I had intended to update an old blog titled Revisting the Basics with this visual. It speaks of a maple tree as a metaphor for building skill in a complex realm. However, it plays a large role in my efforts towards helping BI folks build the business acumen skills required for effective consulting of Predictive Analytics. The major role of Predictive Analytics is to supply the robust logic required for software applications to better serve the complexities of human society than the other way around. 

 

Predictive Analytics is a technology that impacts all industries as well as opening up new types of software companies. In order to

 

The time is near for us to place our bets on the roulette wheel for the eventual economic recovery. I'm placing my bet on "Predictive Analytics for the Masses" both as a stakeholder of Microsoft (I work for Microsoft, my skills are obviously and unavoidably Microsoft-centric, and of course, I own stock) and as an entrepreneur (at this time that pretty much means as someone who wishes to contribute towards Microsoft's BI vision). The roulette wheel is spinning now in the middle of all this economic and political crap; once the wheel stops, it's too late. 

 

I strongly believe that the individuals or companies that possess core competencies around Predictive Analytics (see my blog, Why Isn't Predictive Analytics a Big Thing? and its accompanying links for more on why) will be the players for this economic recovery. It's hard to imagine what companies that would be. It could be the current players such as Microsoft, Google, or Yahoo. It could be a smaller company who's time has come such as SAS (not that they'd want to blossom into a mega-sized company). There's a good chance it could be a totally unknown player as were all of the previous players at one time. SCL?

 

Predictive Analytics is one of the major technologies that decouples rules from procedures. Predictive Analytics is all about automatically discovering rules to recognize phenomenon within complex, ever-changing systems. These "soft-coded" rules are very different from the rigid "hard-coded" rules that dominate our software applications today. It is what makes systems smart instead of inflexible and dumbed-down.

 

So why didn't we always write software that way? Because the hardware and infrastructure didn't exist. As an analogy, think about selective breeding versus genetic engineering. Agriculture shares with computers the fact that production improvements over the past few decades rise in orders of magnitude. That's an important point since population is pretty well aligned with the amount of food we're able to produce. 

 

Genetic engineering is clearly more flexible and fast than selective breeding. So, why didn't we use genetic engineering from the start? Obviously because the surrounding technology didn't exist and wouldn't exist until just recently. People weren't even aware of genes until a few decades ago. But mankind made use of the best technique available until the technology arrived that would open up completely new, infinitely superior approach.

 

We are there with software as well. We don't need to think in terms of automating only well defined processes (dumbing down). We can build systems that better blend with our human ability to think our way through complex and sometimes novel situations.