Predictive Discovery

The use of preliminary organizational work of such cohesive integrity that eventual discovery of what's missing is predictable.


Whoa Nellie! This one is a little hard to explain, so we resort to explanation by example:

Consider the Periodic Table, where chemical elements are arranged by atomic number (number of protons) and the way their electrons are organized. Early versions contained many gaps for elements that were discovered much later. Discovery was predicted with perfect accuracy, because certain elements had to exist to fill in the blanks.

Or consider the obscure Higgs boson, an elementary particle whose existence is predicted by the Electroweak Theory. It has not yet been detected, but should have finite mass and no spin. Even though not yet discovered, it's used to explain various particle behavior.

Or consider the fields of biology and archaeology. Examining the evolutionary chain of a species leads to frequent gaps that are used to make predictions about what fossils are yet to be discovered. Sure enough, they often are. They are sometimes so anticipated that the final fossil finding is more of a corroboration than a discovery.

Predictive Discovery largely falls into extrapolation and interpolation patterns. Extrapolation involves identifying a trend and the existence of more data points, or things, beyond the trend. Interpolation is the same except the unknown lies within the data end points. "Data points" can be theories, abstractions, facts, etc.

How can Engineering take advantage of the phenomena of Predictive Discovery?

1. Process - This creates an "activity framework" where one fills in the blanks, like interpolation. Eventual and correct discovery is much more likely due to use of a process.

2. Design - Many times you will rough out a design with missing elements. By not worrying about those that are missing you can proceed to accumulate more and more elements until the missing elements will suddenly "pop out", due to emergence of extrapolative or interpoative forces.

Use this when your work entails a high rate of discovery.