I vented some thoughts about engine games to my blog (visible on the main page).  It isn’t supposed to be an attack of your game idea.  However, I think that trying to create a game based on simulation is a long and tedious process that isn’t very fun.  You’ll probably have a lot more fun if you come up with a simple game with straightforward goals, and then develop the parts of your simulation required to implement the game. 

For instance, the goal of the game could be to have as many friends as possible.  Here is a quick game that Graham and I came up with:

• making friends or enemies with people affects your friend status with other people
• some people are more socially influencial than others
• some people are more socially competetive
• some people have unique skills, and can help you solve puzzles
• making friends is sometimes based on favors, which may include solving puzzles
• the end goal, which is time based, is to influence a important social event

• a "scholarship contest" like miss america
• an election
• a corporate takeover

I think it would be a fun game, and it is based on some straightforward, simple rules.  It could be implemented in a wide range of environments, from text based to 3rd person 3d rendered and networked.  In the process of writing it, you could develop your ideas of social simulation, for use in other games.  You would also have a code base that will make development of subsequent games faster.

First: will the simulation be complex and interesting enough to explore for any length of time? Probably not.

Second: even if the simulation is perfect, what’s the point? I.E. what will the player actually do if there are no goals?

So I think you’re right; that the game needs to have some goal and some game rules, but I think the simulation may serve as an interesting back-drop for whatever these may be.

Thanks for the ideas.

Kelly Gallagher