Additional Notes
For the People by the People of the World
" Jonathan Robert Banks 2000 jrbanks@1earth.net

21. Computer scientist Danny Hillis with the help of Biologist Eric Lander, "developed an equation that quantified the boundary where the punctuated equilibrium event occurred. The equation is one over e squared (e is a well, established mathematical constant). When the level of the specific genes reached this value, there would be a phase transition and the population would make a dramatic leap to a higher level If we think of civilisation as a whole as Ray's or Hillis's population of a-life entities and we replace genes with memes and enabling technologies, then surely we are close to some massive punctuated equilibrium! Levy described how during periods of stability in Hillis's simulation, analysis revealed 'seething' activity of genes, which were establishing the foundations for the next leap. The world today is characterised by 'seething' activity in philosophies, worldviews, religious and metaphysical ideas, scientific breakthroughs in every field and technological advances from mainstream to alternative." All these new ideas, paradigms, discoveries and inventions are analogous to epistatic genes. "Instead of genetic changes building undetected towards the critical mass, (described by the one over e squared formula) and creating an evolutionary leap, we have memes and enabling technologies leading to the same thing. In a sudden and spectacular leap of evolutionary change­­a total change of our society our civilisation and environment could occur similar to Hillis's and Ray's simulations where punctuated equilibrium emerged." A New Reality p247

22. "Langton developed the Lambda (l) scale from 0 to 1 which represented the degree of information retention and movement in a system. At the zero end of the scale information is fixed or frozen. Further along it is periodic, moving in endlessly repeating cycles. At the opposite end (one) it is totally chaotic. Langton discovered a 'sweet spot' on this scale where complexity was at its highest and where living systems and artificial life thrived! He called this area the 'edge of chaos.' It is located between the periodic and chaos areas. It is a place where information movement has the right balance of stability and flexibility. It is an abstract window or arena in the information universe, a balance point where complex adaptive systems are at their best. When an a-life environment is tuned to the edge of chaos the most novel, interesting and life like behaviours emerge. Wolfram discovered the same concept studying cellular automata and divided the areas into four classes. Class 4 related to the edge of chaos, at 0.273 on the Lambda scale." A New Reality p110

24. Swarm Simulation Projects

There are a range of broad-based areas of simulation that are run by the Swarm system. Within these subject areas, there are many individual program simulations applicable to the AI for the people system. These include:
Ecology:
¨ Gecko - Gecko is an individual-based simulator for modeling ecosystem dynamics being developed at the Center for Computational Ecology at the Yale Institute for Biospheric Studies. It is being developed by Ginger Booth.
¨ Gerard Weisbuch of Ecole Normale Supérieure is using Swarm to study adaptive agents in interaction with their physical or biological environments, exploiting renewable resources such as fisheries or polluting their environment and in interaction with other agents in markets
¨ Claudia Pahl-Wostl's group at the Swiss Federal Institute of Environmental Science and Technology, Duebendorf (EAWAG) is studying ecological and socio-economic networks. A current Swarm model is being used to generate ecological networks, which will allow the investigation of the effects of a network's structural organization and of the properties of the individual network elements on system performance.

Computer Science/industry
¨ Drone - The CAR Group at the University of Michigan Program for the Study of Complex Systems has developed a tool called "Drone" that can be used with Swarm or with other simulation packages to do multiple runs of a simulation while varying the inputs automatically. The CAR Group consists of Michael Cohen, Robert Axelrod, and Rick Riolo.
¨
Economics
¨ Benedikt Stefansson is interested in Computational Economics and is specifically using Swarm to study the dynamics of competition in a differentiated product market and an evolutionary model of Principal-Agent organizations.
¨ Sponsored by the Santa Fe Institute, Brandon Weber has written extensive documentation for the two Artificial Stock Market implementations: Swarm and straight Objective C. The documentation is available either as PostScript, converted HTML or in the original Microsoft Word format
¨ The Swarm Web Interface for Experimental Economics (SWIEE) is a project that aims to develop software classes and examples for facilitating experiments with human subjects.
¨ Microeconomics modeling - oligopolistic competition models.
¨ Modeling effects of customer satisfaction on profits over various time-scales. (http://www.csua.berkeley.edu/~cody
¨ Agent based models in economics, also with neural networks
¨ economic-simulation CA (http://venus.unive.it/~alex
¨ Economic simulations especially Adaptive organizations, network transactions (http://www.geocities.com/Eureka/9845
Political Science
¨ Darren Schreiber has implemented an agent-based model of the formation of political parties. The program unifies four traditional results in political science as emergent consequences of the model: 1) the tendency towards two parties, 2) movement towards the median voter, 3) party realignment with a change in issue salience, and 4) the tendency towards a minimum winning coalition. He is also working with the Empirical Research Group at UCLA Law School on an update of Thomas Schelling's classic model of racial segregation.
¨ Developing Sugerscape into more test environments (not only trade also, war, different types of governments (read: communism etc)) (http://swarm.santafe.edu/survey99/www.wi.leidenuniv.nl/~aal
¨ A Construction of agent based economics models. Learning and induction (helios.unive.it/~fluna/english/luna.html)

Geography
¨ Catherine Dibble is using Swarm model spatial technologies and human settlement patterns and the associated economic, social, and environmental implications of likely changes. She has two relevant papers on the web: Theory in a Complex World: Agent-Based Simulations of Geographic Systems and Geographic Modeling with Computational Laboratories.

Social science
¨ The role of Emotions in Social Simulation (http://www.daimi.au.dk/~werk/
Research into cooperation in adaptive multi-agent systems, using genetic algorithms (http://www.mk.dmu.ac.uk/~jmarshall
¨ I am trying to compare a Vensim model of attitude change with a Swarm model to contrast insights from the two approaches. Also, I plan to use a Sugarscape-type model to investigate gender and class structures in society.
¨ Will be looking at possible cultural vectors of psychological disorders across time
¨ Using physiologically realistic mathematical model neurons, placed within the swarm environment, to examine local circuit dynamics involved in thalamic pain processing.
¨ Developing Social Science models, and a modeling environment to build such models. (http://www.syslab.ceu.hu/~gulya/
¨ Settlement-Pattern modeling of southwestern prehistoric land use
¨ Implemented a version of SugarScape Currently working on Immune Systems for pattern recognition
¨ Modelling the effects of anxiety in human relationship systems, looking for network effects unpredictable from knowing just the individual anxiety operating rules.
¨ I have used Swarm extensively in my Masters research at the University of Minnesota. I am building a reusable library of Swarm objects to be used in building Behavioral Evolution simulations. (http://www.cs.umn.edu/~krumpus
¨ a simulation about formation of coalitions.
¨ Fabrice Chantemargue is a member of the Autonomy Modelisation and Coordination (AMoC) project at the University of Fribourg that has implemented a model of Implicit cooperation and Antagonism in Multi-Agent Systems in Swarm. He has an enhanced application that uses the Vision library to help model this antagonism.

¨ 26. Swarm, a general purpose simulation package, provides utilities for designing, implementing, running and analysing multi-agent systems: systems where hundreds or thousands of singular agents interact in a dynamically changing environment. Trumped as "The best thing to emerge from the Santa Fe Institute" by Business Week Magazine, and developed with Chris Langton's direction, it provides a platform for implementing models of systems otherwise difficult to specify, or that are dynamical, hierarchical or non-homogenous. This is due to Swarm's extensibility and composability. Hence its subsequent rise to become the standard in academia and science for modeling complex adaptive systems. The user can produce models of a complex nature that look into potential future configurations of systems usually managed through heuristics or intuition. It can also be used as a tool for custom-designed, multi-agent simulations. It is maintained and supported by the Swarm project at S.F.I, www.swarm.org/intro the package freely available and licensed under the Library GNU Public License (LGPL). See the Journal of Artificial Societies and Social Simulation www.jasss.soc.surrey.ac.uk/JASSS.html for agent based simulation. See Journal of Applied Systems Studies www.unipi.gr/jass/ for general systems approaches. For those who have little to no understanding of complexity, artificial life and agent based simulation I suggest the S.F.I WWW.santafe.edu as the best place to start.

27. AI for the people system

AI for the people is an integrated system, based primarily on a bottom-up evolutionary approach with unnatural selection. All other tasks, such as tracking the selection and evolution of ideas and people's involvement with ideas, managing the data bank, awarding points, allocating shares and all other management requirements of the system that could be automated could be handled by specifically programmed AI software. It is this entire system that I refer to as AI and not a top-down, representational based AI. We could use anything that makes the system, smarter and more useful.

The original idea for the system was spawned (in general) by artificial life and complexity, but specifically by a classifier system. Since collating the original ideas and arguing that agent based simulations would have advanced sufficiently since the early nineties to make this possible, I have come across SWARM and the evidence for a proliferation of agent based simulations and their growing acceptance and success.

Many scientists would be reluctant to openly suggest we could base decisions on such simple models, derived from such a new science, as no doubt they would be howled down by their colleagues. I argue we have no choice and so we must balance necessity with established protocol, to fit our current reality.

Using agent based simulations we can test ideas, strategies and technologies within a selection criteria designed to drive a simulation towards the sort of societies we always wanted but could never achieve. We could utilise the cold hard dynamics of survival of the fittest in the simulations and by implementing the results in the real world we can take these dynamics out of the real world, thus eliminating the real pain and suffering it creates, to produce a more gentle and nurturing society. We will have a superior selection criteria in the real world, along with a greater capacity to shape our society into one that supports universal ideals.

Many scientists and others believe the Internet will ensure the freedom of the people and the advancing of the free spirit of creativity and knowledge in the egalitarian pursuit of a better world. If this is the case then the smarter the Internet the more powerfully it can serve this purpose. Instead of waiting for the Internet to get smarter, why not create a purpose-built smart website that operates like a super-smart, advanced Internet now.

The AI for the people system will represent a microcosm of a future advanced global Internet system, that is in itself an artificial mind and smart, data bank, ideas generation, system. The system could be seen as a piece of specialised, smart software running on the substrate of the Internet, any improvements in the capacity of the substrate will automatically improve the functioning of the software running on it. Eventually an advanced Internet will do what the people system does and the people system will evolve into a highly specialised node in the global net.


NEWS

On 16th Jan 2000 during a private meeting with a professor at the University of Wollongong, an expert in AI who claimed in an unofficial capacity that the model I had put forward was possible to create and represented a fascinating combination and application of AI processes. He is available for discussion via phone and/or email.

I plan to establish a non profit structure to run the system. I have also developed a small team here and in India for this project.

At this stage I need advice as to what resources are available from corporations, venture capitalists, entrepreneurs, government departments etc to secure seed capital to develop a prototype. Also anyone with relevant expertise or financial resources for the project. If you would like to help then please get in touch.

System potential
¨ be a focal point to distill the essential features of a global vision of the future from the people of the world who use it as a guide to choose preferred futures to move forward with a minimum of chaos.

¨ Act as a global educator to spread the word of complexity and agent-based simulation and of the nature of our current evolution.

¨ Act as a global focal point for simulating every aspect of our current reality and our possible/preferred futures. The system will also be used as the peoples simulator. Using agent-based simulation the people could be involved in simulating all aspects of society and their own lives, to make better individual and collective decision to guide us through the difficult transitions ahead.

¨ Become a new form of people entity in the global sociopolitical/economic ecology that could precisely influence the dominant species, ie, multinationals and national governments and therefore change the global socio-economic ecology

¨ Be a bridge towards global democracy acting as a global lobby platform to influence and support political parties to restructure towards national internet/referendum democracy and then to network, co-operate with other national governments for internet/referendum democracy.

¨ Be a bridge towards the global implementation and distribution of nanotechnology for the people.

¨ Generate ideas, strategies, technologies, products and services.. Manufacture, market, distribute and sell products and services via internal networking.

¨ Produce intellectual property that is owned by the people of the world. Everyone involved in the development and/or realisation of the idea would share in the profit, which is systematically distributed based on standardised agreements.

¨ Initially popularise by direct appeal and advocacy by the more aware, famous, performing artists of the world­­actors, musicians and singers who cut across boundaries of race, religion, political affiliation, age, sex, socio-economic backgrounds­­more than any other group.

¨ A global focal point for collaborative creativity between artists scientists, enthusiasts, philosophers and futurists under the direction of film makers, using the technologies from computer science, complexity and a-life to create awe inspiring and wonderfully captivating graphics and special effects in the production of block buster movies, short films, documentaries and even commercials­­to effectively use this powerful, populist, entertainment medium as a way to realistically demonstrate possibilities, focus attention on relevant issues and inspire the people to participate.

Systems functioning.
1. User goes to AI for the People website and logs on ( they would have to have joined and received a member number.).
2. They proceed to the ideas/strategies input page where they select the appropriate category and input their idea/strategy into the field, they are given a registration number for their input.
3. The system will now track the progress of that persons idea through the system and keep the owner informed.
4. The idea is matched and represented by an agent and connected to a simulation.
5. The idea is then put through a general performance test and is rated. It is compared to similar ideas and ranked against them.
6. The highest ranking ideas in any class proceed to a specific performance test where they compete for the highest score.
7. The winner is posted on the systems site and sent to people and organisations who's activities/resources match the idea.
8. The idea could be picked up by an individual/organisation who apply to be able to develop, manufacture, distribute and/or utilise the idea/strategy. All individuals and organisations in this data bank have had to apply to be part of the system and pass a strict set of requirements in relationship to their philosophy, business structure and practices etc.
9. If the idea is utilised the parties do so bound by a standard agreement based on percentages to keep margins small and fair and retail price low. The payoff is large volumes of sales without any marketing and advertising costs to the retailers as it is done in-house via the system.
10. If the idea is used, sold etc and makes money the profits are distributed according to the agreement which would include all people/organisations who have contributed to the idea and/or its development, manufacture or distribution etc.
11. If the idea is not developed/used by people it could be fed back into the system for refinement and design of technology. This level is done in partnership with people in the system.
12. At each stage of this refinement/design stage progress is posted on the sight for the input of anyone who can assist with this process.
13. When this stage is complete it goes to stage 7.
14. People who want to buy products or services log onto the site and search for the things they are interested and will get information on the closest place to them to obtain the products. They select someone to obtain the product from and purchase online through the system and receive their product digitally or by mail or they may obtain it in person, if possible. Information about new products and services that match members needs could be sent to them on a regular basis.

System mechanisms

Until further discussions with experts and perhaps not until a prototype is being developed, both of which I have not had the time nor resources to pursue, will we know if we can automate each step fully.
If so:
¨ How effective will it be?
If not:
¨ How much people input will we need and how easy will that be?
¨ How long will we have to do it for until it is automated?.

Ideas categorisation. Ideas are assigned to specific topics and classes for example health, economics, environment etc, by input fields. Fields are designed to isolate essential elements of the idea to be represented as an agent and/or agent behaviour in an appropriate simulation.
Ideas representation. Based on the fields categorisation ideas are correlated to a library/data bank of agents and an appropriate agent is selected.
performance test­­general. The simulations for each topic/class are arranged in a hierarchy from very basic/general to more and more specific or demanding, something like heats, semi finals and finals. Ideas are awarded a performance score in each simulation and the best performers go onto the next level while the rest are knocked out. This level is equivalent to heats.
performance test­­specific. This level represents semi finals and the final. It can also simulate the highest scoring ideas individually to test their ability to drive the ecology towards desired futures.
Ideas refinement/design. Use agent and design AI to aid in the re-design/refinement and/or design of technologies and/or strategies based on winning ideas.
Futures simulation. This is a separate area of simulation, that works on bringing together the best of agent based simulations of every aspect of our society, nations, multi-nationals, organisations, bureaucracies, markets, communities, individuals, to brains, CNS and psycho-behavioural-physiology of individuals and processes/technologies for R & D to simulate general possible future paths our civilisation could take under different conditions. This is the people's simulation and is used to educate the people and test out the ability of all new ideas/strategies to contribute to desired futures. This would also involve simulating the effects in our society of the entire system as it develops and grows.
System simulation. In parallel to the above, the entire system is simulated as a process of real-time quality control. This is used to test new technologies and processes for the system itself.
Scientific research. The system could run a separate area for pure and applied scientific research that is open to the input of scientists and enthusiasts. Winning ideas from this area can be fed back into the main system for manufacture/development/implementation/distribution/sales etc.

General performance test mechanism

The simulation would give the idea a general performance rating. For example, in the case of an idea/process/technology to improve the health/wellbeing/peak performance of people, the first level of simulations would establish if the idea demonstrated any positive benefit indicated by a fitness score.
1. The ideas being simulated could include prayer, meditation, diet, exercise, drugs, neurotechnologies.
2. The ideas would have to be broken down into a hierarchy that could be used to represent their scope and impact on an agent in their functioning in the world. This could then be represented in the agent by changing parametres that represent the agents ability to function, which would be influenced by the process, for example, the agents psychophysiological system that underlies its entire behaviour/performance.
3. For example: antidepressants given to all the agents may generate a score of 30 out of 100 (the score would be a composite of agent and ecology performance scores, the simulating of the drug would have to take into account the positive and negative effects on people, its effective range of use, long term strengths weaknesses etc), neurotechnology scores 80, prayer scores 40 and some other idea scores 10. If the cut off mark was 30 then the 'other' idea would fail this level while the rest move onto more stringent and specific simulation/testing.

Specific performance test mechanism

Stage 1. Ideas advocating process/technologies and/or strategies represented as agents are placed in identical simulated ecologies representing a human psychophysiological system and are allowed to run unaided. The ecology is then assessed as to its overall fitness level and given a rating. The results of each idea are compared to create a winner in the process/strategy/technology that gives the ecology the highest fitness score and closest resemblance to a description of the desired traits of a peak performing individual.
Stage 2. The highest scoring agents are place in a simulation of society and their overall performance is rated and the winner identified.
Stage 3. The winner is duplicated and used to populate a simulation and allow it to evolve and rate the ecology's overall fitness. The ecology is perturbed in various ways and the performance of its agents as well as the overall performance of the ecology is rated in its ability to deal with the perturbations. This could be compared to an ecology of agents that simulate the current average fitness of individuals..
Alternative stage 3. Winner agents and normal agents could be placed in the same ecology and rate and compare the agent's and ecology's performance, before, during and after perturbations
Stage 4. An ecology is created to simulate the distribution of the technology and its effect on the ecology.

This basic process could be used for every other area of human activity e.g., economics, politics, social policy etc.

Ideas refinement/design mechanism.

1. Ideas which survive the specific performance test could go straight onto the design stage and/or be posted onto the Internet for a final vote by the people.
2. Genetic algorithms have been used to design circuit boards. Design AI software has been used to design many things. This stage could involve the combination of these two approaches in a feedback loop with the people, who could interact in the process over the Internet, voting on refinements/designs and/or directing the refinements/designs.

Futures simulation mechanism

1. Create a questionnaire for the people that extracts the elements of a shared vision for our future. We go for the essence of what's important, substantial and endearing to human life. This is voted on and the common themes are distilled. This distillation is used as the essence of our preferred future. We won't know the details and the processes that will create and sustain it however we can create possibilities with our simulations. Therefore all simulations that lead to futures that can incorporate the distilled universal qualities are futures that would be preferred and would win the vote from the people. These preferred futures would be used as the final selection criteria for the ideas, technology and strategies that ultimately have to support this distillation and thus the preferred futures.
¨ Alternatively the vision could be built up if ideas were treated as nodes in an ANN and the relationship between ideas were the connections. As more people contribute ideas that are similar the nodes and connections represented by those ideas would strengthen whilst others not being reinforced would weaken. Eventually this would build an ANN that represented a collective vision that could be continually added to. Perhaps this ANN could act like an AI itself and eventually be able to do the ideas selection and/or it could be used as a dynamic selection criteria for the simulations of possible futures and/or ideas/strategies. Feedback between futures selected by the people and ideas/strategies that survive could help to strengthen the connections involved in their selection.
2. Develop general simulations of our global civilisation and the possible paths it could take over the next fifty years creating negative and positive outcomes. This would entail simulating sub-systems of our civilisation e.g. nations, multinationals, bureaucracies, communities and the ecosphere. This could be dealt with by SWARM systems using any combination of agent construction, controlled by specialised AI software. You could simulate human systems using real people represented as agents or people interacting with agents in a simulation. Simulate an economy, market, business etc. with the normal low level cooperation ,i.e., people being selfish, etc then run some simulations with the people making the decisions out of higher order reasons, and demonstrate the difference. This maybe useful for people to have a hands on experience of how things can be different. See SWIEE http://swiee.econ.unito.it/
3. Post the results on the Internet for the people to vote on the preferred futures.

Systems simulation mechanism

¨ Build a duplicate system and use this to test new ways of running the systems. Test different types and structured agents and AI software and different combinations of agents, human interaction and AI software to improve the overall system and monitor quality control.
¨ Use this to evolve agents, classifier systems ANNs, etc, to do the work of representing an idea as an agent and creating simulations. Use ideas that had already been through the real system so that the systems being tested have appropriate reward criteria. For example a classifier system where input from its environment are the elements of categorised ideas and its rules competed to select agents to represent them. It could get its feedback from the results of simulating the ideas, (they are compared to the actual results of the idea that had already been successfully categorised, represented as an agent and simulated in the real system), which would correspond to the classifier actuating a behaviour in its environment. The GA could then breed better rule sets from the ones that got good scores from the matches they made.
¨ Randomly select ideas that have made it all the way to the design stage and run them back through the system and map their path of selection and simulation against the original path to test for any erroneous decision combination that may arise.
¨ Run the same ideas through but this time tweak the system in various ways to look for possible dangers and/or improvements. Professor Paul A. Fishwick from the University of Florida has developed a program to represent software as 3-D models using VRML to create interactive virtual worlds that can simulate real phenomena, which maybe appropriate for the system and some of the other simulations (see. http://www.cise.ufl.edu/~fishwick )
¨ Simulate the growth of the system and its effects in the real world.
¨ Simulate the spread and effects of important technologies and strategies developed and distributed by the system.

Simulating the usefulness of high fitness levels of individuals to our society.
1. Create simulation of human organisation and /or society.
2. Have a fitness scale for agents and the ecology as a whole.
3. Create agents where you can turn down their overall operating capacity. Make some glitched in various ways to represent people with psychophysiological problems.
4. Run the simulation. Note overall fitness/performance scale and perturb the ecology in various ways to graph how the ecology and agents perform.
5. Increase their fitness levels and measure overall ecology fitness, perturb ecology and measure its performance. Compare both simulations.
Then have established increasing fitness levels of ecology make a more stable.
6. Now go on to simulate different types of processes that increase fundamental fitness levels of agents e.g.: drugs, diet, exercise, neurotechnology, meditation, prayer etc. Rate the effectiveness of the agents in terms of their overall performance levels. Create a rating that examines the range of benefits and the broadness of effects in the agents' activities. Processes/technologies could be simulated and give a final effectiveness rating.
7. To do this we would create specialised simulations of the psychphysiological system of a person and hope that system determines cognitive, emotional and behavioural performance. This would be based on neuroscience, cognition and their applications in mental health and peak performance.
8. Also we would need to create a simulation of the process/technology. This may simply be a list of parametres that are used to tweak the performance function of the agents. For example, in simulating neurotraining we would need to be able to tweak attention and arousal in the CNS, cognition, emotional states, etc, or perhaps we simply simulate documented outcomes e.g.; cognitive, emotional, physiological and behavioural.

Ideas and Implications

Brainstorming the endearing essence of what we want from life and society and what sort of future we want is important to create a shared agreement for the future. If we study religious, philosophical and wisdom literature of every culture that has ever existed we could distill this same essence. Human beings have always dreamed of truly civilised, truly human societies. The common themes include; the ability to love and to be loved, to be free from exploitation, violence and repression, to be respected by others, to feel a sense of belonging and worth in the community, to be able to do meaningful work in life, to be recognised and rewarded for our efforts, to be able to dream and plan and work towards the future, to have quality time to enjoy ourselves with friends and family, to make life easier, and to be happy. The Dalai Lama has written a simple book on the concept of universal aspirations, which he argues are independent of religions, philosophies and cultures and constitute a pragmatic definition of spirituality.

This exercise could be conducted all over the world using the ANN model in the AI for the people system. From this individuals and/or groups can create rich pictures and/or interpretations/scenarios of what this would look like by telling stories, writing songs, painting pictures, creating visions, inventing technologies or developing structures, strategies and socio-political-economic policies.

We will not know exactly how the future will unfold or what it will look like, however we can create possibilities based on the essence of what we want. This is very important as Robert Fritz argued, we must be crystal clear about the essence of our creation/vision or what we want. It may manifest in forms we hadn't thought of but as long as we can recognise the essence we will not be fooled by unimagined/unbelievable forms it may appear in. With agent based simulation we now have the basic tools to begin to develop simulations, which can give us possible futures as well as paths to those futures, based on selected technologies and strategies.

This is a tremendous step forward considering our present operational mode. Within the paradigm of a free market economy, we thought the free forces of the market acting as our selection criteria would produce the best possible result for humanity, leading to the sort of conditions that support the endearing ideals people want. For example commentators predicted the computer would make our work more efficient thus giving us more leisure time so people would be relaxing and enjoying themselves but this didn't happen­­the work load increased, people are working longer and with less leisure time. This situation has and will continue to arise as our free market economy is a type of cold, selfish, blind evolution. The selection criteria in the ecology of our socio-political-economic environment are often mutually exclusive to the idealistic outcomes they are supposed to create. So no matter what new technologies we create, or socio-economic policies we implement, they will be skewed towards outcomes different than we envisioned. If the selection criteria, which shapes the evolution of the ecology in a definite direction is based on cold, hard survival-of-the-fittest, then it is natural they can't support the ideals we believe we are working towards. Look around and you see all the problems­­inequities, injustice, corruption, unhappiness, technologies and policies not delivering. What you see is the emergent outcome of the underlying selection criteria of an ecology, which are blindly pushing us in the opposite direction most hoped/believed we were going.

It is imperative to create agent based simulations that model all levels of our global civilisation and possible futures, to use these simulations as the more intelligent and humanistic arena, for utilising the cold, hard dynamics of survival of the fittest, to spare real human beings the pain, struggle and loss that occurs in the real world. Let's face it, we live in a real-time simulation, where we the people are agents manipulated by the leaders of governments, bureaucracies, religions, multi-nationals propagating memes like dogmatic, economic rationalism, etc, in one big experiment where no one really knows what is happening, what will happen in the future and often in blatant denial of the evidence that the situation is not what it was promised to be. What is the definition of insanity? Keep doing the same thing and expect different results.

People have become resolved to the animalistic and cruel reality of our so-called civilisation. They say "it's-a-cruel-world" ,"it's-a-jungle-out-there", "its-dog-eat-dog!". So what have we achieved? We set out to tame nature and civilise ourselves and the world. To protect ourselves from the brutal natural world and make life more elegant and civilised, which seems an utterly natural thing to do for intelligent beings. However we have merely created the same brutal order in civilisation, even pretending it is the most effective way to ensure our society becomes more civilised, thus bringing us our ideals. Our society has become cruel and destructive, where many people in the most developed nations of the world are becoming more disillusioned and unhappy­­the World Health Organisation predicting that within a decade depression will be the number one disease of developed nations. The blind, cold, free market economy with its survival of the fittest mentality is obviously taking its toll.

With realistic agent based simulation we begin with our ideals, our dreams of the future, the essence of what is important to us. We use this as selection criteria, along with quite specific examples of what we want. This way our simulations are always driven towards the structures and futures that support what we want. We can fill them with ideas, technologies and strategies represented as agents and let the fittest survive. We know if they survive they must support what we want, and so will lead to the structures and futures we want. In this way we let the agents endure the cold hard unforgiving dynamics of survival of the fittest with the selection criteria we create that supports what we want. We take the simulation results and implement these ideas, technologies and strategies into the real world, thus dramatically increasing our ability to direct our evolution, to create what really matters to us. A more advanced and sophisticated form of directed evolution would evolve, compared to the degree of direction we command now. We could actually change the selection criteria in the real world because we have done this in the artificial world and modeled the outcomes.

This saves real people and nations from the very real pain, suffering, wastage and destruction that occurs in our society. We take the cold hard survival of the fittest out of our real world and use it in the artificial worlds that mimic our own, making the real world a more gentle, nurturing, healthy environment for human beings. Surely this is a sane, civilised, humane, intelligent and practical way of using this dynamic. Perhaps this would create a more spiritual, aesthetic feel to our society, which many futurists such as William E. Halal predict. This is absolutely essential in a situation where the dynamics of the system show us heading for a series of bifurcation's.

With a vision of a shared preferred future, represented by supportive, civilised selection criteria, as well as the implementation of technologies and strategies which have survived the "dog-eat-dog" world of our artificial ecologies, we can restructure our society to guide it into the future with more control than ever before.