Parallels can be drawn from the natural selection process to human developed parametric logics. As a highlight in the introduction, this chapter is to demonstrate the non-randomness of the so-called cumulative selection process. The example of pebbles by the beach suggests that as a system is under development, non-randomness would spontaneously occur.
Single-step Selection vs. Cumulative Selection –> Regeneration, Subsequence
Single-step Selection: random variation
Cumulative Selection: mutation
Recursiveness is one of the key characteristics of embryonic development, which is a process that computer modeling is designed to imitate similarly just like how genes work.
In computer modeling, value represents genes. All genes exert minor quantitative influence on development.
According to the author, he set the shape of the computer trees to be symmetrical as a constraint in order to define computer biomorphs. Meanwhile, 1. reproduction is asexual. 2. in the example, all the mutations occur one at a time. The author gets the inspiration from the fundamental principle of biology, which shows that genes influence embryonic development of the body yet remaining unaffected by their participation in bodily development, while their likelihood of being passed on may do so.
This is also what that differs reproduction from development in computer model.
A key point in computer modeling is the ability to emerge rather than being built.
Another interesting point the author made is that the so-called create process in biomorph land isn’t random but must-be-efficient-creative-searching process. The perception of developing creative methods to find is quite interesting because I feel it can change the mindset for beginners who have just stepped into the computer modeling world.
Even though the author was focusing on the topic of Biomorph Land, I think what we could learn from the example, or in general, the nature, can be applied to human’s perception of problem solving in general. It is true that betting on the odds might get you a great leap forward, yet the high risks contain to much randomness. On the contrary, although people don’t like disciplines and step-by-step development for their constraints and slow pace, they are more controllable and in most cases, would provide more solid and decent rewards later.