10 o'clock

Return to the Newsroom

When I was a kid, the 10 O'Clock news was always there

The adults watched it on their favourite local channel.
2, 5, 7, 9, 32. Whichever channel, it came on at ten.

These days, news is everywhere, all the time. You get it even when you aren't looking for it. It makes understanding the world around you a little more chaotic. Link farms and news aggregators have been around for a long time, but even those seem to have lost some of their integrity under the weight of their own success or stagnation. The lines between information and blatant marketing have blurred.

What I've tried to do here is give it edges again.

One list of publishers. One hour. One newsroom.

It reads 499 publications in 101 countries and folds the reprints it can detect back into the stories they came from. One agency files, four hundred outlets run it, and most rankings carry that as four hundred items - which puts the story in front of you four hundred times without a second person having checked it. Repetition buys prominence, not corroboration. What is left after the folding is the number of separately owned publishers behind a story, and it is usually far smaller than the headline count. That gap is the whole thing.

The folding is not complete, and it would be dishonest to let that sentence stand without this one. Two outlets under different owners running the same agency copy are caught when they name the wire and missed when they do not. Legwork says so on its own page; it belongs here too.

Publishing the same story more often does not lift it here, because repetition is the thing being discounted. Publishing it through more separately owned outlets would - that is harder to arrange than running it twice, which is the point, not a guarantee. It knows nothing about you either, so it cannot tell you what you want to hear.

The 10 o'clock news wasn't better because it was shorter.
It was better because it wasn't trying to demand your attention.

Every story in the newsroom has been through this.

19677 articles everything published, across 499 feeds

remove duplicates, wire copies and shared ownership

independent publishers separately owned outlets that actually carried it

read the articles, trace the attribution

attribution groups publishers whose cited sources overlap, which is not the same as one traced origin

So here is what it does

Five ways of looking at the same twelve hours, and one way of checking a story you already have. Every number on every one of them can be traced back to the articles it came from, and the whole point is that it takes about ten seconds to see whether a story rests on several reporting paths or on one account repeated.

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 engine has to know something about you to narrow what you see. This knows nothing, and every visitor gets the same newsroom in the same order.

This is not a truth score

Ten independent publishers can all be wrong, and this cannot tell you that they are. It tells you how many of them there actually were.

Who builds it

I built it as a prototype, and it runs on shared hosting. It rebuilds itself every hour and nobody edits the results.

The full working is on how this is measured: every threshold, every feed, what the last run did and what it got wrong, along with the list of every publication read and how politely.