Galaxy Zoo is a very simple solution, to a very difficult problem. We have a robotic telescope which sits in New Mexico called the Sloan Digital Sky Survey, and over the last seven years it's taken images of a million galaxies. It's a good census of our local neighbourhood - the nearest billion light years or so. What we want to do is study those galaxies to try and understand the details of how they formed. And the first thing you have to do is you have to split them into spiral galaxies that look like the Milky Way, and you have to split them from those, from big balls of stars which we call elliptical galaxies. And the best way to do this is by eye, because the human brain is much better at this than any computer - the problem is we've got a million galaxies. A PhD student I worked with at Oxford, Kevin Sevinsky, got as far as 50000. And the one thing we learned from that, so far, is if you look at 50000 galaxies you never want to see another one. So how do we do the rest? All we've done is we've put them on the web at and we're inviting people to come and join in and help us classify these galaxies. So, you register, you go through a very short tutorial - it takes a couple of minutes - take a test that almost no one fails and then you're presented with images of galaxies. And most of these, only a few, a handful of people have ever seen before. The images have sat on a hard disk for years, but no one, until we started Galaxy Zoo, has seen them. All we ask you to do, is sit and click through, and say "Is that a spiral or is that an elliptical?" And from that, that's the key to understanding how these things formed.
We have a million galaxies, but it's not good enough for one person to classify each one because we need to understand how accurate the classifications are as well. So, our target, which we're more than just over halfway towards, is to have 20 people look at each galaxy. So, we need 20 million clicks to classify the galaxies and we're well on the way, but the more people the better. We're also looking at ways to get other images. We will be using some Hubbell space telescope data to try and push this further into the distant universe, to see if things change as you get further away. So, we might be able to see not only these things forming but see how that process evolves over the last 13 billion years or so.


characters remainingSubmit