Did you know the scientific method is nearly obsolete?  Now that scientists have access to petabytes of data scientists no longer need to hypothesize, model, or conduct tests. Correlation is more important than causation. It's amazing! Everything you ever learned about the scientific method is passé. At least that is what Chris Anderson claims in his article in the July issue of Wired Magazine. Mr. Anderson believes we live in the Petabyte Era. In this magical era information no longer has to be explained or examined. It can be taken at face value. If a petabyte of people were alive and all of them claimed that the sun went around the earth, then according to Mr. Anderson way of thinking, we would have the certainty of numbers to know that their conclusions were correct. This is what I am calling the petabyte fallacy; if we just had enough data the scientific method would be obsolete. Truth would stick out like a sore thumb. Well, I disagree. A correlation built upon a petabyte of poo will smell even more than one based upon a small data set.
 
The need for critical thinking and data analysis does not disappear just because we are dealing with an unimaginably large quantity of data. Numbers can never "just speak for themselves." Someone had to obtain the information and odds are they used models to get it. How do we know their models and methodology are correct without testing them? Any perceived correlation that matches our bias will just be blindly accept as verification of our preexisting ideas. No, the Petabyte Era is not the end of models or science. The massive amounts of information we have available to us will indeed lead to new discoveries, and if critical thinking is no longer valued, they will also take fallacious ideas to new heights. Sorry Mr. Anderson, but the need to observe, hypothesize, model and test is as important today as it was in the time of Newton.