10 o'clock

Return to the Newsroom About
← the newsroom

Ranked on how far a story travelled, not on who is reading.

What it ranks on

How many places carried a story, with the countries that carry everything counting for less than the countries that rarely carry anything.

Why not just count countries

The countries are not interchangeable. In one window Britain appeared in 24 of the 40 corroborated stories and the United States in 16. Namibia appeared in one.

“Carried in Britain and America” is close to no information, because those two carry everything.

So each country is weighted by how rare it is in the window. Same inverse-frequency idea the clustering uses on words. A story reaching Ghana, Kenya, Japan and Ireland outranks one reaching Britain and America, because it had to cross something to get there.

Regions were tried first

They do not work. Four buckets, and every large story hits all four, so the ranking saturates immediately. Measured on one window: the top six stories all scored four out of four and the order barely moved.

A measurement that cannot tell its own top six apart is not a measurement.

What it changes

On one window a story carried by 12 publishers across 11 countries ranked first, ahead of one carried by 18 publishers across 10.

And a cricket result with six independent publishers, all of them in Britain, India and Malaysia, dropped off the page entirely. Not less true. Regional, and this page is for the other kind.

Nothing here knows who you are

No account, no profile, no reading history, no cookie that follows you, no engagement signal of any kind. A recommendation system has to know something about you to narrow what you see. This knows nothing.

Every visitor gets the same order, and it comes out of a published list of publishers rather than a model of your behaviour.

What it cannot do

It demotes stories that matter enormously in one place. A national budget, a local disaster, a domestic court ruling: all rank low here by design, and none are less important for it.

It also cannot tell you a story is true. Publishers in twelve countries can all be running the same agency copy, which is what Legwork is for. And all twelve can be wrong, which nothing here can fix.

The rest of it

How stories are grouped, which publications are read, how often and how politely: how this is measured.