Population and Demographics

[Richard Bruns]

I looked at Manurein's old demographic sub-model. It is a decent procedural model, but it deasn't have the flexibility of the cohort system. It does have the advantage of values that are easy to define, but I think that this system should be easy enough to work with as it is.

F_Smith:

I don't understand the complex stuff like enumeration and I probably won't for a while. So, let me try something simpler, like the method for determining the population of one of the later cohorts:

The cohorts after the first one have their creation method defined as aging from the previous cohort. Their population is the population of the previous cohort times the mortality of the previous cohort.

In the Cohort class:

public int aging(PreviousElement)
{
return previous.getPopulation * previous.getMortality
}

Could you correct that syntax?

By the way, I hope to start Java programming soon after I get my new computer next week. By then, I hope to know enough to try creating some simple tech model components on my own.

[Richard Bruns]

I haven't forgotten about this. I'm in the middle of studying the Java textbook so I can discuss the programming more intellegently.

[Mark_Everson]

Richard:

To avoid confusing you (I'm sure there is a typo), when F_Smith wrote:

public void doAging()     {
    return getNumberOfIndividuals() * getMortality();     
}

I think he meant:

public void doAging()     {
    number_of_individuals = getNumberOfIndividuals() * getMortality();     
}

since a void method in the first example can't return anything by definition. Although in my stab at it mortality seems to have the opposite from the usual meaning... So perhaps it should be -= to subtract mortality*pop from population.


[This message has been edited by Mark_Everson (edited August 17, 2000).]

[F_Smith]

Oops:

Mark is exactly right.

Sorry.

That's why God invented the compiler!

[Richard Bruns]

I now know enough Java to understand F_Smith's code and say that it looks good.

Is there anything else I need to do before the model gets coded?

[Mark_Everson]

Um, We don't Need any of this for demo 5. IMO getting the tech model going is Much more important. Unless you think this is a good 'training' exercise that's much more constrained than tech and so has value that way.

[Richard Bruns]

quote: Yeah, oddly enough the 'population' model, the very first, most important model that the entire rest of the game will be built on, was not done.

That was written by F_Smith in the social model. He started coding the Object Builder about a month before I did the population model. He seemed to like my population model, so I figured it would be put in.

By the way, I didn't know that F_Smith had written a population model when I made this. He never said that he had a population model when I created this thread. This would be yet another example of a communication problem.

Then there is the fact that the project lead says the population model is not needed and the social/govt. programmer says it is the foundation of the entire game. I am wondering about this. . .
[This message has been edited by Richard Bruns (edited September 16, 2000).]

[Mark_Everson]

Yep I'm aware of that inconsistency.

I think it is because F_Smith believes that to do the game right you need to OO model the real world. I believe that you simply need to OO model the 'game' world. I characterize the game world as an internally consistent view of the world in the game. The game world does not necessarily match the real world. In this example, I consider modeling just undifferentiated people as acceptatble for a game. He (and I think you) don't.

Whether the model is needed in the long run or not, I don't think its needed right now...

[This message has been edited by Mark_Everson (edited September 16, 2000).]

[F_Smith]

This is actually because of a disagreement between Mark and I about the basic code infrastructure/architecture.

The problem is, my experience in building similar data systems suggests to me that the existing architecture in the Demo 4 code would not work for a full game. It was not scalable, for a variety of reasons -- in my opinion.

Mark disagrees. The decision on the final code is his.

[Richard Bruns]

F_Smith: What kind of population model did you already code? How does it work?

[Xuenay]

quote: At the very least we should define what percentage of the population consists of minors who are not working or reproducing

Minors who arent reproducing? You mean that people wont have sex until they are 18? Now we're talking about something SLIGHTLY unrealistic!


(This was intended as a joke. You do not need to tell me that this would be too hard to implement.)

------------------
"Now Lone Star, you will see that Evil will always triumph, because Good is Dumb!"
-- Dark Helmet

[Richard Bruns]

Actually it will be implemented. The default five year cohorts will have reproduction and labor attributes based on social and economic conditions. So in some socities cohort #4 (15-19) will be considered adults and will be married, working, etc. But in other societies they will be considered minors or students.

But cohorts 1 and 2 (0-9) will never reproduce and will always be considered minors. This prevents the population from doubling every few years with no immigration like it does in some games.

[Richard Bruns]

F_Smith, could you tell me about the population model that is currently in the code?

[F_Smith]

Richard:

It's the 'ethnic group' object hierarchy we're discussing in the other threads. Should I also repeat it here?

[Richard Bruns]

I thought that the "population model" meant the system that defines aging, reproduction, and demographics. Do you have anything like that in there already?

This model has nothing to do with defining EG's. It just takes input from the social model to determine birthrates and economic potential. It can be run once per EG as an option, but it doesn't define their actions or status.
[This message has been edited by Richard Bruns (edited September 21, 2000).]