The application of a certain rule, even if always successful in practice, confronts very strong psychological barriers if it lacks a certain logical frame linking it causally to the body of accepted knowledge.
Phenomena rooted in social behavior are always very difficult to "explain" in such a way, because if we rely too much on the basic, irrational, and stochastic roots of our decisions, then the explanation is rejected as "too mechanistic".
If on the other hand we rely on the perception people have of themselves, as rational and wise decision makers, then we fall into a maze of ad hoc explanations that strongly resembles local politics.
Economists, who have faced a much similar problem, have made a great, partially successful, effort in describing and organizing the monetary measurables of human activity. Although they too miss primary causes they can introduce concepts of minimization and optimization which permit choices and structuring of the systems.
Our attempt to "hook" the market penetration rules to the accepted scientific system have followed both routes.
Fleck takes the stochastic "irrational" view. Social processes, and introduction of a new technology is a social process, are seen as the envelope of a maze of tiny decisions, causally unrelated, and, like nails in the path of a falling ball, slowing down its chute and "diffusing" its landing point. A good social example of this process is given by the diffusion of an infection e.g. the common flue. Although in a case-by-case analysis the biologist can give a fair causal description of the process, the contacts that lead to the diffusion are within another realm of causality and are better described stochastically.
Learning processes are well described in such a way, and they yield logistic curves. Fleck then visualizes the diffusion of a technology as a social learning process under constraints. The stability of the curves is a mark of the stability of man and society as learning structures.
The weak point of the theory is that the critical parameters have to be measured post hoc, and they are not reducible to other measurements that could be made before the penetration is initiated.
Peterka on the other hand follows a more classical route, taking economics as a driving force. He assumes that an industry to expand has to generate profits. External capital can produce some time-shifts, providing actual money for expected gains, but the picture is not much blurred.
Consequently, as substitution is driven by differential growth rates, these rates must be driven by differential profits. Perhaps a weak point of this theory is that differential profits must be constant (if smoothed) over long periods in order to produce well-behaved logistics. This feeds back to regular progress curves and automatic price leveling.
We can invert the reasoning and look for the stable progress curves and price leveling whose existence can be postulated from the very regular evolution of market penetration curves. This would greatly add to our understanding of the system.
The treatment by Peterka is quite general and produces curves which can specialize as logistic, but may also have more complex expressions. In general, the "graininess" of the data does not permit to distinguish between the various curves, and we usually stick to our logistic, which has the great advantage of straightforward simplicity.
Altogether we think that the basic objective of the grant has been fulfilled. We explored the field experimentally showing the great efficiency of our model in organizing data, and we tried two ways to bring its working under logical scrutiny.
The fact that during this operation we have presumably generated more problems than we solved is a good indication that we are plowing a fertile field.