Archive | September 2016

Is the devil really in the details?

Is the devil really in the details? It seems an obvious yes; even with careful consideration. Though I do wish to stress the importance of detailed consideration, I don’t think that focusing on discrete aspects of nature, in this stage of our development is the best way to approach modeling economic systems. This requires some qualification in the form of very strong arguments and I hope that I can do it justice.

With respect to observation and interpretation of data the observer is always a part of the equation. The observer is influenced by the initial environmental conditions; whether or not those initial conditions are coherent. Considering the financial state of our current economic system, one would be hard pressed to find a recent, peer reviewed publication that would suggest that it was remotely coherent. This can cause issues when making observations; as a heuristic, influence on attention can obscure important details. The same heuristic influence can cause issues in interpreting observations; as our approximations of reality are formed by the bulk of information that we have collected. This is to be expected in the absence of some form of cognitive trash collection.

http://web.engr.oregonstate.edu/~funkk/Personal/worldview.html

https://en.wikipedia.org/wiki/Observer_effect

https://en.wikipedia.org/wiki/Observer-expectancy_effect

Though having the ability to apply a maximally accurate approximation of reality is dependent upon the aggregation and correlation of a large number of discrete facts about nature, it’s probably not wise to try to develop coherent systems from the ground up by working out the finer details. This is probably the manner in which our current financial system was developed. At the time Coercion based political influence was likely a factor; as that was the common political ideology. This resulted in a financial system that is fundamentally coercive. Though the same could be said of the laws of nature, the behaviors that our financial system coerces do not correlate with that which nature coerces. This suggests that our financial system influences our behavior to be incoherent as well as our worldviews.

http://www.e-m-h.org/Shill98.pdf

The purpose of rigorous research and experimentation is to gain a more concise, general understanding of nature. This is the goal of intention to fine detail. The importance of this understanding should not be overshadowed by the importance of even rigor. With the development of our financial system it is clear that, the outcome of the development correlated with the initial assumptions of the time. The rigidity of the system, the lack of dynamics, has been a high hurdle not only for innovation, but also for reform. This has been the case for approximately 10,000 years.

Here is a wonderful paper on system modeling that includes dynamic systems:

http://www.cds.caltech.edu/~murray/courses/cds101/fa04/caltech/am04_ch2-3oct04.pdf

Note that a coherent general understanding is the basis for which the system is modeled. Though the understanding is derived from rigorous testing of assumptions concerning very discrete aspects of nature, it’s the quality of the understanding that is needed to produce a system that functions in the desired manner. This in opposition to the manner in which the financial system functions speaks volumes about the importance of a coherent approach. It’s said that, “if you begin with the wrong assumptions, you will get the wrong answers”. In modeling systems, it seems that beginning with a  maximally coherent approach maximizes the chances of being able to find solutions when working out the details. Due to the manner in which humans think and behave, top down approaches are inevitable. This suggests that facilitating this fact is just good sense.

Where it is true that the devil is in the details with respect to research and forming hypotheses and theories, it’s not necessarily the case when modeling systems. Though a bit truncated a notion, when modeling systems, the devil is in the approach with respect to it’s level of coherence.

Advertisements

Foundation for naturalized economic administration?

It seems pretty obvious that a naturalized administration should be decentralized. This would solve many problems with current administrative models. Centralized models are often overwhelmed into ineptitude and fall far short of important services. An efficient, decentralized administration could in principle have greater resources for providing the services needed. For instance, a large number of localized chapters could be effective in addressing local conditions, needs and services. They could also help to distribute resources from areas where they are more plentiful; provided that the chapters are well networked and organized.

I’ve been considering a few different P2P models for organizing such a network of chapters; and block chain is keeping my interest. Since a large portion of the administrations tasks would revolve around data analysis of some kind, some form of economizing platform would be required for data entry and dismissal. This has me considering a pool where initial entries can be evaluated for entry into the main database. The initial entries would of course need to be tagged with descriptors that would help to sort and expediently evaluate them. This might mean sorting by location, date entered, last date accessed, number of times accessed, candidate for dismissal, candidate for consensus, and a confidence value. This would of course require other tags for the purpose of statistical analysis upon entry into the main database; that should probably be left to the discretion of the researchers keeping and analyzing the data.

Since the data would have a confidence value, I thought maybe a block chain based sharing system could be appropriate. The confidence value would be a currency of sorts with a target confidence value making the decision as to whether the data would be entered into the main database. A minimum confidence value would of course decide if the data were to be dismissed. The tags for entry date, last date accessed and number of times accessed would be for the purpose of evaluating the data in an efficient manner. It stands to reason that, if the data posed for dismissal were to be dismissed in a maximally short time frame, the storage resources would be being used in a maximally efficient manner. If all of these tags were to be dynamic, issues might be better solved and reconsideration might be better facilitated.

Since the overarching system itself would be Proceduralistic; allowing all to participate and therefor enter data into the pool, coursewares for initial data entry would be a very good idea. Providing an access card upon completion might minimize a wide range of issues. The card might also allow access to the main database; for the purpose of finding solutions to issues; however not necessarily allowing editing. That of course would require the completion of additional coursewares. Maintenance of the database would of course be an Epistecratic subsystem; however this wouldn’t necessarily be an issue if educational resources were freely available.

The network of nodes might be composed of a large number of light clients that can be accessed directly or remotely, from personal devices surrounding a smaller number of more secure, full nodes that can be accessed directly or remotely, from the light clients. I’ve been considering Ethereum specifically because it boasts of being a platform that applications can then be developed for. Because of this it may be versatile enough to construct a more secure P2P application upon.

Our economic system has changed very little considering how long it’s been implemented. Unfortunately it’s been in place for the entirety of recorded human history. This is a large issue as we don’t have sufficient real world testing of any type of naturalized economic solutions. What we do have however is a good understanding of some of our more basic needs and wealth of scientific and engineering data. The best way to approach this issue may be to construct dynamic systems that organize such provisions and use the collective wisdom to solve the issues via self-organization. A natural model itself would be one that would self-organize. Creating a foundation for such a system seems to be the most rational approach.

Game Theory vs Self-Organization

The outcomes of Game Theoretical behavior have been demonstrated to produce effects that fall short of what is generally desired. Many would argue that this suggests that humans are naturally greedy. This however isn’t what Behavioral Science has shown.

We like to think that our life is constructed on a long line of choices; however the evidence concerning it appears to be iffy at best. For instance, consider Dr. Mate’s statement about the power addiction that could be associated with being a board member of a large corporation. The maximization of profit can be considered as a consequence of responsibility to ones stockholders. This however doesn’t mean that the choice to be in that situation isn’t relevant. Taking it even a step farther, the choice to employ a non-corporate model isn’t as likely to produce success; as the corporate model is capable of producing a more favorable price point. Corporations have been winning against small businesses for decades now; because of this. This however doesn’t suggest that the choice to run a business under these conditions is irrelevant. It also cast a shadow of doubt on the notion that it’s actually a choice; since there are so many social and economic pressures to contend with.

It’s common to try to simplify arguments to second order logic in the explanations of them. The rigor that goes into constructing the argument however goes into higher orders. For instance, considering the implications of second order logic produces third order logic an so on. This becomes a more complex task; that is easy to complicate. The social pressures that tend to complicate this process are often political and financial; and are also often based in more impulsive responses. This is where higher orders of logic — or to be more precise, lack there of — can become more of a manipulative influence; as opposed to a more cooperative one. This is not only found in business models but it is also taught in the educational system. This is of course a large problem. The Randian implications here are centered around individualism; which is being shown to be fundamentally fallacious.

What is found in normative behaviors that are associated with selection pressures is cooperation. This is probably the engine of self-organization. Being cooperative with a critical mass of subsystems appears to be what produces fit systems. Part of being self-interested is the subsystem being interested in the health of the system that supports it. Extinction and existential risk are naturally part of the equation.

The lack of third or even fourth order logic concerning the implications of our models could have severe consequences. Financial influence on political policies is obscuring the environmental awareness that is required for an individual to make informed decisions. This isn’t necessarily a danger of centralized decision making; however the initial conditions are promoting this outcome. It’s in essence, maximizing Entropy while promoting the absence of Normalization. This could be considered dangerous as it is an algorithm for extinction or existential risk. For this reason, it appears that the model is fundamentally flawed. Self-interest, by the very laws of nature cannot be both individualistic and an example of fitness in the same instance. Interest must scale to the supporting systems in order to minimize risk. This would be a much more viable model for an economic system.

The operative concept may be that predispositions to behaviors should be accounted for. This doesn’t necessarily mean that they should be manipulated; however it is easy to argue that that is clearly the case now. Rather, employing negative utility concerning that which hinders natural self-organization might bring about the most desired effect. Humans have already been endowed with the ability to be cooperative by millions upon millions of years of selection pressure. I would contend that artifacts of complication may be our most concerning issue. Working toward locating and removing these aspects of the social model, rather than trying fruitlessly to control the environment and having great expectations on human behavior might be the more rational approach.

Rather than trying to figure out how to coordinate, realizing that we are predisposed to coordinating behaviors under very naturalized circumstances might bring about an approach that would in turn produce the desired effects. Much of our attention has become distracted with artifacts that are far from coherent. Our immediate environment is generally of our own construction; and this has seemed to sever our connection to nature to varying degrees. I would contend that this is the root of the more concerning social issues. One can observe the cooperative behavior of animals and humans in more natural settings. This of course doesn’t discount the distribution of pathological behaviors found in all settings; however, it does show a positive sum spectrum with complexity. For instance infanticide, which is common in the animal kingdom is much more acceptable to male lions than male great apes. Great apes have behaviors that can prevent such occurrences; like the promiscuity of females that confuses the males ability to determine which offspring are his. Larger brained species like ourselves have the cognitive ability of weighing the value of emotional bonds against the value of an infant’s life. It may not be practical to try to extinguish all pathological behaviors; as we may not be able to have a concise understanding of all of them.

What has been interesting to me lately is systems for decentralized organization. What’s most interesting to me however, is the doubt that they are indeed decentralized; or that anything really is. The perspective that centralized models are only centralized if they are the product of an artifice seems to be a little truncated when one tries to scale them into perpetuity or infinitude. This at least for my limited understanding, keeps running into a grand centralizing structure; ie the laws of nature. I’m concerned that our awareness of our general lack of control has created a complex that has us neurotically striving for it. The dissonance between our localized perception of being in the driver seat vs the evidence that study has produced is bound to have some psychological effect. This could be the root of the issues that we have faced in building models since the dawn of civilization. This is even present in the dichotomy of free will and consequences in religion.

The disconnection from nature that civilization has produced is probably poisoning the well of our collective perception. That being the case, the most appropriate approach to allowing human nature to flourish with it’s natural predispositions may be as simple as the negative utility of removing the obstacles. It’s my contention that the obstacle is finance and it’s coercive aspects. The evidence throughout recorded history seems to point to it as a common denominator… and I’m not alone.

http://www.e-m-h.org/Shill98.pdf

 

Entropy, Disruption and the AI

fourd

All of the variation, the diversity and everything that is interesting is rooted in entropy. The flawed nature of nature prevents emergence from becoming repetitive. There may be nothing more natural than variation. This however comes with a consequence that is just as interesting. The variation isn’t restricted to only the matter in the universe. Variation over time is the observation. This results in the bulk of emergence becoming extinct. Everything becomes extinct; either through abrupt failure or through success and transcendence. Change is always in the cards for everything.

This makes human loathing of change seem a bit counter intuitive. This is because entropy is only one module of the machine. Normalization is the process by which the successes propagate. This is extremely interesting because normalization is not by a long shot… well… the norm. Like most human endeavors, most natural experiments fail. Abrupt extinction is by far the most common occurrence. Entropy produces more variation than normalization can produce successes. Even as such, success is still subject to variation, transcendence and thus extinction over longer periods of time.

The evolution of the modern human is credited with some form of systemic entropy. The Savanna Hypothesis suggests that human evolution encountered a disruption that accelerated the variation and produced interesting and useful characteristics in our ancestors. This would be one particular instance where entropy was exacerbated and very disruptive; resulting in novelty. Of course normalization played a role; as the characteristics that Natural Selection had endowed our ancestors with influenced the manner in which the novelty was expressed.

The understanding that resulted in the crafting of artifices has proved to be extremely disruptive to the biosphere. The combination of the natural predisposition to normalize and fill a niche’, and the constant change associated with the resulting technological progress seems to have created an inner struggle in modern humans that doesn’t seem to exist in the whole of humanity; but rather is more evident in the developed world. Civilized humans are interesting and unique for many reasons that are affirming and concerning as well. The concerns of extinction risk revolve around the idea that maybe a more general instance of normalization means the abrupt extinction of civilized humans. At this very moment, the implementation of a strategy of “mutually assured destruction” has normalized the most dangerous scenario that one could imagine. This doesn’t just entail the extinction of civilized humans; but also the bulk of life on the biosphere. We have put all of life as we know it at risk as well. We have even narrowly escaped this scenario on one occasion. We forget this; in the illusion of the safety of our brick homes, in the fog of political ideology and deodorant ads. Cognitively, we attend to our immediate surroundings; which is an artifice that divides us from the natural world that bore us. Our perception of reality has thus become an artifice. We have forgotten who we are; and our place in the universe. We have created silly notions like property, justice and inalienable rights; while teetering on the destruction of an entire global ecology that appears to be rare in all that we can survey. This is an inconvenient fact for the optimist.

The cause of these issues is clearly the production of entropy in the absence of normalization. I would like to suggest that the absence of normalization is rooted in the disconnection from nature that our modern conveniences and memes have caused by hijacking our attention. Science has the potential to mitigate some of these effects by directing attention to nature; however natural distributions may not allow for this to become the norm. If attention to science is proportional to scientific endeavors in the immediate surroundings, it would seem that we are currently playing a negative sum game. Opportunities for people to be in nature and thus connect directly can also be a viable influence, however in a world where working in a building and going home to a building with a television and a beverage both marketed as a “reward” leaves much to be desired as well. I can’t help but see this as more exacerbated entropy that would only enable similar types of dangers to that which we have created in the past.

My contention is that this is most likely to end with the extinction of specifically the technologically progressing human, by a combination of the normalization associated with niche’ existence and by the entropy and resulting novelty that is associated with transcendence. I find it difficult to reason out a dichotomy of human and machine when the competitive advantages of merger is so great. I also find it hard to reason out a place for an evolved species in a high tech environment, for the reasons that I have previously expressed. I also find it difficult to buy the notion of humans becoming pets for the AI as the AI is not likely to be tied to the biosphere as legacy humans are and the opportunity to explore and advance might be more appealing than carrying the dead weight of a bunch of freeloaders. This of course requires some qualification. When humans began domesticating dogs, they had a practical purpose. The superior senses of the dog helped people in hunting and protecting the group. It was symbiosis. What practical purpose can a human provide for super intelligence?

Work as we know it is only about 100 years old. Before there was essentially only business and subsistence. I would suggest that is likely to make a comeback. This of course requires some qualification as well. There has been an ongoing trend to refine production methods with technology and pass the technology to the general public. This is happening with technologies like 3D printers, for instance. This movement toward automation would only simplify this process. A combination of availability and necessity seems evident in the coming decades as the technologies are in the works and the crises are as well. There has also been a long running trend of decentralization since the time of Adam Smith, the birth of the New World and the enlightenment. There doesn’t seem to be a defeater to this trend. It seems that the end of the era of the 5 / 40 job may just return us to a state of relative self sufficiency like we previously had lived under. I’m speaking of legacy humans of course; though I can’t imagine the remaining humans numbering as much as a billion.

The notion that the AI would take the proverbial crown from humans is based on the fallacy that we wear it now. If there is anything that we are not, it is in control. We are on the verge of “causing” our own extinction every day, with every “decision”. We are not the drivers. We are driven. We are the product of and a vehicle for natural law… and the AI will be too.