Big Data: Risks and Implications

Review of:   Big Data: A Revolution That Will Transform How We Live, Work, and Think by Viktor Mayer-Schonberger, and Kenneth Cukier.   Strongly recommended, for reading through or browsing.  Part III.

I continue my look at this book. Finally looking at the sections on implications and risks.  Here too the authors do an excellent job of outlining, with a number of good and recent examples, the risks of relying on Big Data to understand the world.

First are the implications for privacy.  As more data is gathered, it is not only the primary direct use of the data, but also the derived implications that can violate privacy.   A very non private world can be created.

The the authors suggest that the quantification of systems have led to their dehumanization of our world, using as a primary example McNamara's Quantification of the Vietnam War.   The enterprise I joined early in my career was a very quantified entity.  Even in recent years they are attempting to 'digitize' their operation in more complete ways.   I was brought in specifically to do analytics, and by doing that, optimize the enterprise operation.   It worked well.  So I am defensive about this angle.

This comes back to the main point of the book, that Big Data is about What rather than Why.  We now have the data to record and potentially keep ALL the cases pertinent to its operation.   Even when these data are varied and rapidly changing.    Which leads then to the ability to rely on correlation rather than causation.    Much easier than figuring out the causal drivers, but potentially risky in its inaccuracy.  While getting causal results can be useful,  with Big Data they are not needed to get value.

So in a number of ways the chapter on risk argues against the earlier chapters which strong argue for the value of Big Data.    Can these two worlds exist together?

Again, an excellent book that lays out the argument of the dilemma.