HPC blog When it comes to Big Data, I’m as geeked out as the next guy – if not a little more so. For the last three years or so, I’ve been telling anyone who will listen (and plenty of people who won’t) that Big Data and enterprise analytics are the "next big thing" both in business and computing. Today, it’s widely accepted that Big Data is going to make big changes to our world.
But not everyone is on the bus. Or if they’re on the bus, they’re not entirely sure that it’s headed in the right direction. An article in Datanami summarises an MIT Technology Review interview with Dr Peter Fader, co-director of the Wharton School of Business, in which the good doctor tosses cold water on some of the most hallowed Big Data precepts.
Dr Fader (and the interviewer, to be fair) also coins the terms "data fetish" and "data fetishist" to describe the belief that people and organisations need to capture and hold on to every scrap of data, just in case it might be important down the road. (I recently completed a Big Data survey in which a large proportion of respondents said they intend to keep their data “forever”. Great news for the tech industry, for sure.)
Even though I’m a big proponent of Big Data, I found myself nodding in agreement throughout the summary and the full interview (available here). Dr Feder makes some great points that cut through much of the hype surrounding the trend.
What rang particularly true was his comparison of those who rely on Big Data to technical stock analysts. They’re the guys who predict future stock prices based exclusively on past price moves. They use cool terms like "Head and Shoulders", "Abandoned Baby", and "Dark Cloud Cover" along with TRIN, TRIX, and ADX to describe price and market behavior.
The problem is that their analyses don’t work. Or, more accurately, they don’t predict individual stock moves any better than random chance. Their models don’t take into account the fundamental reasons why a stock moves up or down – bad management, bad products, etc.
Dr Fader makes the point that today’s data scientists are exactly the same as the stock chart jockeys. They simply look at loads and loads of data, fit some patterns to it, and then start making decisions based on that model. He predicts that the resulting success will not only be lower than expected but, as he puts it, "surprisingly low". ®