Where are the new posts about the law of accelerating returns? I had promised a daily-series of hacks that showed how something as complex as the eye could come about from "simple" evolution. I've either lost steam, or I've been really tired these past few days.
It could be the steam thing though. A similar situation happened to me when I went through Consciousness Explained... before reading that book, I was really curious about how consciousness worked... and wanted to answer questions like, "yeah, where exactly is the red when I see a red dress?" Thing is, I read about a quarter of the book and got a gist of the illumination process... like he provided some key metaphors that made it clear to me that it could be proved or described... and once I knew it was possible, or felt the mental gestalt that occured when the unknown and mysterious seemed attainable through more iterations of examples, I lost interest. I'm going through the same feeling with Linked. Like I read the first few chapters and got that "woah" feeling. But then as I went through, I thought, well, let me guess, everything's connected in an amazingly ridiculous ways and that power laws hold. Same with Godel Escher Bach.... it's like, "oh, let me guess, there is some sort of thing or trick within mathematics like infinite recursion or incompleteness from which you can gesture that consciousness arises out of it... all the while relating it to escher paintings and bach's complex compositions." It's like all I care about is the general feel. I do care about content, to some degree, but incompleteness doesn't bother me. As long as I understand the basic concept, get a few good anecdotes, and experience the author's unique style, then I'm done.
Danger, danger, danger... I'm either a pseudo-intellectual or have a high bar for novel content. I read the first four of the five books in The Gay Science (the 5th book he wrote many years after the first four, so I had a good excuse)... but I read that form beginning to end, everything he said was novel and exciting... not the same hat repeated.
Of course it sounds bad to go through this kind of process of incomplete books, and of course that's not how you were raised to do things (well, actually in high school I felt that was THE way, the only way in order to get by crappy courses, by skimming through for a gist). But just as they say, having read one book really well is better than reading many books poorly... why couldn't that "one book" be shortened to "a portion of a book." Plus, for three-quarters of a book I don't finish, I get free time to check out three other books I could dabble in.
I dunno, I just felt like washing my laundry here this morning.
Read more about tools and algorithms that aid biological evolution, and in tandem systems of accelerating complexity.
Groups of related organisms are 'variations on a theme' -- the same set of bones are used to construct all vertebrates. The bones of the human hand grow out of the same tissue as the bones of a bat's wing or a whale's flipper; and, they share many identifying features such as muscle insertion points and ridges. The only difference is that they are scaled differently. Evolutionary biologists say this indicates that all mammals are modified descendants of a common ancestor which had the same set of bones. (Introduction to Evolutionary Biology)
Mammals for example, all share a similar underlying framework, while the various common parts are simply scaled differently. A bigger brain here, a longer leg here, shorter mandible there, and voila you get a wonderfully different organism.
Scaling is also easy, as we discussed with runaway sexual selection. Relative bone size increments or decrements within a specie could be selected for constantly, allowing those parts to grow over time. So imagine a little bud of a claw on some rodent, being stretched into the huge leg of a dog. Or imagine if the females of our ape-like ancestors were attracted to males with the largest skulls of the group, then overall capacity of the brain would definitely expand.
Templates with even a modestly sized feature set can allow for tremendous complexity. Take the 206 bones of the adult human body. If the size of 30 of those were cut in half and another 30 doubled, the nature of our existence would be dramatically changed. We'd seek different food, we'd have different capabilities, the tools we would make would be different, etc. In other words, our niche in the ecosystem would shift.
Since the nature of our actions is based on the combination of all of our parts, changes to a small set of features propogate large changes to the whole system.
This ability for small changes to have large impacts is important for rapid adaptations. As environments change and competition for resources becomes fiercer, a species cannot be re-written from scratch. So groups of organisms that can make small modifications to a templated organism will be better suited at discovering new niches as old ones disappear.
Templatization appears naturally in other systems. Look at blogging for example. The ability to create date-based journal entries existed about as early as I can remember on the Internet with places like BluesNews. It wasn't until somebody made a standard template of imagery, dated entries, and links that we have this "revolution" called Blogging. By the simple scaling of those three features we get a tremendous range of complexity in blogging. Take Howard Dean's Blog for America. There we see the imagery using a patriotic color scheme, with his picture in there showing he's active, and a general cleanliness that evokes a sense of pure democracy. Then look at the scaling on the dated entries. His blog updates so frequently that he puts himself into the category of a "mega-blogger." But even beyond that, these entries are scaled up by allowing for mulitple authors form his campaign team plus the comments of visitors. Then look at his blogroll or list of links. The number and choice of various pundits and other Dean sites turns his site into a portal for political change. All Dean did was cater the style to a certain message, scale up the date-based entries, ramp up the size of the blogroll, and all of a sudden Dean has created a standarbearer for an emerging genre of "blog-portals." (Note, by Dean I mean whoever created his blog).
Similarily, the same blog template can be used for a personal diary. I could scale the blogroll down to zero, and use a default style, and focus on daily entries of the same tone.
Once again, small changes in scale, large changes in purpose.
Templates are readily apparent in human systems, such as business hiearchies, programming systems such as object-oriented-programming, or religious frameworks such as "Christianity." If you make a few changes to any of those templates, you get a unique organization with a rich new set of pursoses to fulfill.
Templates are a powerful expedient of complexity. And since they occur naturally within systems that seek creative ways to fulfill niches, it's no wonder than accelerating returns happens as fast as it does.
My conversion from Catholicism to Agnosticism coincided with my first exposure to Darwinian natural selection in the 9th grade. In Natural Selection I saw the beginnings of what seemed like a better sketch than Christian genesis. However, I had my doubts about Natural Selection; Darwin's theories still made it seem like it would take forever to bore out only average or gray species. How was this simple thing supposed to ever spontaneously produce life, or create something as complex as the eye?
These questions remained unaswered for me until I read Kurzweil's Age of Spiritual Machines, wherein I first learned about hidden mechanisms and algorithms laced within the structure of order and time that showed how rapidly increasing complexity was the rule.
This begins the first of a long-series--hopefully daily--of metaphors and vignettes that will make you feel more confident about the power and truth of the law of accelerating returns.
The first one that's simple is runaway sexul selection. Sexual selection is a subset of natural selection. This is a process where features are selected for during the mating process, such as having big muscles or colorful plumage.
Here is an excerpt from Introduction to Evolutionary Biology
Evolution can get stuck in a positive feedback loop. Another model to explain secondary sexual characteristics is called the runaway sexual selection model. R. A. Fisher proposed that females may have an innate preference for some male trait before it appears in a population. Females would then mate with male carriers when the trait appears. The offspring of these matings have the genes for both the trait and the preference for the trait. As a result, the process snowballs until natural selection brings it into check. Suppose that female birds prefer males with longer than average tail feathers. Mutant males with longer than average feathers will produce more offspring than the short feathered males. In the next generation, average tail length will increase. As the generations progress, feather length will increase because females do not prefer a specific length tail, but a longer than average tail. Eventually tail length will increase to the point were the liability to survival is matched by the sexual attractiveness of the trait and an equilibrium will be established. Note that in many exotic birds male plumage is often very showy and many species do in fact have males with greatly elongated feathers. In some cases these feathers are shed after the breeding season.
In this model, features can explode out of nowhere. For example, among human populations, there could be various subpopulations with different preferences for male height. In one particular subpopulation for example, the women could favor men who were at least one standard deviation above the mean. Over time, this would pressure the mean height to rise. If this increase were successful for this subpopulation in its survival rate, then this preference for height would propagate to other human subpopulations.
This illustrates a sexual selection based on intraspecies relativity, and not just feature-existence. By women having a preference for the best of a group, a consistent pressure for quality-increases emerges. Eventually, the utility of the quality-increases could surpass its use for survival. Well, having a knack for speedy quality-increases could be what gives a certain species a competitive edge over other species.
This also makes it obvious how intelligent homo sapiens could have evolved from not so intelligent ape-like ancestors. If subpopulations of women (or men) continuously favored the handful of smart members of their group, then the overall mean intelligence would drift up.
To further accelerate this process, throw in the growing ease of communication among humans throughout history (dark ages excluded). The more connected we became, the faster trophy mates would become norm, giving rise to a higher standard for trophy mates. Even today, celebrities probably get laid the most for (previously) good evolutionary reason. A phenotype that is celebrated the most by a species should have its genotype spread proportionally as well in order to improve the overall quality of the species.
The impact of increased connectivity on evolution would be an example of a network effect. More on this later.