(The Orrery)

SIMTIME (c) copyright 1992 Anselm Hook All Rights Reserved
THE ORRERY (c) copyright 1992 Anselm Hook All Rights Reserved
Project #1317
Revision History,
1/15/1992…Memes
12/17/1992…Initial Layout
Concept,
Imagine if you could create your own country.
You’d want it to be a good place, a place without prejudice, with fair and just laws. You would want to be able to design the laws and cultural behaviors which make people act the way they do.
SimTime is just such a place, an excuse to wander over all of human past, to go back and forth through time, changing events and history to "try and get it right!".
1. SIMTIME
2.1. Introduction
*2.2. SIMTIME is a simulation
*2.3. Time Lines
*2.3.1. SIMTIME Earth time-line.
*2.3.2. SIMTIME Solo Play time-line
*2.3.3. SIMTIME Speculative time-lines.
*2.3.4. SIMTIME Fantasy time-lines.
*2.4. Themes
*2.4.1. Social
*2.4.2. Technology
*2.4.3. Environmental
*2.5. Feasibility
*3.1. Rule Based Simulation
*3.2. Programming Limits
*3.3. Agents
*3.4. Time Simulation
*3.5. Fractal based math
*3.6. Reversability
*3.7. Controlability
*4.1. General Play: Open Ended
*4.2. Solo Play Scenario: Catastrophe!
*4.3. Multiplayer Competitive
*4.4. Solo or Multiplayer Voluntary Constraint Scenarios
*4.5. Solo Player Game Scenario: Alien Escape
*4.6. Multi Player Game Scenario: Aliens Escape
*5.1. Topographic Map
*5.2. First person perspective
*5.3. Cross Time Map
*5.4. Composition and Editing tools
*5.4.1. Terrain
*5.4.2. Celestial phenomena
*5.4.3. Rule Composition
*5.4.4. Actor Composition
*5.4.5. Memes Composition
*5.4.6. Inventions and artifacts
*7.1. Exploratory at multiple tiers
*7.1.1. Tier 1:
*7.1.2. Tier 2:
*7.1.3. Tier 3:
*7.1.4. Tier 4:
*7.2. A Testing Ground
*7.3. Power
*7.4. SIMCITY
*7.5. More thoughts and questions
*9.1. Spirit
*9.2. Systems and Institutions
*9.3. Mechanics:
*9.4. Comments on Mechanics:
*9.5. Thoughts about memes
*10.1. Stimuli handling:
*10.1.1. Stimuli Sources:
*10.1.2. Doing Stuff:
*10.2. More thoughts about memes
*10.3. People Editing
*10.4. Life Forms other than humans
*10.5. Culture and People revisited
*10.6. Knowledge
*10.7. Goals Oriented Behavior
*10.8. FAMILYS
*10.9. MEMORY
*10.10. BELIEF CHANGES
*10.11. SESSILE ACTORS
*10.12. THE MUSE
*13.1. GENERAL DISPLAY
*13.2. TOP VIEW AT A DISTANCE
*13.3. TOP VIEW ZOOMED
*13.4. PERSON EDITOR
*13.5. CULTURE EDITOR
*ECOLOGY OVERLAY MAKER
*CROSS TIME MAP
*13.8. INVENTION MAP
*13.9. TRIVIAL EXAMINERS
*
Design
Introduction
Make history the way you’ve always wanted it;
Stamp out Evil or Setup a New World Order.
It’s your choice in SIMTIME.
SIMTIME is a simulation of history. It lets the participant explore and manipulate human decision making on a grand scale. The participant can zoom through time, altering human decisions at critical historical junctures and then rush back to the future and witness the cascading reactions which then happened from that point in history onwards.
Like SIM-CITY you manipulate society on a grand scale, altering events at a variety of levels, from the propagation of a certain thought to the occurrence of a natural disaster which perhaps changed history.
As humans we represent the end-product of hundreds of thousands of years of social, political and economic interaction. Much of what we are and have always been is a social animal, our interactions have always been concerned with each other, not just the physical world. We live and cooperate by a system of decision making based on a history of moral and ethical exploration. This would be more visible if we could interactively simulate this history; to recreate a sense of the deep swellings of political thought which characterize our complex socioeconomic tapestry.
We might see that we are the way we are because we have been filtered through a kind of giant four dimensional sieve. All the emergent complexity we talk about, the way we act today as a result of the events in our history are, when viewed from a higher order of dimensions, simply a pattern upon a wall. And even then only one possible pattern out of a wide variety. History is neither so straightforward nor so fickle as we would imagine it. The events in our history are arbitrary, but the successful events, those decisions which survived the test of time, are harder to peturb, and radical successful solutions to the problem of living are not as easy as they might seem. Our history is arbitrary, but the results of it are not.
SIMTIME and products like it mark the beginning of a period where the interactive experience will become the dominant for of media, education and entertainment in western society. The "experiential" revolution is just beginning. We are among the last merely literate generation, the last TV generation, and among the first to see the potential for experiential knowledge as a way to develop a rapport and feel for the real world. SIMTIME aspires to help us understand why things are the way they are, to work to bring peace to some of our thoughts, and possibly, like some of the essays, poetry, thinking, and literature from time past, to even bring about some much needed real world change. Being able to for the first time purely and simply simulate-out the ‘what-ifs’ of questions moral and physical that we’ve all being presented with may lead to some startling insights and real world change.
Only the past can be changed,
The future is unknown and therefore fixed.
The behavior of individuals and groups is mechanically simulated such that a history evolves algorithmically. Game Actors occupying the stage of SIMTIME react in a self-reasoned way to outside stimuli. It is up to the participant to decide what that stimuli is going to be. The participant can "push" on events changing their outcome - and the resultant outcome of all history from that point.
The simulation is intended to be accurate enough to be satisfying, to capture enough details about the human mind and human interactions that it is worth playing. Whether or not game actors have a tendency to group together or splinter apart depends largely on the circumstances of their lives. Their needs, beliefs and environment will largely dictate their behavior.
For the participant, the thought of being able to alter a history - a story which for all intents and purposes could be their own - is highly entrancing. Controlling the destiny of entire cultures and civilizations by the well placed drop of a pebble has the most appealing sensation of power, apprehension and comprehension. Simulation resonates with the mind, we map experiential knowledge much better than ‘passive’ or literal knowledge.
Human belief systems in SIMTIME are implemented using the idea of ‘memes’ as promoted by Douglas Hofsteder. A meme is an idea which replicates like a virus, usually providing an action or benefit that is tightly bound to its replication needs. Humans allow themselves to be infected by powerful idea agents which use humans as a substrate for their propagation, and as humans we seek out the most powerful most successful ideas to propel ourselves through our lives, to carry out the agenda of other kinds of internalized agents. Memes provide leverage points, coupling us loosely to what appear to be points of immobility in the universe, and like a swimmer we can grab these points and push off against them. For example, one of the reasons humans gather in large groups is it provides fertile grounds for new memes, as a propagation strategy many memes tend to encourage grouping.
At some point there is a difference between a gang of thugs and a government. 200 people is supposedly minimum breeding stock for the survival of the human gene, and 150 people is supposedly the ideal size of a community and a number that our minds are attuned to working with.
Our simulated people may be few in numbers but there will be a disproportionate presence of the muse, the spark or fire which ignites the mind and body and compels it to action. Each actor will be the equivalent of a Plato, a Malcom X or a Kathy Acker, each raucous, loud and influential. Our game is a stage which dramatizes and exaggerates actions and reactions in an ever complexifying space.
SIMTIME shines as a sheer exploration of time.
Not only time past but time altered,
time never visited, and time yet to come.
SIMTIME Earth time-line.
SIMTIME will come supplied with a number of pre-generated theme worlds, exploring various "what if" events in our particular history. Users will be able to trade their own postulations with each other, and debate over the rules and pertubations which guided the simulation.
The "main time-line" of SIMTIME is an "earth analogue history", an evolved history that has been nudged throughout its growth to appear similar to our earth history, much in the way a bonzai tree is formed.
Particular attention is paid to the fact that the world is much broader than the western mindset, and that even within the western world there is a dangerous populist view of history which inappropriately reifies or generalizes the ownership of events and knowledge to all the disparate groups of people loosely labelled under that branch.
The main time-line will allow a natural branching off point from areas of our relative expertise. If a participant changes something in say the year 1200 then perhaps America ends up being fully colonized by the Eastern Chinese and micronesian tribes or the African Olmec or even the Vikings instead of the Europeans. Or perhaps the earliest arriving American Indians become more formidable to the European oppressors. Meanwhile other areas of the planet are unaffected and continue to exhibit their pre-evolved behavior until the changes avalanche to impact them as well (thus a local effect may cause global changes but the ripples may take hundreds of years to exhaust their full course).
For example, what if Galileo had not lived during a time when there were a number of brilliant astronomical events - two supernovae and a number of cometary phenomena. Would the peoples of the western cultures have taken longer to as a whole generally assimilate a more cosmos centric view of their world? Many other questions are also explorable: What if the internal combustion engine was teleported back to the Romans? What if the west had never had the dark ages? What caused the dark ages? What if Jesus had not come into existence on the world stage when he did?
In the year 4000 the sun goes nova, but humans have already expired due to congestion. Save humanity. (See Winning)
Because SIMTIME is a simulation and our history is one of the simulations available to it, it will be possible to simulate into our future to a short degree. Of course the simulation will soon break with reality because the past simulation has been carefully nudged ‘here and there’ to be in accord with our particular political evolution - and the future part of it will not have those nudges except where the participant chooses.
The big question is can the simulation be as dynamic as our human story? In the early 1990’s Yasser Arafat’s plane crashed in the desert during a sand-storm. He was missing for 15 hours but was found alive. Imagine the change his death would have had on that volatile Middle East situation. That kind of event would throw a kink into the future direction of middle east peace talks - and the possible branches from that point are precisely the kind of situation SIMTIME could explore. A world with and a world without Yasser Arafat and the result that would have on the politics, alliances and wars in the middle east in the years to come.
For those interested in alternative history, SIMTIME will feature fantasy modules based on parallel earths taken from popular fiction and Science Fiction. Many of these modules may be generated by the participants themselves and may be in the public domain.
For example, one intriguing history is explored in William Gibson’s "The Difference Engine". It postulates a kind of hand-crafted Neo-Victorian technology, or as some put it, a ‘Steam Punk’ universe – where the need for computation was so urgent that it predated better technologies for it. Ideally SIMTIME must be able to simulate a chance condition which would lead to computers being in wide spread use before the vacuum tube and then allow an exploration of that space.
Earth history is only one case. There may be other synthetic worlds included with a different yet exciting history. Definately to consider are fantasy worlds where magic really works or theme worlds following famous Science-Fiction and Fantasy novels’ universes: RiverWorld, Foundation, Xanth, Narnia, Tolkiens Hobbit, Dune.
Social
The most exciting use of this simulation could be to explore how our society and our people came to be in the position we are in now, here at the end of the 20th century, with our rich and diverse cultures, and with the good and bad ways we treat each other and our world.
For each disempowering negative aspect to our society, there seems to be a strong contrawise postive aspect. And historically there seems to have been progress in western thought, from the Magna-Carta to the Charter of Human Rights.
Today we have a technology which promises to let our species survive beyond just a single planetary history. Technology plays a role extolled ad nauseum.
Its been argued that our environmental problems are caused not by a series of separate blunders, but by a fundamental problem, a kind of lack of respect for creatures other than ourselves. Our constant struggle for our own local survival has blinded us to the repercussions of actually winning.
One special aspect of the modelling of the interrelation of physical events on the planets surface to each other is the recognition and designation of rules to govern those relationships. Pollution in the Hudson Bay affects its fish populations according to a specifically defined relationship. To iterate the simulation is not difficult, a given pollutant affects parts of the chain of life in certain ways. What is difficult however is to discover and "map" the relationships, and having a necessary level of abstraction for physical computation issues. Being able to model these kinds of relations stochastically rather than purely iteratively may be challenging.
If the simulation was not just used as an abstract play tool but was programmed with all the data from the real world - the status of everything in the world around us: fish populations, animal-life populations, tree-populations, soil-richness, air-quality, ozone-quality, water-quality and the best knowledge about the relationships of everything to each other. Then presumably we could simulate into the future a short degree of what would happen if for example we developed the Hibernia Oil Fields, or developed the James Bay project, or went ahead with California Big Green. This kind of look-ahead has never been available to us and it would be guaranteed 100% superior to the kind of look-ahead we do now, even though it would have its own limits. Local communities might find a greater sum profit from not doing something than doing something. And a single individual could potentially debate on an equal footing with larger organizations, coming with a non-obvious solution that on the simulations using the same relations as conglomerates postulate was clearly demonstrable as being superior to others.
One problem we have as a society is that we do not have good ‘cultural vision’ tools yet, although this is improving. We cannot map small scale widely dispersed events. All we can map clearly are mega-projects, single clear ideas where the simplicity of operation and the profit yield is just scaled up into a feasable profit range. We don’t clearly know what the effect of a blue box program is - and somehow a blue box program just doesn’t seem to be as romantic as an offshore Oil Rig. Certainly even if a blue-box program is more profitable as a sum that money doesn’t pipeline out into one place so its not really appreciable by the people giving the go ahead. But if you can show that even huge amounts of money appearing in one place don’t solve the kinds of problems that bother people locally - then perhaps that’s a foundation to be able to sometimes reject those projects.
Consider SIMTIME as compared to the real world. Of course we can’t copy the real world into a computer game. But there is a lot of boring drek in the real world. All we really want is to capture a little bit of the fascination.
In transporting the real world into a game context we can think of it as a series of metaphors - shrunken but not losing that original spark.
Our Metaphors:
Human Cultures and Individuals:
Replaced with small set of incredibly rich dynamic "actors", each one exhibiting the dynamic of a hundred people. We support up to say 32 of these genius level actors, with a supporting cast of a 100 dumb actors at any given time. Actors die and birth at a speedy rate but these bounds are not exceeded. We can even play a little respect to the Ten or Twelve tribes of man and have the game start with only Ten Geniuses.
There is the idea of The Muse - which travels along and strikes an individual making that person into a genius, their hopes and aspirations suddenly taking center stage.
Human Enviroment, "The Stage" and its time frame (4000 B.C. to 4000 A.D) at say a resolution of iterating every person once every single day during that time:
Replaced with a smaller stage, of say 10000 iterations total, 1000 "major" time periods (ie: years) and 10 "minor" time periods per major period.
Human Events, the observable procession of Human catastrophe, successes, the vigour and activity of Humans time on Earth:
Replaced with a special tool which permits observation of these phenomena, and intentional construction of above sections such that they have a self induced criticality more easily, ie more volatile.
Note: There is a degree of predictability, but mostly of hindsight - luckily the game is played with hindsight being a valid kind of sight!
So basically a miniature world - but compensated by being incredibly energetic.
WANTED:
ONE WORLD SAVER.
You are a caring person with a deep understanding of the earths social, political and environmental ills.
You are capable of making the strong decisions required to save this world.
This position is available immediately.
No experience necessary.
This product is a rule based simulation. The entire game history is regenerated on the fly every single round of play. When a participant commits any action anywhere in time the result of that action is completely unplanned and completely realistic with respect to the rule based behavior of the actors in the game history. There are no ‘decision trees’ or designer pre-scripted futures. The entire game must be developed at a suitable level of abstraction that much of the game experience is emergent, otherwise there is simply too much effort required to create it.
One problem in creating this game is the reconstruction of the game history every round of play. Consider that if the participant alters an event in say the year 400 A.D. then the entire game history up to the present must be reconstructed in an expedient and novel way. Depending on the resolution of detail of that history this process may take a considerable time. It is finding reasonably computable analogues for this process which the technical design focuses on. Finding computational solutions involves understanding the problem space, and finding some optimizations through identification of classes of behavior that can be treated stochastically.
An agent is any recognizable object, such as a culture, a wind-storm, a person or an idea. Agents may contain or breed other agents.
All agents in the game have their own rules for behavior, and the released version of the game is just a full run through of an entire fantasy history based on an initial set of actors with an initial set of rules (with a bit of a nudge here and there to make it seem as similar to Earth History as possible).
The SIMTIME concept of space-time is such that any change in the past is immediately propagated to the future. Space-Time is treated as a dynamical instantaneously stabilizing field. As the participant you commit any change and instantly get to view its net effect on the present.
Space-Time is also designed to be unstably chaotic such that very events may cause radical repercussions further along space time. Self induced criticalities are susceptible to slight perturbations. These effects are known as ‘the butterfly effect’, or ‘the straw that broke the camels back’, or ‘the snowball effect’. However, at the same time, there is a clear fitness landscape which limits what the successful strategies and outcomes are, and this will tend to mitigate and stabilize the simulation.
This chaotic vision of time is called the TIME-TRAVELLER syndrome: Time Travellers don’t exist because any effect they have on the past escalates so rapidly out of control that they destroy their unborn existence. This is opposed to the RED-LIGHT theory of time travel which suggests that time is self stabilizing and that ripples tend to die out rather than have wide sweeping repercussions. The second idea is a self-normalizing vision of time involving the idea of fate and irrevocable destiny.
Every variable which is not evolved should be generated from an appropriate fractal equation, or a number generator which produces the same results given the same input and is self similar at vastly different levels of scale.
Natural phenomena will be fractally based, and may be touchable by events however I may want to vary the degree of outside influence. It would be nice if the participant could always expect a certain storm at a certain time, then they could base battle tactics on storm conditions for example.
At a fundamental level there is reason to the rhyme. Events occur for meaningful, analyzable reasons. Its important that the participant not feel that the game is making its decisions in a capricious and arbitrary way. The participant may not be able to necessarily predict events, but once events have happened the participant should see the valid reasons for such.
The physical geography is fully resolved, and participants should be able to:
Rewrite the landscape at will, changing terrain, local weather and large weather patterns such as ice ages, people, buildings, water-routes, mineral deposits, magnetic poles, the orbit of the planet, its celestial partners and other strictly physical phenomena.
Rewrite abstract phenomena, such as the presence of memes, the migratory routes of animals and people, what urgencies and needs a particular individual or group is undergoing, the memory of a culture or individual.
Rewrite the types of organisms which exist and their suitability for their environment. Create extinctions or create new species.
Visualization tools must be provided to allow interrogation of peoples, their beliefs, their coarse cultural divisions and their individual and cultural memories.
The technological and artistic progress as reflected by various ideas must be viewable. Art and artifacts, literature and machines, both for physical materials processing and information processing must be identifiable and mappable to something in familiar to the participant.
A concept of ‘the muse’ is used to send some game actors into a more active leadership role, this must be visible as well.
There are a number of different user interfaces in the product. The interfaces show a view of the internal SIMTIME computations. They are detailed enough to allow the participant to generate mental predictive models on the workings of the SIMTIME universe.
You may find that taking the path less travelled has indeed made all the difference… but if it didn’t you can go back and try the other path as well!
General Play: Open Ended
In this scenario the game is open-ended; there is no specific win situation. The game can be thought of more as a "time construction set" rather than a linear progression of moves to a final win situation.
You return to a brooding desolate earth. The sun, in the early stages of becoming a red-giant is a huge blood red arc covering the horizon. The remains of a technocratic civilization are all that you can find of the seed of genius that you planted here so many eons ago. Being a God you can travel back and forth thru time, making changes to human behavior and seeing what effects it has. Can you make it such that humans survive till the year 4000 with sufficient technology to help them escape the suns growing grasp?
Multi-participant games provide competitive play scenarios which have clear win situations. It is not always beneficial to go back to the furthest point in time to effect change, since that perturbs ones own simulation as well as the opponents. As well, time pertubations have a ripple effect, creating an expanding cone through time which is stopped by certain natural ramparts.
The participant competes against other participants by trying to make the world scenario end up in some determined state first, or in some state that is mutually exclusive with the other player. They do this by nudging actors or events and thus changing and related actors life-history, causing a snowball effect that - hopefully - changes the entire evolution of history.
In this scenario the participant is all knowing but not all powerful. Here we’re trying to get away from the ‘god-like’ syndrome. In the simple version it was expedient to allow the participant very large control over nature - here it should be prohibitively expensive to throw that kind of weight around; the participant must look for the seminal opportunities much more carefully - and they cannot step into characters bodies and take them over etc; it should be what an ordinary human could do in the real world - that might have an effect if we were all knowing but not all powerful.
Ideally the participant would be limited in knowledge as well - but that compromises the particular character of the game. Let the participant see opportunities both crude and elegant - but encourage logic. Its better to make every single thing controllable: placement of mountains, rivers, actors-memory, taking over actors bodies - everything - but to just make it inaccessible by virtue of the raw amounts of power which it would require. Presumably there might be brief moments where the participant has enough power to effect one mighty change, but that should be the exception rather than the norm, since it defeats the elegance of the strategy to be that blunt; to end a battle one can either create a mighty earthquake, or simply drop a well placed banana peel.
The premise of this scenario is that you are an alien entity which crashlanded into the earth a million years ago. In the later days of your existence hominoids came upon the earth and multiplied across it. You find that you can manipulate their minds, and you feel that if you could do it properly you might be able to get them to excavate you and send you on your way. Your strongest area of influence is the local surface directly above your head - but you have some influence everywhere.
When the game starts you find yourself sitting at the end of time, the year is 4000 A.D. The human race has been extinct for 2000 years. In the year 4000 a solar flare sweeps the planet and kills you (cuts off your time continuum at its farthest end). Up until this point you were content to sit and ruiminate for millions of years - but now you have a new agenda, you want to get off planet. The question is given the tools you have can you save yourself? Humans seem to be the obvious tool - they are intelligent and can be controlled somewhat by you. As well because they live in a linear time continuum you can keep adjusting their history over and over until you get it right. Unfortunately there is that little puzzle of their entire continuum stopping at the year 4000. Looking at that issue will probably be one of the first items on your agenda - of course you can always put it off until later since the flow of time doesn't really mean anything to you. And of course you'll probably want to figure a way to encourage them to develop a technology and society sufficient to save you.
Alternatively you could look at the dolphins...its a bit more of a climb technically - but their minds are much more powerful.
When the game starts up you are faced with a view at the end of time. This is the year 4000 A.D., there is no sentient life on the planet.
You are a sentient being, having run through time, trying desperately to escape your enemies. You were caught in the solar systems gravity field and as a last attempt managed to crash-land on earth - it being the only planet with life in the system. It is 4000 years ago, and you're conscious of the ascendancy of the humanoid. You're idly curious - you play with them a bit - fascinated by the concept of an organic intelligence - then goto sleep. Its 8000 years later, you feel the anguish of death as the sun flares out of control.
There is no human life on the planet at the year 4000 AD - so if you want help you're going to have to figure out why not, and help foster it along.
Unless you can find some other willing sentient.
You are presumably still being pursued, your opponents may be on earth, they may still be waiting in space. It might be a good idea to promote humankinds ability such that they can help you out of this mess...
You have no mobility personally, but neither do your oponents if they are here.
You know that your "leaky" brain is prematurely advancing the human race. You know you have an area where you're power is at a maximum, radiates from u. You know that you are small enough to be held by human hands. You know you are tough, but can be destroyed.
Your opponents are the other players. Your areas of influence is small with some overlap between yourself and the other players. This means that the unfortunate humans in the overlap area are under two or more players control.
If you kill another player your power will quadruple.
Here the win scenario is to try and foster a technological development in the human race sufficient to get you off planet by the year 4000 – its conceivable that two or more players may win.
Oh yes, by the way, you still have other enemies in deep space - so you're going to need to bring people along with you to protect you until you make the 200 year journey completely out of the systems solar field.
Topographic Map
[ please refer to top view map diagram ]
The Topographic Map shows a geographical top-down view of the planet at any particular moment in time. Geophysical information may be displayed such as elevation, mineral reserves, tree populations, temperature extremes, magnetic field lines, radiation levels and such like. Other necessary if less obvious information such as antelope populations, total biomass counts and new foreign pollutants may also be shown. Even more abstract information such as average complexity of number of relations between active agents in the physical area, and the inter-agent bindings on those relations is visible. Various styles of visual representation capture these salient details in a way that is useful for the human participant, one common way is as a series of two dimensional overlays over the landscape, other methods could be as a volumetric display with mutual exclusion filters or edge detection filters to accentuate certain phenomena.
The mapping algorithms actively collect and display information to the game participant such that they can make wise informed decisions about their game play.
New mapping utilities can be added to create new layers of visualization. A facility may be provided for users to add new mapping layers.
Selectors are available to rewind or forward through time.
It’s possible for the participant to scroll the map around - there really aren’t any edges other than those arbitrarily settled at the moment when the entire map is being shown on screen - certainly characters in the simulation don’t every come across an edge-of-the-world (unless the participant wills such edges to exist).
Cultures are visible as different unique colors on the surface map. Successful cultures for example will typically cover more of the map than unsuccessful ones (of course long term success is a different
matter).
The map shows geographically related information such as location and spread of sub-surface mineral resources, vegetation cover and health, spread of pollution from its source and suchlike.
Based on mapping information participant is able to make informed decisions about possible actions to take to strengthen or weaken a given culture.
The first person perspective humanizes and romanticizes the play experience, showing the real world repercussions of decisions and implicating the time traveller more truly in their simulation. The traveller must search all that more carefully for that snowflake which triggers the avalanche - that moment where the fate of millions slides one way or the other.
The particular memes which make up the cultural milieu are mapped to conventional earth themes, so that Northern Warrior tribes look like Vikings, and that arboreal tree folk look like elves, and subterranean dwelling dwarves and valiant medieval knights, castles and the like all exist in a reasonable mapping. The artistic and technological heritage of a culture is a property of their memes.
Participants can ride whatever conveyances are available in the era, and talk with the leaders, or visit key events in first person, and interact in all the ways that other actors can accept input.
There is some question if the participant should inhabit the body of an inhabitant of the age. This seems like a good way to allow for maximum participation.
As far as actor behavior goes at a face to face level - this is an explosive topic today and is discussed in great detail later. Over the next few years these kinds of agents should be easier and easier to write as programmers become more experienced with the complexities.
[ See:
Cross-Time Map Drawing,
Conceptual Drawing of Actors in Space Time ]
The Cross Time Map provides a better long term vision of the migration and flow of cultures, by providing a volumetric view of change over time by stacking sheets of the top view on top another. A cultures growth becomes clearly visible as an expanding bole, and if that culture fragments or is killed off, a clear splitting or terminating branch is clearly visible.
By choice the default game world is shaped like a long thin ruler, and the cross time map is presented edge on only, presenting a simplified view which is easy to understand but still captures most of the salient details of change and migration of most large phenomena.
The Cross Time Map is a sleuthing tool to watch the growth of a culture, helping to find the points of self induced criticality of transition or collapse of a culture. From this perspective it is easy to select a key juncture and then visit that point in time to find out what key events occurred there, and then tracking back those events in turn, and or possibly changing them.
In a multiplayer session, this is the primary way that players do battle with one another, by propagating events out from early points in time to try and affect alien cultures. This interface is also the primary meeting place for all the players in the game – it is the central scoreboard showing a history of successes and failures, and often the first place a player returns to after affecting a change to a point in time.
Cultures themselves are loose collaborations of memes, and a colorization process is used to differentiate memes and cultures from one another.
The Cross Time Map is regenerated when the state of the game changes.
The game database saves information for the entire state of the games progress in order to allow the generation of the cross time map.
Terrain
The product lets the participant alter surface features such as water routes, location of minerals - even placement of a culture. Basically every aspect of what the participant sees is under the participants control. Some game scenarios may postulate limits on the participants abilities but that is a voluntary limitation of control discussed elsewhere.
The shape of the landscape is also under user control. A flat thin map is a convenient environment to play in, especially for a cross time view, however a sphere is more realistic. Having large polar regions is one way the user can have both features. Where the cross time map cuts through the land surface is also under participant control.
Beyond the simple landscape features, other phenomena are alterable as well. The distance from the sun, the orbit and number of planets and their hospitality, moons, tides should be under user control. This can goto any level of abstraction, however the focus of the game is to model aspects of the universe which strongly affect people.
Actors in the simulation interact in a rule based manner. The rules governing their interactions are bound to the actors, and implemented by simple algorithms. Some emphasis will be placed on rules which allow for stochastic prediction, but retaining reversability.
The term ‘agent’ is sometimes used to refer to a rule, such as ‘the agent of change’ or ‘an agent operated on a given actor’.
Actors themselves are just a collection of rules operating on actor bound state. Both the rules and the state of an actor are settable.
Actor state is fairly tightly bound to the actor, however some agents might bring or act on assumptions about default state if the actor does not have such a state.
Rules are divided into two main categories, fairly ingrained rules passed genetically such as rate of replication and diet, and cultural rules discussed separately. These are under user control, but the expectation is that they are not changed much. There is no evolution or mutation of these basic rules although there is no particular limitation on rules undergoing such changes.
It would be useful to be able to see all the memes, events motives, moods, memory, cultural memory, needs and wants that an actor is undergoing. Other information such as societal status, familial relations, spatial position would also be helpful.
Memes are a variety of rule which attempt to propagate between actors. A meme can be any expression however it usually won’t propagate unless it has a viable vector, usually by doing some kind of service to the actor. Memes interact in an information sphere, using cultural alliegences, likes, dislikes, memories, and other memes as inputs.
Inventions are divided into two rough categories which map to the two ongoing revolutions of humankind, the mechanical inventions which permit better processing of foods and such things as primarily related to the key substrate the actors are imbedded in, and information inventions which permit better processing of inputs that guide actors through their material substrate.
Evolving useful inventions and memes may be difficult in the extreme, and it may be necessary to provide a wide number of invention components or meme components which have certain permissible combinatorial patterns that can be used to manifest a wide variety of useful and not so useful inventions and memes. Inventions may also always be directly bound to a meme.
The cultural reef of artifacts generated by actors should be editable as well. There may need to be SIMCITY like concepts for manipulation of large buildings and their role in a society. There is a question if these new buildings should be treated as common facts, or extraordinary gifts from the gods.
Essentially the simulation should be editable to the level of its granularity, both for ease of development and pleasure of use. The granularity itself should be sufficient to capture the simulation of humanlike behavior through time, and sufficient to observe and operate on moments of criticality. Interference patterns generated by the highest level of granularity will actually increase the resolution beyond that of the simulations scope itself anyway.
The initial focus is on building a reasonable model of a solar system, with appropriately rich physics so that planets can work. This is a foundation component which is more robust than the initial needs of the application, but which provides a good growth path.
The earth surface modeller could be done fairly completely without too much trouble. And the various temporal displays are fairly easy.
The agents and actors are much more difficult and require significant analysis. Some kind of meta-meme componentry is needed to simplify the potential space of memes and inventions.
Exploratory at multiple tiers
SIMTIME is an exploratory game designed to let the player explore what it would take to save the world. It also offers a historical perspective, allowing the player to fully explore time, and what the effects of decisions that were not made could have been. By flitting back and forth through time, and changing events in time, the player can construct and nudge a future that then perhaps won't end in the same tragedy as ours. The game can be thought of as four tiers - the higher supported by those below it:
Tier 1:
A way to test ideas for saving the world. And catering to that audience which feels a real need, and would get a real satisfaction from at least being able to "playtest" their ideas.
A game which lets you see in infinite detail the repercussions of your decisions, or of events naturally propagated by the simulation. Tier two is more of an education process, to educate the player as to the kinds of dynamics and complexities that a real world might really offer to confuse and confound even the most insightful policy makers.
A simulation of a whole earth history which lets you go back to the beginning and construct and nudge an entire history until its perfect. Tier 3 is the assesment process. Its the way we determine a win or a loss situation - the actual application of the players theory. If you can bring a viable world of living succeeding human beings into the 25th century then its considered that you've won.
A real serious tool for understanding the dynamics of human and world interaction. More than just a game, a prototype of what will one day soon (on a much larger scale) become the system by which all citizens of this world understand and participate in a new global electronic participatory democracy.
The precise pinpoint of what SIMTIME is all about is that it’s a "testing ground" - you can test policies - test them in combination explore the layered result of changes to your world - and then scrap them and start all over or partially over if you don't like how the world has turned out.
Lets you be monarch
and do social planning
and see what decisions are best
before committing them.
Its not a testing ground exactly - in actuality you *do* commit decisions but you can uncommit them, so for example if you find that some policy you implemented way back in the 1500's turned out to be really really bad in the 1950's - you can go back to the 1500's and alter it, or annul it thus changing the entire future course of history up to the present.
Its like a chess game where you can take back your moves - and keep repositioning yourself for the optimum play.
But like playing against a chess master - being able to go back and change moves still doesn't guarantee a win situation. The simulation - like the problems that face us in the real world - is so tough, that you're virtually guaranteed to have to go back over and over and change and change your position until you find a real solid stable foundation for a viable future.
The concept behind SIMTIME is that "SIMTIME lets you, the player, explore and manipulate human history on a grand scale.". You can travel back and forth through time, seeing and interacting directly with SIMTIMES "actors". You can change events at any moment in time that you wish - and then rush back to the future to see what affect your changes had on the world as a whole.
Imagine being able to create your own country. A country with with proper laws and a fair and just morality shared by all. A country where politicians actually make good decisions - even if harsh. A country where irreplaceable natural resources are not squandered by the few.
This is the fascination of SIMTIME.
Perhaps you feel the immorality of our current age is bringing us to ruin. Perhaps men and womens lack of love for fellow men and women is the cause of all evil. Perhaps the politicians are impotent, heaping injustice upon the poor masses.
In SIMTIME you can change all that.
You control the life-history of an entire world. You have special tools which let you pinpoint the root of a problem. You can go back in time to exactly where a bad decision was made, or a bad event happened, and actually change what occured. Your powers can be subtle, such as encouraging a snowstorm on the day of a battle, or blunt, such as taking over a leaders mind and changing his or her decisions directly.
If you make a mistake then you can return to the past and undo your decision, or further alter it.
The premise for gameplay is that "your world starts out polluted and dead and its up to you to rewrite history such that this tragedy never happens".
SIMTIME lets you scratch that mental itch - the frustration at what the world is today, and what it seems certainly doomed to become. You can try to remake the world in your image, with your history, your beliefs, your rules.
The ability to change time is a popular idea in our culture. Many shows popularize the theme, among them are "Back to the Future", "The Terminator", "Bill and Teds Bogus Journey" and "Quantum Leap". We feel that because of special programming logic that we can create an interactive game that has the kind of rich dynamics of probability that these movies popularize.
By carefully scoping out the domain of the simulation we can recreate an entire 8000 year segment of an earthlike believable history. One which is as fully detailed - and has as rich a probability as the movies popularize. Everything is simulated, the behavior and interaction of people, even the planets ecosystem - right down to little ice ages and seasonal animal migrations. The player can treat it as just a game, or as a serious explorational tool for our world and our common future.
SIM-CITY was a success because people care about cities. I and many people around me have extremely strong opinions about how a city should be run, "how things should be". We constantly point out the terrible mistakes that government makes in structuring cities. When we are thrust into the role of a city-builder we all of a sudden have to consider the other side of the coin, how to make a city "work" - to have a functioning industrial base with rapid access between the facets of production, without making it inhuman. We get the chance to make things the way we want, and we get the satisfaction of successfully pushing the envelope between what the simulation demands and what we want.
SIMTIME has a different yet equivalent appeal. Here the player has opinions about how history should have been, and thinks about the stupidities and blunders of our ancestors. Here the player can play God, and rearrange how cultures grew, evolved, fought. They can try to create a world that is much better, that was spared the apparent stupidity of politicians. But maybe they will find out that it wasn't so easy - even for a God - to make things always work out the way they wanted. SIMTIME turns TIME into a 3 dimensional board game, bringing TIME itself - and all its fascinating probability - into the reach of the player. In a sense it is the ultimate game - a metaphor of ourselves and our whole society.
A funny story: a friend of mine, S0ren Gr0nbech, once wanted to make a game which was just a simulation of a city, you could plant buildings and make its denizens happy, and watch it grow. We of course ridiculed the idea "Who would want to play that?" we said "Just a city? But how do you win?".
Here is the player playing the game. What exactly does the player see or do? I believe that they should see something romantic. They should see something that is logical and reasoned. Their resultant actions should be constrained so that they do feel some frustration.
A) A parallel earth fantasy world, devoid of any real technology change, with all manner of individuals roaming its surface, doing wild and daring things, becoming victim to hapstance accidents or evil intent.
B) Making anticipatory decisions based on limited input, yet with a best-fit analysis of pending situations - and current and past circumstances.
C) The behaviour they see is generated from a series of primitives that in
composite appear to be very freehanded and radical.
D) Overt rational behaviour, that the player can have explained to them or can infer clearly.
So the player sees actors who:
1) May be doing something radical.
2) Is able to justify that action (more or less).
3) Possibly sweeps other actors up around them.
The player sees people at a top view level - but the real intrigue comes not from watching these dots move around like ants, but the knowing of whats behind their actions. Basically either they commit their action in a fascinating way - which would preclude the top view because its just a dot at that level - or their intent is fascinating.
We'll presumably support a 3d view anyway so fascination of intent will be
communicated, as well as servicing the desire to get away from the top view kind
of action.
Presumably in that case the top view will not show individuals, rather just the power blocks of established or emergent cultures as placed by the general physical location of member individuals.
If there was a specific computed locus, would this mean that people of one culture stay together? No: rather people become influenced by the area they are in - or they drag others along with them, thus appearing to stay in one area (hopefully).
Its safe to say that people have relationships to one another in this game, and that that forms a basis for many actions.
There are smart-actors and stupid-actors. Actors who are smart can be said to be strucken by the muse. The muse may be a wandering force, or one that
stays with an actor and their children throughout history.
Given exceptional circumstances, a person at the center of a locus of
intriguing events may become a smart-actor?
Or should we bind smart actors to people? - throught their lives (i like this way better)
So relationships with one another is an example of a reasoning mechanism
driving individual actions.
Yet this adds a measure of complexity, for example who drags whom? When a people split is there a drag force seeking to rebind them? What kinds of memories do these people have if you query them? ( presumably there is a memory of relationship driving their actions ).
Their memory in general would be the general case of this example as well.
If there is a lot of memory management then it may be wiser to keep memories on a cultural scale , a Jungian memory reservoir rather than individual memories.
So lets say we keep a memory resolution of every person which each person can back-track and analyse for information.
Now presuming information arrives (an as of yet undiscussed problem):
That person has to make decisions based on their input, of course sessile actors will do nothing, but smart actors will act dramatically.
They have to be able to do the following analysis for example: - This is a person whom I know and feel x for (a memory). - This person is in this situation, or I am in situation relative to person. The outcome of this situation will be ok/good/bad (future meme). My actions should be x, and are based on: 1) my past feelings (feelings are critical) 2) expected outcome of important current situation.
Is this how I envision Orrery? Is this how the player will envision it?
Is this how I see the romance of the product? Cultures colliding, coming apart, and the actor level decision making that drives these actions.
This could be a proper envisioning if these individual actions can become major avalances. It requires that the Orrery simulation promote the buildup of avalance situations.
Is watching individual behaviour romantic?
Yes: where that behaviour is the visible root of large events.
For example:
An actor who gets a vision and founds a society based on it.
An actor who goes to war because of somebodys actions
An actor who decides to explore the planet
(ie: things that are or become culturally visible)
What triggers actor actions?
There are action stimulators, all stimuli roots are based on:
Memory
Observation
Stimulator responses are going to vary widely, angry responses etc.
There are abstract actions (probably hardcoded):
To take Covet things, people, "necessities -> metal,water,food".
To give
War Involves Ability to drag along followers.
Explore
Religion
What about the Complexity of Individual Lives?:
Much behaviour must be pre-built into the individuals, seeing as how we cannot expect them to develop a desire for growing crops and marketing them. So presumably we should not be dealing with them on that level. Rather these people should not have any "needs" per-se except in a general sense (as in a quota - needing to be over rich and fertile lands etc).
Now people may still have occupations like farming? Or is that again too specific to implement in all its variety?
Cities will presumably come into being and individuals can be fractally assigned life duties while in an area (life duties may change). People in less dense regions presumably are farmers, conditions permitting.
How can we avoid the need to be able to show a farmer going to market?
Maybe we do want to show that since its one of the things that can be
intercepted...a farmer might be killed while on the way to market. That might start some kind of war or something.
This implies a market however, and different social structures have different kinds of markets...so if it was a Communist society the market structure would be different than if it was Capitalist etc.
Cultural Level Memories:
Having different market systems implies Cultural level phenomena (since all actors of a culture would presumably know enough to goto the same market).
Each culture type is presumably a composite of primitives. Each one decides all its political structure, market structure, social structure, etc.
A group of people will have their own social/religious beliefs which may be overlayed by the culture they are in, however if they are not disenfranchised by being separated from each other (as children) then they'll take on their parents beliefs.
Ok so here is a mechanism for which culture is carried on:
Children take on the values that they are associated with. If their parents are of a strong religion - yet do not have an enviroment to support and practice that religion among like minded individuals, then once the child is of age its coloration will change - usually to that of where it happens to be - or of that of a travelling meme.
For a culture to be a culture it must hardcode certain behaviour (i don't have time to allow systems to be fully natural).
I think that ideas are one of the more potent instigators of change, therefore we're going to need a way to have ideas travel - since we can't really expect to have books and suchlike I believe that we should have either memes travelling directly, or we should have travelling minstrels who preach their beliefs etc etc etc. When a child is exposed to somebody with a belief system then there should be the opportunity for that child to take on that system. Its possibly a question of visible profit of that system as compared to their own system.
Whenever children are exposed to one another there is a potential for a transfer of beliefs (usually occuring once, perhaps more, depends on richness of child (capacity for change)).
Each child has a number of beliefs which affect its behaviour political beliefs (as excercised by culture it is in).
There is quite probably a "culture" structure which each child has either imbedded in it, or is dynamically assigned per culture, it tells people where to goto market etc...
SIMTIME models the earth to act as a source of dynamic events which may motivate actors to suddenly change their behavior. The earth model is limited however, it trys to only model critical, people affecting events. Since our focus is people and culture the earth simulation is only detailed enough to provide a backdrop for their interaction.
All land phenomenon can be overridden and controlled by the user. If the game knows about it, then the user can alter it. Of course within the context of the game the users abilitys may be limited, but for creating a world the user has unlimited powers.
The planet surface itself is fractally generated, the surface consists only of a land elevation index. Land below a certain height is presumed to be water.
Some stellar and freak phenomena are simulated. The daily behavior of the moon, including eclipses of the Earth or Moon. Comets, meteors, and super-novas, as well as rains of frogs or other creatures. A solar 11 year cycle is simulated (affecting world temperature on an 11 year cycle). A longer term 200 and 2000 year temperature cycle is also simulated allowing for little ice ages. Other freak phenomenon include earthquakes,
volcanoes, airborne volcanic ash clouds, glaciated rivers, tsunami.
The world has no night or day specifically recognized. Its assumed that it does happen, but that its not shown or made a part of gameplay (other than occultation of the moon by the earth).
The land is fractally assigned various mineral wealths, soil richnesses and water carrying and reserve capacitys. These reserves may be depleted, water reserves can rebuild.
The planet has two poles, and the temperature decreases towards each pole uniformly.
Air flow currents and water flow currents as well as air temperature and water temperature are not in this version of the game.
Plant cover is not fractally assigned, it is simulated as a function of land viability, this is a function of land soil richness and water reserves. Natural plant cover (known as "forest") does not deplete local land resources, and can stretch from fertile regions into sparse regions. Human crops deplete water reserves and soil nitrogen/richness. Some other crops may reverse some of the factors.
Climate and weather is dynamically fractally created. Weather patterns are evolved over planet surface over time, but will be designed to be seasonal. Rainfall increases water reserves (water reserves affect plants). Snow is allowed, as well as long term snow cover.
Cities and the structures that people create are handled elsewhere, although actors can dam rivers and suchlike which are internally handled here.
Spirit
The big feature of Orrery is getting to watch cultures evolve, grow, die, spin off.
The player gets to do get to try do a better job, but has the kind of perspective that human beings just haven’t had. And their perspective is from a cultural viewpoint - the player sees the big picture by seeing how the planetary cultures respond to their actions.
The player can see each culture on the TOP VIEW map as a different colored blob - each one usually having its own geographic area - possibly overlapping - (as discussed in the main document).
The player can see the evolution of each culture on the CROSS TIME map (as discussed in the main document).
It seems that culture implies many institutions,
ECONOMIC SYSTEMS
Free Market economies, Centralized economies
Ownership of land, rules on bartering, taxes
Welfare programs
SOCIAL BEHAVIOR
Wars
Marriage
PERSONAL BEHAVIOR
Rules of conduct, beliefs regarding violence etc.
LAW AND GOVERNMENT STRUCTURE
Patriarchy
Local Area Government
Centralized Government
Pure Democracy
Basically a culture is the highest level entity of SIMTIME. This is different from the real world where a country may have many different kinds of culture in it. Perhaps the term culture should be thought of more as ‘country’ here; and the question then becomes what defines a country.
There are these necessary algorithms required to visually render a culture (at any given moment in time):
1) Determining what constitutes a culture.
2) Determining the main drawn region of a culture.
3) Determining the members of a culture.
On the face of the planet are a number of individual actors. Each of these actors has a slightly different set of rules for their beliefs/behaviors. Luckily actors who are geographically bound will tend to think and act like one another so there is usually a geographical correlation to like minded thought.
All these individual actors on the planet are going to be conceptually sorted into a number of coarse groups depending on how similar their beliefs/behaviors are to each other. Geographical location is one of the sorting parameters (to speed up the sorting process).
We determine the likemindedness of a group of people by examining their beliefs/behaviors from a sufficient distance - or lack of resolution – such that they appear the same. A special algorithm accepts as an argument the coarse beliefs/behaviors of a person and returns a color guaranteed to be unique to that set of beliefs/behaviors. I imagine this algorithm is going to take the three most important parameters and simply generate a RGB 3 dimensional color cube and select a color within that zone. This guarantees uniqueness and also guarantees similarity of color between similar groups.
So basically a group of like minded actors is a culture.
There is the concept of continuity - a culture which has existed should be shown to be continuing to exist. If the flavour of the culture changes – some global paradigm shift - then the color at large should reflect that change in flavour accordingly.
Sometimes cultures may be self reinforcing. If the meme of nationalism has infected a culture then its members may toe the line much more rigourosly in terms of beliefs.
Actors are going to be able to know when they are of the same culture. If their cultural beliefs permit they are going to react as a group to threats against the group or individual members. As well they will seek opportunities which better the group. This kind of abstracted behavior is going to have to be set aside as a note for further examination.
The algorithm for rendering the drawn area of a culture is: colorize the region that encompasses the main body of all the participants of that culture. So we don’t encompass the lone explorer in enemy territory and distort the perception of the wellspring of the culture.
Note that this model of culture suggests that drawing culture is an entirely computational process which takes place every time the map is regenerated, and that actors are the only repository of culture.
I believe however that once a culture is computationally discovered that it should be remembered into the computers memory. And that the color of the culture can serve as an "ID" for discovering if actors are members of it.
All I want to allow is the player to select a culture right off the map and interrogate all its attributes - examine the general belief systems and behavior (and rationales) that make up all its composite rules.
Its important for the player to be able to do this because this is how they determine the success of their game play - by being satisfied or dissatisfied with what they have created. Its the critical form of feedback.
Of course the player can interrogate individual actors as a separate process from examining the culture at large. While the culture may only show big memes there may be finer grained memes that are actor private.
The act of actually viewing a culture or actors memes involves actually bringing up a text window with a text printout of the meme involved. There is a discussion on meme components elsewhere. Suffice it to say that a meme involves the idea of a belief motivated behavior, and that that belief has a rationale which the player can read about. At some level, all memes are hard-coded by the programmer - its just their interrelation and combined behavior that is un-preprogrammed.
Large scale events are going to be the seeding grounds for cultural change. The idea of human migrations, exodus, natural cataclysms, war, exploration suddenly opening up new lands, harser ways of living causing new kinds of rules to come to the fore, more evolutionarily successful memes suddenly overpowering a cultures previous memes. These are all methods by which cultures are going to be created or destroyed. There shouldn’t be any worry that there isn’t going to be change. The cultural level will function perfectly fine if the actor level can be made to behave dynamically.
Culture can be thought of as a group of binding ties between like minded individuals. Of course thats a nationalistic view of things.
The only problem from a computer aspect is that the conceptual and ramificational structure of these ties are quite unique and difficult to evolve.
For example a culturally inspired set of beliefs may dictate:
Economics, Free Market, Communist, barter...laws of ownership or lack of. Religion, Patriarchal, Mother Earth, Single God Complex…Social Behavior
All things that permeate individual actor behavior down to a very fine grain level.
We are particularily interested in:
1) How culture is perceived as a whole
2) How it justifies individual behavior
3) How to construct cultures
In regards to 3):
While it is possible to construct a series of hardcoded cultural rules which moderate individual behavior it is difficult to come up with the core rationale for those rules <- presumably the mechanism by which those rules evolved.
For example, when does a culture become aware of itself, as an entity with a definate "character and destiny".
Now in human society the evolution of culture is fairly visible - certain ideas and ethics seem to "come-together" at critical or opportune times to become the enduring foundation of the above mentioned dictates (economic, religious and social behavior).
Sometimes a people are thrust together by circumstance - like minded people refined by circumstance from a melange of peoples.
I perceive that we could simulate culture by having each culture be the consolidation of a number of composite beliefs - some of which may be mutually exclusive, and all of which occupy one or more cultural "slots" such as the above mentioned dictates (economic, religious and social again).
Alternatively we could take all the memes of each actor and pass them into a black box which would return a magic number. That number would be their culture. The black box would be taking their most important memes and generating a number from a multidimensional space of which each meme was an axis.
Now only problem remaining to the dynamic construction of a unique
culture is its birthplace - what forms the birth of a culture?
I can see some mechanisms, as a specific algorithm:
When people are grouped together they form a consensus of opinion or leave the group. Splinter groups?
Basically the culture is moderated by the unique beliefs of all the citizens - and in turn the citizens begin to reflect each others views.
Using this mechanism it might not be necessary to have a "culture structure per se". Yet this in turn yields the question of how the culture is rationalized: "We belong to this culture because of the great event xxx - and the talmud, constitution, bible, myth... has ordained our behavior".
Codeification of cultural beliefs must be visible to player!
2) ??? (the above mechanism seems sufficient...)
3) I do like the idea of "refining" where like minded people tend to group. Thus hapstance may play a role in the initial members of a group.
Grouping would be a physical phenomenon, as well as a response behavior (ie: in given circumstances these people will act thus).
Of course similar growth situations may yield similar beliefs.
Now some other little issues:
It looks like all these issues are actor individual issues rather than cultural level issues.
Actor Behavior thoughtsThere is a fixed number of hard-coded stimuli which an Actor is capable of perceiving. There is a fixed number of responses with which an Actor can respond to the world around themselves. The soft-coding is the connections between these two arrays. Each actor has their own evolved soft-coding for responses to stimuli. All the possible connection types are hard-coded, but the particular ones an actor has makes for a soft appearance.
1) Stimuli Handling
2) Stimuli Sources
External World input
Self-input from desires
Actor to Actor dialogue input
Mental associations input
3) Doing Stuff
Mission request handling:
Mission scheduling:
Mission doing:
Stimuli handling:
Input is in the form of one of 1024 numbers. Each number means a different thing that the actor can perceive. The number of things that the actor can see is extremely finite. As well it is extremely abstracted. Input to the actor compares itself to an array containing three fields: the input type, the meme and an action suggestion.
When input matches an input type then the meme code is called to evaluate if that input is worthy of a response. The meme code is manually written and can access what's known as "desire bases" to assist its decision. The meme decision is to either submit a "mission" for possible doing or to do nothing. Many memes will be codeless - just a direct connection between input and suggestion of action.
Ramifications: Since each actor has a unique set of memes as inherited from friends and parents - its responses may be unique. Mission submission does not necessarily mean mission will be done. This fully defines the input handler part of the product.
Stimuli Sources:
A piece of software pre-processes all information about the world around the actor and submits each item one by one to the actor for evaluation. Only new events are submitted.
"Desire bases" are basically things like "anger", "hunger", "jealousability". "Desire bases" like the tempermant of people in the real world is inherited genetically. Every actor vision iteration we also update the actor desire base. The actor will grow hungrier and hungrier for example. When a desire base gets high enough then it will trigger its own input into the input path. The memes will evaluate the desire as if it was a stimuli.
Actors near each other may propagate information to each other by a specific action. The receiving actor gets the input as stimuli.
An additional feature which is purely optional is to create mental linkages between closely occurring stimuli. There would be a finite number of links allowed per actor, and they would be re-cycled (weakest first). Any two timewise nearby inputs would be linked. When the actor received an input the related one would be "refed" back into the input path. This would give the actor an associational memory between events. Each link could have a counter to strengthen it, and each input could have a volume. If a link was very strong then the input would be submitted at full volume, basically the actor would anticipate something upcoming. This is purely optional, anyway the memes can do this kind of thing themselves.
The input routine (above) will submit "missions" to do. There are 1024 possible missions an actor can undertake. Each mission has a "hotness" field and a duration. A suggested mission is undertaken if the actor has nothing better to do. This is determined by simply comparing the hotness of the current mission to the one suggested. The previous mission is often discarded. There is one special case where if a new mission requires a very short period of time and the old mission is of sufficient comparative hotness then we'll push the previous mission onto a limited size task stack.
The task stack runs the current mission till it is done. It then pops the stack to the previous mission and begins running it. If there are no missions then the actor does nothing.
Missions run under a non-preemptive multi-tasking. Where each mission does a number of commands then calls "wait" - or issues commands which implicitly require a wait (such as "sail boat to foreign lands").
While a mission is waiting it is possible for other missions to pre-empt it more easily, but when it is done then special code forces it active again - preempting and discarding the running mission (I may allow a special check to to allow the currently running mission to close down).
Virtually all missions are instantaneous. Most of them are immediate responses to changes in the world around oneself.
Each mission is hand-coded by the programmer. They range from tilling the fields, to killing the dragon, to career occupations. Missions
make the actor do stuff. The code affects the actor directly – calling a number of primitives to effect change on the actor - such as "goto this place", or "kill that person".
There is a meme pool of fundamental discoverable memes - these are hardcoded right into the game. Each meme is backed up by an event or a physical object (such as a bible). Its possible that memes may map onto events that happened in the games history. Although all memes are hardcoded they may interact to form composite memes. Therefore memes should be simple enough and direct enough to not overlap. The idea of rigid memes suggests a certain number of behavior-slots which the memes fill. This is true, behavior regarding any possible thing has a slot and there can be a kind of behavior to fill that slot. Behavior with regards to enemies for example. Memes will basically compete for survivability in the human organism they control.
We can't ignore how signifcant Inventions are to history. SIMTIME is an exploration of the evolution of history - not of a stagnant history so this is an area we address. There is going to be some algorithm which allows for the mapping of discoverable inventions onto individuals. Inventions will probably be discovered when the social climate warrants it (a combination of need and pro-inventive atmosphere) and by the right people etc. Inventions must be characterized as to the need the fufill and how it affects people as well as the incidental effects they have etc. I guess in a way inventions are a kind of people in that they draw on resources and have a contribution to the meme pool or at least the carry-awayable-idea pool.
There is the basic earth surface simulator, as well as the associated cross time maps and top view maps. The 3D view and any culture-mapping is here too.
Memory of people is an interesting concept. I don't know how I'm going to handle that. Perhaps a sparse matrix of people against the events they remember where maybe socially shared events can be coalesced into one node. People must have a memory because it so much affects their behavior.
People interaction is important - when people collide they are going to share memories - especially if they are urgent - in fact the carrier may not think a memory is important but the recipient may - and may probe (perhaps I should compare all of recipient responses to all of carrier memories and if one triggers a significant alert status then pretend it was communicated (given that both people are on the same side)).
Aside from memory exchange and the growth of a jungian memory pool through collisions there is also the critical concept of meme exchange - and susceptability to another persons memes. This depends on various computable factors such as the karma, the success and the gullibility of people involved. The meme exchange really dictates how the game is going to be and how cultures are going to be percieved. There is a deliberate normalizing influence on people in physical close proximity such that a culture can be formed.
Actions - This is not so much a component per-se but it is something that is basically the foundation of the game, and a result of the above pieces: Memes really dictate actions and thus individual actions are going to be accountable - you'll be able to interrogate a person and see why they decided on a course of action. In SIMTIME when having a lack of good information my characters will be pro-active rather than sessile. memes are the collection of two pieces 1) a meme action, 2) a meme rationale - possibly pointing to a historical event. Meme actions are triggered when the meme is triggered as a result of a related event happening in the real world being active enough to get a response. different memes have their own differing sensitivities. long elaborate schemes as carried out by actors should be possible given the right conditions.
Each person has a habit manager which is just a hardcoded way of making them do the things they have scheduled - so that they can interact with other people. presumably this is congruent with their goals and their lifestyle - 2 new issues.
Looking at personal interaction with memes only:
Person:
can chain to an infinite number of memes - each meme has a trigger
each person can acquire as many memes as possible.
originally we had behavior slots so that external phenomena knew which
memes to trigger - but that is too limiting - lets just to a linear scan
through list for collidant memes, redundant ones can fight it out also.
Events:
all events in real world get some interpretation to the personal level. there
are specific events and they are rendered into something that the meme can
detect and collide with - possibly triggering a response.
i supose that interpretation should be subjective - dependant on the meme.
the brain viewed from this way is a machine for digesting and getting work
from memes - like the eyes are a machine for eating light.
Meme pool array:
list of memes, their triggers their rationales their resultant actions.
basically these should be composite non-conflicting behaviors but they
may instead of being only memes may be whole behavior sets.
i don't know if memes can really be composite and if they can make
composite actions - the time relation factors seem to be a problem -
a decision to do something then followed by another related decision seems
complex - it would be easier if it was a top down process where there
was one overriding goal broken into several components steps to achieve it.
So we're looking at the way people react to events, what events what reactions?
Well we can take it for granted that people are stimulated by outside phenomenon, such as they find out that something happened to something they like or dislike, or they encounter something they like or dislike Basically encountered stimuli is compared to internal responses and a reaction to that is formed. a sufficiently strong reaction promotes a physical action based on previous action and which is broken into a series of doable steps. more global actions reassert themselves once immediate stimuli is taken care of - reaction to immediate stimuli is inline with global actions it seems. - a re-statement of my AI ideas.
The above AI idea is possibly too complex. Its possible to encode at a higher level of behavior - decide what all the stimuli are and what all the reactions are - then simply allow the memes to migrate as usual thus giving each person different reactions to the same stimuli.. In this case we should have have behavior for life-things the long term behavior - and for immediate distractions as well. people can come under the influence of memes affecting either.
As a youth you get exposed to the meme idea of being a farmer and the benefits that it entails, as expressed altruistically, personally, socially in some sense of benefits and reason, justification, rationale.
You adopt the memes and the habits and behavior which it engages you on, depending on your commitment you follow it more or less properly when better ways of farming come along or you come along to different practices then the way your farming meme gets expressed is moderated by those new phenomenon. (ie: the invention of the tractor, or introduction of horse)
Other competing memes such as be a doctor are moderated by your circumstances - they should appear less appealing if everything is setup for you to be a farmer, so how that analysis goes on is undecided as of yet - and of course if you are already a farmer it probably takes some fairly strong outside rationales to change course. here we encounter the idea that the meme thrives because of circumstance - and its acceptability depends on how closely the circumstances it demands meet the circumstances that are present.
We model people directly. A person of a culture is a certain color and when map is showing a culture and the area it occupies what it's really showing is the loose colored zone around all the people in that culture. From the overhead view you can see each person and click on them to get a breakdown of their beliefs, their culture, their personal position and daily habits and suchlike, if you wish you can watch them as they progress about their daily chores.
Each person has some fundamental facets:
0) Age
1) They have a meme for each kind of basic behavior that they exhibit. This includes daily ritual habits, and weekly or seasonal habits. All types of occupations and their behavior are hard-coded.
2) They have a linkage to their culture.
3) They have a simple memory - which is used to propagate knowledge and to remember enemies or special friends.
4) They may or may not be under the influence of the muse.
5) If they are under the influence of the muse then they may have unusual memes, and/or they may have inventions.
6) A seeded random generator for text, poetry, art, name, general habits, appearance, demeanour, intelligence, strength.
7) They may have some riches - money, or work exchange systems, various assets. These assets may not be on person.
8) You can view the genetically related family of a person if you wish.
9)As well you can view the people whom they are influenced by or who have influenced them.
As with all of the game you can change anything that you can see. You can change beliefs, and you can inspire them to new heights if you wish.
The game allows certain large wild animals. These are animals which bear a direct relationship to man and affect him: Whales, dolphins, bears, tiger, deer, horse, oxen, tunafish and geese (as migratory creatures). There are a number of herds of these creatures modelled all the time (in essence simple actors) they follow basic migratory routes, bear young and have a growth rate and young care duration. they are classified as tameable, predators, carnivore/omnivore/vegetarian/ solitary/herd/pack, and have a habitat (prarie/forest/highland).
These animals will all be simulated either as herds, or will be randomly and spontaneously have a temporal presence as a function of their probability in a certain area. A horse for example only exists as a herd, or as a flag signifying owned item of a person, it doesn't take up any memory. Although a temporal, memory using version may exist between ownerships.
Animals exist also as a sign-marker of how well humans are doing. When herds decline it suggests human pressures. This planetary animal populations and their behaviorial change as well as decline will become markedly noticable towards the end of the game as the human population grows and colonizes most of the planet, as well as cutting off most migratory routes and aireal route stop-overs.
People exist as individuals, but are simulated within the context of their cultures. they live and grow families, their occupation is based on their parents, or local influences. an isolated rift of people will sometimes become their own culture. most cultural dynamic behavior is pre-programmed: cultures goto war, exchange goods and services, use resources, pollute, exhibit various aspects of the world.
Any individual in a culture can take on leadership traits, these traits will cause local conflict in a culture in a number of pre-defined ways: strikes, civil war, suffrage and emancipation movements, helping the poor movements, religious things. all these behaviors are hard-coded, and there will be a sufficient number of them to give the game an apparent dynamic behavior.
Cultures will have a pool of culture type rules from which they will take their behavior - as based on the behaviors of cultures previous to them. this includes aggression, colonization, respect for world, and a number of other factors influencing decisions based on new phenomenon (such as being attacked, or discovering new land, or under a drought situation or any critical cultural level pressures).
Cultures will discover inventions, as propagated from a single person, inventions are assigned based on a tree of critical path (certain previous inventions must exist) and when the need is sufficient, or sometimes just when the culture reaches a certain size. it should be possible for inventions to radically change the makeup of a culture, switching it from agrarian crop raising to industrialization for example. this would occur by the invention changing the profit origin point for a culture - each culture will have certain energy requirements to grow (memes allowing) and that may come from trade, crops, or other.
Actors won't have memes privately anymore, although as constituents of a culture they do have access to cultural memes, and there may be some residue from previous cultures.
Also actors won't be simulated individually anymore, they'll just be part of the culture. although it will be possible to point out individuals who contributed to cultural direction change.
CULTURAL KNOWLEDGE
LOCAL KNOWLEDGE
There are two kinds of knowledge, culturally encoded knowledge, accessible to all people exposed to the sphere of influence of the culture, and private knowledge.
All private knowledge is reacted to right away hopefully - such that actors don't need their own memory? No.
Some kind of judgement is going to have to be applied to knowledge? No...
When knowledge is communicated to a third party they have to be able to estimate it somehow, such that their response is appropriate.
Theft of a wife, a member of an enemy raid found, the buffalo coming...
These may or may not trigger strong responses.
(Typically individuals are pro-active anyway, so in the absence of input they do anything (without constraint or intelligence)...rather than nothing.)
Responses may be based on prescient anticipation or on gut-reaction. MEMPOOL It looks as if each person must have a memory, and that that memory must be communicable to a larger pool of individuals (more clients get access to memory node) as person talks to other people.
(What about corruption of memory? -> communication in times of stress? Or indirect communication?)
Yes - its easy to support - contact w. an individual gives access to memories from before that contact (ie: individuals point to other individuals w. date)
(what if person is an enemy?)
if person is not an enemy.
(what about private memorys?)
hmmm...
Actors have primitives, archetypes which bring up their behavior these primitives respond to environmental stimuli & actor internal stimuli So actor behavior can be envisioned as an interface between environmental stimuli and internal stimuli - a mediating interface between these two junctures.
ok.
So what primitives -> well what excitement?
Culling followers
Basic areas of exploration
- politics
- science
- religion
- business
- exploration
Constructing a society of given beliefs
- what are the belief fundaments?
- law
- government
- reaction to other cultures
I imagine actors doing the following:
Going on great quests - with their followers <- motive, opportunity: natural and societal
Gathering the people and moving them to wars <- personal will, opportunity (enemys and discontent/wants)
Discovering technology <- personal will, base technology, need, raw materi als
Gathering people as a unit, a culture <- personal will, a void as such.
Killing dragons and monsters?
pogroms
create
destroy
influence
change
invent
physical
institutional change
social change
ruin enviroment <- be sensitive to
children
Degree of polarization of peoples depends on strength of circumstance and the other major influence type actors...
And each of their actions implies,
1) certain available pre-conditions, these preconditions come into
existance because of very tiny initial states.
2) certain willpower of that actor
Above that is a "vision-extraction-tool" which does two things:
1) Keeps a database of what is visible to an actors field of vision. This is based on contact with other actors and information flow.
2) Makes highly visible those things which are most attractive or most repugnant to an actor. This is kept as a pure list of geographic coordinates along with phenomena type per actor. Determination of visible phenomena varys per actor based on current actor desire statuses. Actor desires are determined both by absolutes and by belief systems, there is some kind of "hierarchy of needs" at work.
The question is how do beliefs motivate primal behavior?
And how much abstract/complex behavior do I have to synthesize?
On a slightly different vein as well, how much abstract/complex type input propagation do I have to synthesize? Ie: books / minstrels...
Actors will have families - or at least close bonded units whose relationships will be mapped onto familiar family type structures for the player. the evolution of actors will also keep in line with a family type structure. since the game starts with only 10 people the initial mapping should be very simple. initial family relationships will be something like "child of no woman" and i may spontaneously create people throught the game to continue having these kinds of things.or else i may have all people born to creches or something...
Actors can recognize things that they want (ie: loved ones). Actors have a memory of events they have been exposed to. Actors remember their homes and remember their daily chores. Smart-actors have a perception of the future.
Beliefs affect behavior, beliefs act as a filter on behavior and stimuli.
Whenever children are exposed to one another there is a potential for a transfer of beliefs <- usually occuring once, perhaps more, depends on richness of childs own beliefs.
Whenever actors have a chance to interact they may swap belief systems, this primarily occurs for disempowered children, and may depend upon a more-powerful-if-weak value assigned to the meme as well as may depend upon number of times child has been changed, as well as upon opportunity to practice such a system. should the belief be of the right kind and or strength then the child may leave home. typically the childs political colorations stay the same (ie: real coloration stays dormant) until opportunity is given.
Children who grow up in a belief system that has rewarded them will be less likely to change no matter what the stimuli - and in fact may act to change other people.
Actors are always sessile - that is they will not react to external events, theft of valuables, great opportunities etc. Their success strategy is to stay put.
There is the idea of the travelling muse - who when it latches onto a person it increases their charisma and intellect and actionability. these actors always act on external stimuli in a very strong way.
There are a finite number of muses in the game at any time.
Actors in general will be completely incapable of manipulation - ie they will not be able to act through subterfuge per se. In this sense they are going to be remarkably stupid - unless some special mechanism can be devised for such.
All surface phenomenon are going to be vector mapped, with all their state values resolved in floating point. For example: a corn field effectively exists over the entire planets surface, but only in some places are the amount of corn values anything but zero.
Basically there is a sample point array existing in the x,y direction the size of the entire planet surface, and some points of it have non-zero values. Now corn may have an actual structure per point, instead of merely a value (but we’ll try to keep it compressed).
There exists a memory pool for surface features.
The client can allocate chunks of a surface feature type.
The client can expand / decrease chunk size.
If possible the memory pool size should be such that for all non mutually exclusive phenomena - the largest total memory required is <= pool size. (This won’t be possible - but we can print out the difference). We probably also want to byte map the surface or something for quick scans.
Hmmm....to decide the best route we should make a list of all phenomenon, look at what is mutually exclusive, and try to see what we can optimize.
Basically what I’m thinking is having a number of "full size" arrays, and flagging into them a number of states.
I figure civilization shouldn’t be greater than 1000 people, so each person will need say 100 metres equidistant , so 10000 metres by a 1000 metres y should be adequate. now i’d like to resolve surface details to the width of a metre to make sure the 3d mode looks good. So that means that my array will be 10000000 (10 million) data points. I can scrape up lets say a megabyte of ram - so one million bytes.
Ok so we’re going to have to considerably downscale the size of the simulation.
Lets say 100 people max.
100 metres per person is 100000 metres square. 1 K wide, by 100 metres tall.
So these peoples world is the width of a football field???
Well we can downscale the people, range, eyesight, size...
Maybe we should have some form of scaleable architecture, or vectorized architecture?
Ok, lets say we design the world as a series of vector objects, each one mapped out to some low degree of resolution - where that map is perhaps imbedded in a shared field not mutually exclusive with other objects.
Each object has global properties common over the entire domain of said object. As well each object has access or has some portion of a bytemap for graduated aspects of itself that would be memory consuming to vectormap.
Alternatively we can fractally generate graduated behavior inside of a bounded region - just using a fixed seed. Possibly we can have enough control over the seed that we could deliberately select for very low graduations, or very high, etc.
So a farmer might have:
If a fire burns part of his crop then we would have to create a new vector zone for it - until it recovered.
This vector idea actually works very poorly for many things, like animals may defacate in large herds thus add swaths of nitrogen to the ground.
These kinds of things i’d like to map out by sample points versus vectors.
Also vector optimizing will be a bitch - what happens when two similar
vector objects don’t quite touch? I’d like to optimize for them but it seems like a bitch.
As a fire eats away at a forest i have to adjust the forest vectors, or overlay them with fire damaged forest vectors.
Well i guess vectors with some overlayed bytemaps is the best i can think of.
Vectors allow concept of region size, this makes for fast handling.
Vectors are precise - more so than a bytemap.
Vectors become dirty and must be