(I hope I was able to describe the problem rapprochement. As you can see, the theoretical matter is not my problem, I am logician and scientist, but the English language. I hope I could make myself understood and inspire.)

1. Thesis

I know of no game that is not deterministic. Although our human reality is usually not deterministic, humans creating games like simple machines, which become after a while predictable or linear. Let me explain this.

After a median level of play, games usually collapse. In role-playing games we gain enough objects, that everything getting boring or unusual predictable (victorious). Strategy games changes their organized towns or nations into clocks, which produce no errors from now on, which works like abstract factories. It seems that only the first 1/3 or 1/2 part of a game is unusual undeterministic, because of under development or organization lacks. In Civ there is a point, after a state or society becomes no more threats, no more riots. It is possible to build a well developed economy and nation (utopia) without inner state confrontation or systematic clumsiness forever, and it is nothing difficult to do so. For me, this is too far from realistic representations of communities or societies.

Although we know, that everything and everybody is lethal, and that everything dies or changes after enough of time, and that none money, none superior machine can create a solid state, we build games like abstracts clocks. And I think, this is the reason why today’s games seems only particularly realistic. It is a modern problem of development of large logical systems: and a strategy game is one of the most complex structures that can be abstractilly imagined, as it tries to portray a society after all.

As you maybe know, money and organization is only a small factor of a society. The Elemental game tries to imagine fantasy societies. As you see in German and European debates about welfare states and support of unemployed, there is a reasonable but unpredictable complexity of social systems, which you can not weight in money or deterministic evolution of production levels. Although a society spends millions of money, the people get unhappy. In manner of Niklas Luhmann (sociologist, system theory): We can not build systems that do not require changes. Systems change constantly from itself. It means: self-organization. All systemic change What is on the microscopic level change (dissatisfaction of an individual) may lead to a macroscopic change (overthrow of the king).

2. Examples

I am asking me, how a modern game like Elemental, which will get the same deterministic problems after all, could be more realistic, how to solve it in a realistic manner of “less deterministic”.

It is a mathematical problem. Think of complex adaptive systems (http://en.wikipedia.org/wiki/Complex_adaptive_system).

To establish the uncertainty of a realistic system, we need unknown factors which are capable of making a quiet system restless. These unknown factors can not be merely coincidental. You need to have microscopic causes, for example, that someone have to be an opinion leader, because he has a new idea (the leader of a new society).

When you look outdoors, you recognize a society which is (in manner of active police and firefighters and hospitals) restless, but also kind of pretty quiet (within an attractor of). Even you invest millions in this society, it could (but do not!) become restless and breakable and changeable. Your investment is indicative, for orientation, but not fully regulate. Deterministic systems, however, act as if everything would be fully adjustable.

I miss such logical and sociological circumstances in every game I’ve ever played. It is interesting to observe how developers deal with the problem of “wear and tear”. Usually you'll use “maintenance costs” to depict this factor. Houses getting older, trousers wear off, people are getting older, etc. But “maintenance” is too easy. For example: in Civ you getting after a while economical strength enough to solve every inner state issue. It could not be. Because in reality there are problems that arise, even if everything is quiet or well paid. It is called self-organisation of complex systems.

In most games I build a house that will never collapse, will be never out of date. It is simply there and can if necessary be replaced. But the house in itself caused no damage or disruption. This is a logical or sociological problem, because only a tiny number of systems in our real world are at all capable of such phenomena.

In games only the builder, the player, and a random counter (desasters) have the ability to change the deterministic perfection, if a player builds a new industry area or harbor. Then prices rises or decreases. But prices never decreases by fashion, or boredom, or unusual movement from the “inner logic of citizens” (self-organisation).

I think these models follow an old, classical model of the 19 century of complexity and mathematics and logics, where (important) changes take place only if they are influenced from outside.

It is the reason why all or most games are unrealistic and only realistic at the beginning, what means: they are only at the beginning a major challenge. Because at the beginning you, as a player, never knows everything or if, your “game society” or city is not fully operational. All your beginner operations are focusing to a well-done and superior production structure. And this is not realistic, because there is non super structure, because every level have it’s own reasons for descrease. And you're always confronted with the self-conscious of actors of your company or your system. This is the reason why complex systems are never deterministic.

3. Games solution

Now we could ask, how to develop abstract socities (games), which are capable of modern mathematics and processes of change, attractors, adaptive mechanics.

Really, I have not thought about it. But I think it would be good if the programmers would consider these modern aspects of mathematics or logic.

I don't know, maybe it’s useful in game

- to get older, not only main characters, but your citizens
- houses falls into disrepair, a house become “unrepairable” (too many fixes)
- society settings (culture, fashion, government) get “expire date” (old school)
- a king is only great as his followers does, it’s not enough to be born as
- and followers changes, with age, with (I don’t know) reasons
- or think of society changes which seems at first unreasonable
- unreasonable riots: some idiot found a new religion, new dogma
- or he tries to cut off the king (for personal reasons: to be king)
- and not randomly, but reasonable somehow by your actions

I think all this has much of a logically and mathematically problem of game design and to understand how a system evolves.

In most games there are only change effects by:

- player’s moving (reasonable factor, mostly unpredictable)
- randomly natural desasters (randomly, in manner of outside the system)
- other player’s (diplomacy, outside the system, by clash of systems)
- or sth. else outside the own society

As you see, these designs work with closed deterministic systems. Because the systems are boring, are predictable, they are only affected from the outside (also by the human player), but never by itself (self-organization). A “system“ means: a single nation in Civ or Elemental. This makes games less realistic.

Because the bigger the game society (my empire grows), the more deterministic the entire game world becomes (because I rules the outer effects). In this case we recognize that the model does not always correspond to our reality. Larger systems in our reality (huge nations, huge companies) often breaking itself, even no one extern contributes something and it is a lonely system, far from enemies. Or translated into computers: There is no computer system that would run indefinitely. At some point, somebody needs to come and share something, although a computer system seems deterministic (within itself: but is not true, and every technician knows the effects of hardware or software with bugs). It has mathematical reasons.

And I would like to know how we can make games more realistic and less deterministic.