about

Nothing here is measured. Every assumption is published.

The Everything Network is a set of small live counters for ordinary things. Each one answers a single question, and each one shows the working behind its answer.

What this is

How many people are eating right now. How many are asleep. How many are in the middle of a shower, a kiss or a walk. Each site answers one of those questions, and nothing more.

Every site runs the same software. They differ only in their subject, their palette and their arithmetic. Adding another one is an entry in a list plus a folder, which is why the network keeps growing.

How a number is made

Every figure on every site is built from the same three things.

one A population How many people could be doing this at all.

two A rate How often the average person does it.

three A duration How long each occasion lasts.

Population, multiplied by rate, multiplied by duration, gives an estimate of how many people are doing the thing at this moment. The local clock is the fourth input, and it is the one that makes the numbers move: the world does not share a schedule, so the estimate follows the sun around it. Eating rises through each country's mealtimes and falls overnight, and the total drifts rather than climbs.

A worked example. The eating site assumes about four eating occasions a day, lasting roughly fifteen minutes each. Four occasions of a quarter of an hour is an hour a day, or a little over four per cent of it - so a little over four per cent of the world has a fork in hand at any given moment. That number is not measured. It is the assumption, written down where you can disagree with it.

Why the numbers are not real

Nobody is being counted. There is no census of people in the shower, and no site on this network can see into your bathroom, your car or your kitchen. Each figure is a model: an estimate produced by arithmetic from stated inputs.

That is a limitation, and the sites are built to make it visible rather than to hide it. The rate, the duration, the source of the population data and the date it was taken are all printed at the foot of the site you are reading. Where a model needs a rate that nobody has published, the rate is a guess and it is labelled as one.

If an assumption looks wrong to you, the assumption is on the page, in plain language, where you can argue with it. That is rather the point of publishing it.

The one thing that is measured

Each site also carries a vote: are you doing this right now, or not. Those counts are real. They are people who pressed a button today.

They are kept deliberately separate from the model - in their own panel, labelled as visitor responses, and never added to the estimate above them. The two figures answer different questions: one is how many people probably are, and the other is how many said so. Merging them would make a better number and a dishonest one, so the pages never let them meet.

What happens to a vote, and what does not, is set out in the privacy page.

Who makes this

Built and maintained by PAJ. The sites are deliberately small, deliberately free of accounts, and deliberately honest about what they cannot know.

Questions, corrections and arguments about the arithmetic are all welcome at info@ing-now.com. Corrections are the most welcome of the three.

Where to go next. The hub shows every counter in the network at once. The privacy page sets out what is stored when you press a button, and what is not.