New Zooniverse project: Planet 4

Regular readers will know all about the Zooniverse stable of citizen science projects, where ordinary people volunteer their time online to help process large datasets. My favourite examples include Galaxy Zoo (sorting and classifying the millions of distant galaxies photographed in deep field surveys) and Moon Zoo (identifying and counting the billions of craters photographed by the Lunar Reconnaissance Orbiter, to train computer algorithms to continue the task automatically). The idea behind all these projects is that modern research methods return enormous amounts of data, far more than any group of scientists could ever process within a single lifetime. Usually this would be a job for computers, but some tasks turn out to be incredibly hard for machines to solve. Classifications based on simple patterns, that are particularly easy for our human brains to process, manage to confuse the computer every time thanks to subtle variations in shape, colour or brightness. By simplifying the task down to something that needs only the most basic training, and then crowd-sourcing it through the Internet, the job can be divided between thousands or even millions of people, and completed in a reasonable time.