UCLA professor Terence Tao, one of the world's top mathematicians, has just solved a famous problem dating back to the 1930s—and he says it was a comment on his blog earlier this year that sent him in the right direction. He also built off earlier crowdsourced work to solve what's known as the Erdos discrepancy problem, which, as New Scientist explains, involves "the properties of an infinite, random sequence of +1s and -1s." For the technical minded, Nature presents it this way: "(Paul) Erdos, who died in 1996, speculated that any infinite string of the numbers 1 and −1 could add up to an arbitrarily large (positive or negative) value by counting only the numbers at a fixed interval for a finite number of steps." For the non-technical minded, the important thing to know is that "Terry Tao just dropped a bomb," as Iowa State University mathematician Derrick Stolee tweeted the day Tao's paper was published on the new open-access journal Discrete Analysis.
Tao, who met and worked with Erdos when he was a kid in the 1980s, joined a few dozen mathematicians in 2010 as part of the Fifth Polymath Project to harness the power of many human brains working on blogs and wikis to try to finally crack the code. While they made progress, they ultimately gave up. Then earlier this year, Uwe Stroinski, a mathematician out of Germany, commented on Tao's blog that the problem might be linked to what is called the Elliott conjecture, which Tao at first dismissed as being only superficially similar. But after careful review, "I realized there was a link," Tao wrote. He produced the proof less than two weeks later. It still must undergo peer review, but most think that will wrap up quickly. And yes, Tao has thanked Stroinski for his contribution. (This billionaire will pay $1 million if you solve his 20-year-old conjecture.)