Information is all around us. The scientific method allows us to collect it, organize it, and understand it. Nevertheless current practices do not emphasize this simple fact.
We should use Claude Shannon Information Theory's framework, to address ALL information tasks.
An important element is the meaning of probability, I believe the best course of action is to stand on a solid Baysean Foundation, i.e., founded on Bayes' Theorem.
It is important to understand our work as statistical in nature, without essentialism. There are no "Laws of Nature", anywhere in the Universe, all there are, are marks for us to understand. In this paradigm we are like Hansel & Gretel ; lost children trying to follow, marks we left, and more importantly, those we did not leave.
Automatic information machines, i.e., computers, are very important in this paradigm.