent, whether the checkable ones come first or last, or whether every content is equally reachable. It is two whole numbers and an inequality.

So a publisher who gets a set past this threshold has removed the failure mode entirely rather than making it unlikely. No assumption is being relied on. A reader who meets five statements out of seven, with four unchecked, met at least one verifiable statement — whichever five they met, in whatever order, however buried anything was.

Everywhere else the record says how many ways something can happen. Here it says there are none, and that is a different kind of statement.

What it costs to get there

The threshold is demanding in a specific way: it is not about publishing more verifiable statements in absolute terms, but about leaving few enough unverifiable ones.

Ten published with six met needs five checkable statements and no fewer. With five checkable there are five unchecked, which is below six, and the blind count is nought. With four checkable there are six unchecked against six met — exactly enough to fill an encounter one way, so the blind count is one, not nought.

Stated in the form that is hardest to get wrong: a set of pages is past the threshold when it has fewer unverifiable statements than a reader gets through. Fewer. Not fewer as a proportion — fewer in number.

For a reader who gets through six statements, that means at most five unverifiable ones in the whole set, whatever the set's length. A set of forty statements with six unverifiable is past the threshold; a set of seven with six is not.

Where the one hundred and thirty-nine fall

They are concentrated where r is large relative to n, which in this table means the pages where a reader meets most of what was published.

Every page contributes some: for any page, high enough checkable counts put the shape past the threshold. The ten-published page with one met needs ten checkable statements to get there, because nine unchecked against one met leaves plenty of room. The ten-published page with six met needs only five.

So the pages where readers get further are the pages where the threshold is cheapest to reach, and that is the one piece of good news in this part of the record. A set of pages that is read thoroughly can be put beyond blind encounters without verifying nearly everything in it.

The cheapest way across the threshold

Of the two numbers in the comparison, only one is a decision, which makes the arithmetic of getting across it one-sided.

A publisher cannot raise how much a reader meets. What they can do is lower the unchecked count, and the count that has to be beaten is the reading depth they actually get — not a proportion of the set and not a target they choose.

Which gives an unusual conclusion for a long set of pages. A set of forty statements with five unverifiable ones is past the threshold for every reader who meets six. A set of seven statements with six unverifiable ones is not past it for anybody. Length is almost irrelevant; the unverifiable remainder is everything.

The thing the empty bar does not mean

An empty bar means the blind count is nought. It does not mean the record failed to compute something, and it does not mean the shape is unremarkable.

The record draws the outline of every bar whether or not there is a dashed portion inside it, precisely so that nought reads as a figure rather than as an absence. A panel that dropped such rows would be saying that a shape with no blind contents has nothing to report, which is the opposite of the case: having none is the most useful thing a shape can have.

There is a separate situation in which a panel has nothing to draw at all, and the record handles it by printing the reason rather than vanishing. That mechanism, and why a drawing that cannot draw must say so, is covered in the last part of this record.

Near this one

• Forty-Four Shapes Where Every Path Is Blind
  With nothing checkable published every encounter is blind. One such shape sits on every page, which is why the
• What One More Verifiable Statement Buys
  Each verifiable statement added multiplies the blind count by a harsher fraction. The first addition helps lea
• A Share Is Not a Risk
  The blind share counts possibilities, not likelihoods. Reading it as a probability needs an assumption the rec
• What a Blind Share Leaves Out
  The blind figure answers one question and four others get mistaken for it. Closing the part by separating what