Moore’s Law: When Will It End—or Will It?
Everyone loves to predict the future. Whether it’s the soon-to-be newborn’s gender, lottery numbers, or the outcome of sporting events, people love making completely random guesses—while others love pointing out when those predictions don’t pan out. I’ve been poring over the recent headlines of countless experts predicting the upcoming end of Moore’s Law, created by Intel co-founder Gordon Moore in 1965. Wikipedia explains that Moore’s prediction:
…noted that the number of components in integrated circuits had doubled every year from the invention of the integrated circuit in 1958 until 1965 and predicted that the trend would continue 'for at least ten years.'
The hypothesis has remained true for more than forty years—without ever slowing down or being outpaced.
Former Intel CEO Craig Barrett humorously gives one explanation for why each year his company’s “new crop of engineers” has managed to keep the law ringing true, at least on Intel’s behalf. They’re each informed that “it better not end on your watch.”
While no one can deny Moore’s accuracy during the past four decades, skeptics of the theory are divided into two camps. There are those who believe computational power cannot continue to double forever, and there are those who feel that Moore’s Law limits the power of today’s pace of innovation and that the law most certainly can be topped.
Even Mark Zuckerberg has recently added his two cents, not to debate the power—or lack thereof—of Moore’s Law relative to computational power but to extend the theory to the world of social media. Zuckerberg recently told Wired: “Okay, in 18 months, someone’s going to fit this many transistors on a circuit board—we’d better be the ones to do it or else someone is gonna eat our lunch!” Zuckerberg referenced Moore's Law when he stated:
I look at this the same way. Three years from now, people are going to be sharing eight to ten times as much stuff. We’d better be there, because if we’re not, some other service will be.
Whether you’re on “team Moore” or not, the power of today’s computers, software, and the speed of production is both awe-inspiring and somewhat daunting. A recent BBC interview with computer scientists at MIT explains that whether Moore’s Law ends in the next ten to twenty years, one area that is not keeping pace with the power of computers is that of job creation—that is, jobs for actual humans. The job forecast that they and many others see is eerily robotic.
