BioSec: Evolution

Purpose:

To retell Darwin's argument associating the biological with the technological.

The sections are as follows:

Correlated Variation

(Annie)

Character of Domestic Varieties

(Annie)

Difficulty of distinguishing between Varieties and Species

(Annie)

Methodical and Unconscious Selection

(Elizabeth)

In discussing the selection process that takes place in the breeding of domesticated animals, Darwin distinguishes two kinds of selection: methodical selection, where breeders start with an idea in mind and deliberately attempt to create a new and superior species; and unconscious selection, where many people try to get ahold of the best animals and by doing so, improve the breed. Darwin is primarily interested in unconscious selection, and notes the importance of observation and documentation, so that these changes become noticeable.

In the world of computing, parallels of the process of unconscious selection can be seen in many situations. Some examples might be the process of code development – where the same task might be coded in several ways, and eventually unified into the most efficient process once all the possibilities have been explored. A more economically motivated example might be when multiple products are competing in a market. Though no particular consumer has any intention of improving the field of desktop publishing, all consumers want to use the product that creates the most beautiful documents. Thus, the program that most people buy gets further examination and development, leading to improvement.

“Man can hardly select, or only with much difficulty, any deviation of structure excepting such as is externally visible; and indeed he rarely cares for what is internal. He can never act by selection, excepting on variations which are first given to him in some slight degree by nature.”

Darwin reflects briefly on the nature of unconscious selection, and discusses how although humankind has shaped the development of various species, the ideas always come from nature, and humans must work within the limitations of the domain.

How is this true of computing?

Circumstances favourable to Man's power of Selection

(Elizabeth)

In the final section of the first chapter of “The Origin of Species”, Darwin discusses the kinds of circumstances that facilitate the successful domestic selection of animals. Several of the factors Darwin lists are contextual factors. He remarks that a high degree of variability in the population is needed to attempt selective breeding, and that someone must have access to and control over a large population. In addition, he remarks that there must be some obvious value in undertaking this challenge.

“… the animal or plant should be so highly valued by man, that the closest attention is paid to even the slightest deviations in its qualities or structure.”

From the perspective of facilitating the development and selection of computer programs, it seems possible that similar contextual factors are needed. In place of a high degree of population variation, we could perhaps substitute a high degree of knowledge, expertise and experience. It seems entirely sensible that an organization who has access to large amounts of expertise, experience, equipment (and presumably, budget) would be positioned to motivate and commission work in the area. Darwin’s remark that “... the [product] should be so highly valued by man, that the closest attention is paid to even the slightest deviations in its qualities or structures” can be directly applied to the construction of computer programs. It seems believable that development and evolution of computer products (whether hardware or software) are likely to be more successful when there is a clear need or use for the product.

(What kind of organizations will have these privileges? Large corporations, like Google, sure. But what about people developing open source software? Crowd-sourced efforts, like Wikipedia? )

Darwin also notes a few factors that affect selection that have more to do with the nature of the creature at hand. Clearly, determined selection will be much easier if the species can be easily crossed, but it is also necessary to be able to prevent species from crossing at will or at random. For computer products, these two factors seem relatively easy to accomplish. Programs usually change only in the way specified by the developers and programmers, and if left alone, most computer programs do not change, merge or reproduce. (What about computer viruses, things like that?)

Part of Darwin’s discussion is of the limits of natural selection – how far can selection be pushed? He acknowledges that some limitations are defined by the capability of the product, but he also notes that external pressures affect the limitations of development. For computers, I think these limitations include hardware limitations (presumably some products will be developed at the time that the hardware exists to implement them), but I think that the external limitations come not only from the perceived need for the product, but also from the need for someone to have an idea for the product. We cannot develop products we have not thought of, or have an idea of their use or necessity.