# Why is a day divided into 24 hours?

23 February 2016

## Question

Why do we have a 24 hour day cycle, and could we make it metric?

Stuart - So this is really interesting because it's actually an historical question. So historians have attributed this division of the day into 24 hours to the ancient egyptians. And the idea is they had 10 hours for the daytime, 10 hours for the nighttime, plus 2 hours each for the dusk and dawn, and that's what formed the 24 hour clock. And there's an idea that this might have been linked to the number of stars that astronomers were seeing in the night sky at the that time, that maybe there was around 12 stars that occur during different phases of the night. And that's what gives us our 24 hour clock.

Kat - But where does the 60 minutes and 60 seconds come from?

Stuart - So that even more interesting. It's not from the same place and so this idea of 60, this base 60 system. When I say base 60, I mean when we start counting from 0 to 59, that's the number of digits that represent our time. That kind of goes back to the Babylonians and the idea that they were fascinated with this idea of 60 and we're not too sure why but historians think, or people have speculated that it's due to the fact that 60 is a very good number for dividing into different fractions. And that's the way the base 60 minutes and 60 seconds comes from.

Kat - Well would it work if we say had metric time. Would that be possible or just completely confusing for everyone?

Stuart - So it's important to specify what we mean by metric time. Our SI unit of time, the base unit that scientist use to measure time is the second and that's standard.

Kat - You get milliseconds though. So that's going from base to into decimal.

Stuart - Yes, so that is already a metric, a decimalised form of time and that's commonly used throughout the sciences because it's easy to work with; we like counting to ten. And actually it's a real problem because if you think about how computers work, which are reliant on date and time, how to record that on a computer is a really challenging problem. So a lot of systems, in particular ones that are based on an architecture called Unix, use a specific point that's the start of time. They use January 1st, 1970.

Chris - 1970. And that's what lead to the millennium bug, wasn't it?

Stuart - It's one of the reasons. The millennium bug specifically was about what happens when the programmes were being a bit lazy and only recorded the last two digits of the date, like 00 or 01. But actually, there's a proposed problem coming up in 2038 which is where, basically, we ran out of space to store the number of second since 1970, which is how computers count time.

Chris - So what will they do about the Unix time stamp problem then when potentially run out of space? Is there a new solution to surmount that?

Stuart - Yes. So the problem is down to basically how much space you allocate in the computer to count the number of seconds. So what Unix time means is that the computer is just counting from zero seconds (1st January, 1970) up to the current time in seconds and that has it's own problems because, actually, the solar year and the solar times are all complicated. There's leap seconds, we need to adjust for the rotation of the earth moving - it's a very complicated issue. But what you can do, of course, is just redesign your system and if you implement, basically, software updates. That's what you need to do - update your system - very important.