On the poll, I agree

On your 2 proposals, both sound good, but I'd need to playtest both to really make a decision on which is better.

All,

On the "political" vs. "negotiated" systems, I think LGJ has pretty much summed up my thoughts on the subject of what a real political system is like, I feel the negotiated system handles it better, especially when you take into account the other parts of the system, the riots and social models. IMO, you cannot look at the gov't model alone to decide if the negotiated system will work, you must look at the three models, gov't, social, and riot models as a whole (at a minimum, the gov't and riot models). If you just look at one model, it's like looking at only 1/3 of a very large puzzle and trying to figure out what the picture is, you might get it, but it's unlikely.

F_Smith,
btw, that's why I haven't posted any of my thoughts on what is going on in the beast right now, not that I haven't had a few peeks at it, it's just that IMO we're testing an unfinished system and the test results will be totally screwed up until the rest is in there.

That 'uncertainty' is the goal, the 'game' part of the political system.

When a player tries to change a government policy, that should be an uncertain thing. No 'automated' governments. The player shouldn't know for sure if a policy will pass, just like they shouldn't know for sure if an army will win a battle.

• You order your special police to bribe a leader opposed to you.
  
• You call for a vote.
  
• You then wait until the votes are cast to see how well the bribe worked.
  
• You realize this character you bribed has an honesty of 0, and took your money while voting against you.
  
• Next turn, you order a hit . . .
  

So it's a govt 'game'.

I think this could also help make the game a bit more interesting during those long turns when nothing much happens for a long time, which any game that sweeps thru centuries is going to have.

* * *

Lordy:

The 'politics' system is designed more along the lines of 'politics is about power'. It is built on the principle of 'coalitions'.

It seemed that '51%' has a special meaning in politics. That *is* total control. You still have to deal with the unhappiness of those out of power, don't doubt it. If you only have a slim margin like 1%, then any power play by another group can rob you of your majority. You'd need at least 60% to be secure, I'd think. Maybe more.

And hopefully the 'politics' system is designed to encourage a player to accept *less* than 50% of the power, leading him to more 'modern' governments. Because you can still build a coalition to pass changes you want. You wield the secret police, bribery funds, and the power of the military.

* * *

Ya'll:

I do enjoy chatting about this, but I do want to repeat that this is only one option, and not the default system. So I'm not sure how 'on topic' this rambling of mine is.

If possible, do still post any feedback that occurs to you when you glance at the beast.

You'd be amazed at how fast those eyeballs move the development of the program along.

The reason for ya'll to look at it is to keep me on track. Like keeping an eye on a builder to make sure he builds a house exactly to your specs. You may not learn about how the system really operates until much later, true. But you will learn how the 'game' works, which is almost more valuable.

Ya'll are 'debuggers' (we call them 'Business Analysts', or BAs). You glance at what I, your programmer, am putting out once a week or so, and say 'I like that. Change that. Dump that.'.

If ya'll don't, then the software will not get written the way *you* wanted it to. It'll get written the way *I* thought it should have been, or the way *I* thought you meant it to be. And this conversation should show you that programmers like me see things *very* differently.

Can you imagine taking something as complex as these pages and pages of models and 'writing' thousands of lines of prose to describe them *exactly* to a computer . . . in your spare time, as a hobby? I'm sure I've screwed up a hundred time just last Tuesday.

Ya'll better keep a close eye on what I write. There's no telling what I put in at 3 a.m., after that 10th Mountain Dew. The caffine can make ya nuts, I swear . . .

Besides, this is not just true of me -- it's an industry joke.

• Client requested a bicycle.
  
• Salesman sold them a motorcycle.
  
• Architect designs a unicycle.
  
• Programmer codes up a space ship.
  

It's kinda funny how close to home that analogy of the builder and the architect hits...I'm in school studying to be a draftsman, of which architecture is a major part.

But the thing is, if you don't know already, I was developing a new social model while Rodrigo was working on the gov't model, and we worked closely together. I had to abandon the social model because of my job and school, so I already know the inner workings, although they changed somewhat, it's fairly minor, so if you're unsure what I think, like I said, 99% of the time I vote with Rodrigo.

IMO "how the game works" is not much more than a very well laid out, easy to use, interface, and "how the system operates" is what we've been working on... the models. I just personally cannot conceive of testing the gov't model until the riot model is also in place (given my knowledge of the inner workings). For an example, I played with the Beast and set everything to polar opposites of what it was at default, and got things like, tax rate = 84% when it was 43% the previous turn. This is very discouraging taken at face value. But considering there is no riots model in place it may turn out to be ok, we won't know until then. That's why I said it's an unfinished system and all results will be screwed up. With that said, there are some things worthy of being tested at this stage (like how the procedure unfolds), but the end result of policies is not one of them.

I understand that turning these paragraphs into code is no easy task, especially as a hobby, but IF we can get the code to behave the way the equations behave in mathematical tests (and please don't underplay these tests... mathematics is the basis of nearly everything nowadays, and if it works there it can work nearly anywhere), I think we'll have the perfect system.

No, not testing the govt model -- testing the govt code. I need you to be a 'Business Analyst', a 'debugger' -- all of ya'll, from time to time. I'll post 'Test Case' threads, with specific instructions. Shouldn't take long to do, then please just post a quick note saying you did it, ran fine/lousy/fast/slow/etc.

Hopefully, that's what's going to seperate us from all those other teams out there that came up with grand demos, treatments and design docs but never produced an actual program. We're going to all work on and develop actual game code. This means we have to work together. It's the only way to do it.

And I must protest, if you're designing logic for a game program, I'd really like to think you would benefit from knowing something about how the game itself works, at least the basics (it's far more than 'just an interface', please believe me on that). At least take a look at the basic datastructure. That's the 'frame' of the game.

* * *

Believe it or not, we've actually got to design code that works better than the equations. You can't code a game like this procedurally, in one lifetime. You'd go mad trying.

I do not like the uncertainty of not knowing if something will pass or fail. That makes the game chaotic and requires micromanagement. IMO all of the uncertainty should be in the consequences of the action. I think that the game should only let the player propose something that will pass. The reaction to the policy is not defined clearly, but the actions available to the player should not fail.

I, as a player, do not want to micromanage the foul play. I want it to implement automatically. If my agents can't bribe someone, they should be able to automatically go to the next step of arresting the person. If that doesn't work, they should order hits. They should keep doing things until the measure passes, and then the player should deal with the consequences of the foul play. The foul play is calculated behind the scenes, and the player should deal with the results of whatever happened. Leaders of countries do not generally tell agents exactly what to do. They just say "make them cooperate by whatever means necessary."

I would like to propose a new slider color scheme:

Green means that almost everyone will support the measure and there will be very few negative consequences.

Yellow means that the measure has at least 51% support and will pass, but that it will make some people unhappy and there will be a few more problems.

There should be a gradient from Blue to Yellow, indicating different shades of popularity. For example, 100% popularity would be dark green, 75% popularity would be yellow-green, and 50% popularity would be bright yellow.

Orange and Red mean that foul play is required. The shade of the bar shows approximately what degree of measures are required. Pure orange means only a few steps need to be taken, and dark red means that extreme measures are required.

Black means that nothing can get that policy in place. Those regions of the bar are not options.

I'm glad to see a lot of interest in this thread. Please all of you keep saying what you think.

TK: Even if you agree 99% with me, I think it's very important to help F_Smith giving him feedback for the beast. Even comments on very tiny things are very useful. Please do it everytime you can.

Mark: You say you don't care much about what's legal and what's not, but I think it's good to have them differenciated for flavor. It's cool in game terms to do things behind the eyes of the public risking to get cought... and even cooler if you get away with it!
It also makes clear to the player that in some regimes he cannot do whatever he wants and there're things he is supposed to not do.

Richard: You made comments about how informed the player is when choosing his preferences. It's my mistake. I never said anything about that. All along what I envisioned is the player having a window where he can browse the political status. As detailed as checking what each class wants for each particular policy or more aggregated info like what ideologies are more supported or an "overall" (aggregated) people's preferences for policies. I understand now that part of the discussion around colored sliders had to do with this type of info helping the ruler in what to decide. Since I was always considering the ruler had info, then I took the colored sliders proposal only as a matter of what's possible or not for the player in terms of how despotic he is. Just another communication problem! What's possible and not and what's popular and not, can IMO be integrated in the same interface if we want it that way. All the info needed to take political decisions can be as sophisticated as we want (the model data can be analyzed in a lot of ways), so for me the main issue here is how policies are set assuming a well informed ruler.

Policy-setting methods:
I agree with almost all LGJ has said about politics. That's the way I understand politics too. I think it's possible to have two or more systems for setting policies in the game and let the player choose the one he/she likes the most. But, in a general perspective and beyond this particular part of the game, I believe it's good getting an agreement on how each part of the game should work. It helps a lot having everyone looking at the same thing and making an effort for that thing to become fun and realistic instead of dividing efforts just to let each one of us have his very particular point of view implemented. It will save us time and coding effort.
The beast is a good place to test different methods and I really hope it will help us choose one.

F_Smith says IRL policy setting doesn't happen via negotiations and the 51% rule is really everything. In his own words, "Once they have that [51%], they don't need the other groups with political power, and can ignore them. [...]People with minority power in govts get over-ridden [and] they have absolutely *no* effect on the final results that come out of Congress"

It's of course a matter of opinions. If that's your vision of how real life works, there's really no much I can do. I see things differently and maybe all I can do is try to convince you a negotiation system is more realistic... can I try? Let me try!

The 51% rule IRL can or not be total control. It's not a fundamental law of nature. Maybe the US senate works like that. I'm not that familiar with it to know it. But there're several other systems. The 51% rule is the "absolute majority" rule. The "simple majority" rule is another. The "quorum restrictions" is another, where things can be changed only if, FE, 75% of all votes support it. The list is long and we can also include other features like the ability to veto a proposal. Even more, in many countries more than one of these systems is used depending on what is being decided. So, the 51% rule is just one system out of many and shouldn't be seen as THE law by which politics must be driven.

We can of course simply choose one of these and work with it, but if realism and flexibility is what we want, then it doesn't sound good to me to take how a particular nation works and apply it to all civs. It would be more general and realistic IMO if we can implement a system where interactions take place in a more abstract, less precise fashion.

You say it sounds unrealistic if a ruler having 80% power couldn't implement anything he wishes. That's only because you're looking at it like if the "absolute majority" rule was the *only* way to go. That's understandable if that's the reality you face in your country, but you have to admit you're just being "blinded" by it. It'd be like a muslim fundamentalist saying "Islam is the only valid religion". We know that's not true.

Your statements imply a system where a person or coalition can take just two status: "in the ruling body" or "out of it" leading to two levels of control: "total control" or "none at all". I don't believe reality is like this. Democrats in the US have had less then 50% seats in congress many times, yet the country has never taken a fully republican shape. Democrats have managed to stop laws and other inniciatives in these situations, which tells me they in fact have some sort of control over what's happening.

The all-or-nothing system (law passes or not) doesn't recognize real life stuff like agreements between parties. In countries with a tradition of multiple political parties, interactions like this can take another level of sophistication: "Your party wants to increase tax rate by 5%. Our party will vote against it and you won't get it. But if you change your proposal to just a 3% increase, we're willing to support you and, together, we can make it happen". Or "our party will support you in this inniciative of yours if you support us in the voting for that other inniciative we're pushing". Democrats tend to get a little bit republicans and vise versa to make the machine work. For any given time, the US has always been partly democrat and partly republican and this tells me a sort of "mixing" between "pure" preferences is taking place instead of the "take this OR take that" outcomes the elections methods produce. Even if the country in fact works with elections!!

This mixed shape laws can take is practically unrestricted. In january we had elections in Chile. The new president comes from a coalition of parties ranging from socialists to center-right. It's pretty obvious that for this coalition to be possible, a common agenda had to be created, which means the "coalition agenda" is something in between the original "pure" preferences of each party. And that's, of course, the main direction this country is going to take in the following years. The other politically important block in Chile is the right-wing coalition. They didn't win the elections, but yet they have an important amount of seats in the senate. For some laws to be changed it will be imperative for the in-govt coalition to gain at least part of their support. And to achieve it, these laws will have to be presented to the senate somewhat more right-winged. So each side, no matter how different they're, has at least some level of influence on what the final "policy values" will be.

Laws are not simply passed or not, politicians make counter-proposals, suggest minor changes to inniciatives, accept them if others are changed, etc.

That's real politics. Contineous interactions, negotiations, arrangements, etc. It's really unnecessary to know if the actual detailed system is a 51% rule voting or what, because as a whole a govt can be seen simply as the interaction of actors with preferences and powers.

That's for democracies. What about despotism? Recently in Venezuela the new president Hugo Chavez closed the senate and in its place he putted a "Constitutional Assembly" whose mission is to re-establish "the institutional order" or something like that. To the eyes of all us it's just a move toward some sort of dictartorship, but the point is, even with that, Chavez is not as despotic as, FE, Stalin was. This is just an example showing that "degrees of despotism" should exist in our game. We can find a lot more examples, like the slowly decreasing level of despotism England's rulers experienced in the last centuries. And I feel this things are attractive for a game like this. Then, it doesn't sound good to me if a ruler with 80% can do the same things one with 100% could do. Someway we must handle these differences for extra flexibility, realism and fun.

I'm not defending the negotiation procedure particularly. All the above has as a goal simply showing why I don't see the "51% rule" system as realistic and I used several real life examples to sustain my thoughts. I believe it runs too short to what is desired and could only be called realistic in a very very particular political system. I hope some of that convinced you, F_Smith. If not, well, I had to try!
---------------------

My inclination is to use a system where all the interactions are modeled in a general and abstract form for simplicity, realism and generalization (usable in a democracy and in a divine monarchy, FE). The negotiation procedure (in its lattest form) is one possibility, but I'm sure there're others and this thread is indeed showing other roads can be taken. I saw Richard's idea and it looks interesting, but the fact that negotiations only happen sometimes (and only to correct a weird effect Richard himself acknowledges as being present)is IMO not good because all the rest of time you lose all the good things of interactions between actors, not to mention the all-mighty-ruler in all the ruler's pol.power range 51%-100%, which is a characteristic derived from using the 51%rule.
-----------

The slider is intended as an easy way for the player to see what kind of coalitions can be formed to get something passed. At one glance, the player can see what values can gather a coalition support.

As I outlined above, the colors of the bar represent the amount of support, or the size of the coalition that can be put together to support something. If you can get parties (or classes) with a total power of 80% to agree on something, the value is green. If you can get a coalition with 60% of the power, the slider is yellow. If you currently have a coalition of 40% and have to play dirty to get the other 11%, the slider is orange. If no one will support you and you don't have enough power, the bar is black and you cannot try to implement that value.

That color scheme works very well, I like it.

But -- if it's okay, it can not include the effects of 'variable' actions, like 'foul play'. They have to be 'variable', or unpredictable. Because that is the 'game' part that seperates this from an 'equation' -- I *want* uncertainty in the outcome of the vote. You make a move, you wait to see what happens. Just Like I want uncertainty in the outcome of a battle.

I don't want my 'people' to automatically escillate to a hit, or an arrest. I want to make the choice of how far to push, since each action will have different effects on society. Bribery would be relatively safe. A hit, a little more dangerous. Outright arrest would be critically dangerous.

Understandably, if you don't want to 'play' politics, if you don't want the outcome of a vote to be uncertain, you probably should leave the default 'automatic' system on.

The 'politics' game is designed as a game within a game, with it's own uncertainty and management requirements.

* * *

Rodrigo:

Yes, that definitely mirrors my observation about all political structures (like governments, guilds, etc) from the beginning of time -- just like your elections.

The power brokers -- your 'coalition govt' -- (those who hold the '51%+' power) negotiate amongst themselves on acceptable policies. Then the smaller groups with power make their wrath known in a 'reactive' fashion (unhappiness, lower productivity, riots, political demands, etc). Then the power brokers either take them into account next time, or decide to just live with the results (or 'fix' things in *other* ways?).

That's really the only difference between the default system and the 'politics' game. The negotiation in the default model happens automatically.

The negotiation in the 'politics' game is in the player's control.

The default system assumes everyone negotiates with everyone, and that everyone gets a seat at the table.

In the Politics game, the player decides with whom to negotiate, and whom to stiff. Then they deal with the wrath/effects of stiffing that group. If you want to negotiate with everyone, making a utopia in which all groups have a say, you are free to set the policies at that 'average' value. Altho, in this game, it won't always make everyone happy . . . people who want a 20% tax rate are not usually going to be 'happy' just because they managed to 'negotiate' you down to 50%. They may still demand a larger share of the political power, or threaten to resist/revolt.

The default system doesn't give the player a choice -- everyone with a vote gets a say.

The Politics game puts the player in the seat of that President of yours, who has to build a coalition just to hold power. Think about it -- the Politics game would let the player play that role, making politics a fun, simple yet deeep game in it's own right.

If a Prez/Player like yours wanted to pass a policy, it would have to be one acceptable to all his 'allied' groups. Sometimes you could also 'bribe' some support out of the opposition (trade backroom promises, etc), sometimes you decide to take more extreme measures . . . like Venezuela has.

Oh, and there are many forms of despotism available in the Politics system. 51% is *very* different from 75%. 90% is very different from 100%. The political outcome is not different -- both can pass just about any policy they want -- but the 'social' costs are enourmous.

Also, you actually bring up a great point, about being able to vary the % needed to pass a law. That is a definite enhancement to this politics system that should be included. A 'Political Body' -- 'Council', 'Senate', whatever -- can give a civ the power to control it's govt in that manner. Each govt who discovers the 'Political Body' tech and built a working 'Council' can control that % needed to pass laws to some degree, so some govts could be very hard to change policy in.

So beside the slider, there could be a botton that says "foul play limits." The button opens a window where the player determines the maximum escalation of foul play. The player could tell the game to stop at bribing, or set foul play to unlimited. The game would start with the most benign actions, and if they don't work it would keep implementing more harsh actions until the objective is met or the player's limit is reached.

Cool Idea: Sometimes overzealous agents might go beyond the player's limits. Then the player would have to deal with the consequences their actions.

The ruler will be encouraged to take a 'minority' position in the govt because groups' 'happiness' will be directly related to their representation in the govt. And Happy people will perform/produce more. So the only way to have a top-flight econ is to give the people a big share of the power.

And with the 'politics' game, the player can still have some decent amount of control in a more 'modern' govt, in which the ruler is in the minority. With the 'default' system, the ruler with minority power will have very little impact at all.

For the 'Middle Ages King' you mention, in real life that king passed laws. As long as he controlled the military/police, the laws were enforced. As you said, the Church could 'excommunicate' him, re-acting, trying to force their will on the player. But they can't force the King to change a law or policy. It is up to the king/Player.

One other thing -- the 'default' system doesn't seem to work at all for groups with a definite opinion, as most groups in real life seem to have. As I said, if a ruler with 50% of the power wants an 80% tax rate, and the people with 50% of the power want a 20% tax rate, the people are *not* going to happily compromise at a 50% tax rate, as they will in the default system.

* * *

Richard:

I suppose we could have that as an optional GUI. But again, if you take away the uncertainty, there's not much point in using the politics game anyway.

I think we've stumbled onto an interesting point here -- there are, broadly speaking, two kinds of 'games' we can make here.

1. A pure 'sim'. A series of equations that takes inputs and generates outputs. The player alters inputs to get different outputs.
   
2. An 'uncertainty' system. The player has a choice of several actions, each of which may or may not work as planned. The player chooses an action, then watches the outcome.
   

quote: Originally posted by F_Smith on 08-31-2000 12:59 PM The ruler will be encouraged to take a 'minority' position in the govt because groups' 'happiness' will be directly related to their representation in the govt. And Happy people will perform/produce more. So the only way to have a top-flight econ is to give the people a big share of the power. And with the 'politics' game, the player can still have some decent amount of control in a more 'modern' govt, in which the ruler is in the minority. With the 'default' system, the ruler with minority power will have very little impact at all. For the 'Middle Ages King' you mention, in real life that king passed laws. As long as he controlled the military/police, the laws were enforced. As you said, the Church could 'excommunicate' him, re-acting, trying to force their will on the player. But they can't force the King to change a law or policy. It is up to the king/Player. One other thing -- the 'default' system doesn't seem to work at all for groups with a definite opinion, as most groups in real life seem to have. As I said, if a ruler with 50% of the power wants an 80% tax rate, and the people with 50% of the power want a 20% tax rate, the people are *not* going to happily compromise at a 50% tax rate, as they will in the default system.

1. The government may be encourged to lower his power for happiness level, but lowering to low can cause sluggishness which causes unhappiness in the people and they want someone to take more control to get things done (Julius Ceasar).

2. The ruler, if he has atleas the minimum of what is needed to pass laws or whatnot doesn't need to compromise nor worry about swing votes because he can still control the outcome. That's why the player will try to micromanage his stats to keep them at 51+ or whatever is needed. They can have otherways of keeping people happy and whatnot.

My view is that if a player really wants something to happen, that player should be able to get the thing to happen. For example, suppose that I really needed to raise taxes to support a last ditch defense of a valuable province. I wouldn't want to micromanage a lot of petty politics and run the risk of my vital tax increase being rejected. I would want to set foul play to unlimited and run the tax rate up into the red zone. I would want to make sure that I get my money.

Of course, I would still have to deal with the consequences of my actions, and those consequences are uncertain. The people might agree with only a little persuasion, or they could all get really mad. But if I decide that keeping that province is more important than making my people happy, I should be able to eliminate the possibility of not getting the money.

The ruler can indeed fight against giving power to the people. The player can keep the 51% in his hands. He'll see his civ outpaced by more 'free' countries, in production and research, but if he'd rather keep the reigns of govt for war purposes so be it.

And it actually would be a simple thing to track a group's 'opinion' of the ruler. That sounds really neat. I think I'll use it.

* * *

Richard:

If a player does exactly the same thing twice exactly the same way, should the result be the same?

To me, that's a mistake. I don't think it should.

I want a game in which the outcome of each turn is uncertain. That's what I mean by 'uncertainty'.

Remember, the seasoned player is going to know *exactly* how their decisions will turn out, in the 'default' model. Because they've played it a dozen, or hundred times. So there will be no uncertainty in their end results. You do this, it makes the people unhappy, you do something else to make the people happy again. Repeat.

This is what I meant by you can't judge rodrigo's system without the riots model:
Quote from you, "One other thing -- the 'default' system doesn't seem to work at all for groups with a definite opinion, as most groups in real life seem to have. As I said, if a ruler with 50% of the power wants an 80% tax rate, and the people with 50% of the power want a 20% tax rate, the people are *not* going to happily compromise at a 50% tax rate, as they will in the default system."--

quote: Originally posted by F_Smith on 08-31-2000 09:50 PM The ruler can indeed fight against giving power to the people. The player can keep the 51% in his hands. He'll see his civ outpaced by more 'free' countries, in production and research, but if he'd rather keep the reigns of govt for war purposes so be it.

Also in modern day look at China. Barring the recent collapse in the asian stock market which also affected democratic countries like japan, China was outpacing every other country in the world as far as how fast its economy was growing.

On the other hand, there's Russia, a democracy that is in many ways doing worse than it was during the cold war (some progress in some economic areas, but on the whole not).

Actually, right now, the people don't have any say about taxes at all. That's been left for later. Even the he ruler just sets any taxation level they want, in the default game.

I've been thru the riots model pretty thouroughly.

Unless I missed it, the 'riots' model (which should actually be renamed to something like 'happiness model') doesn't cover taxation levels. At least as of now. It makes allowances that later, the Upper Class may have a reaction to taxes, but that's it.

* * *

Lordy:

It's not 'arbitrary'. We want to encourage a ruler to develop governments like the ones that have actually come about, don't we? If we build a govt game that encourages people to keep all control in their hands, we'll end up with a game dominated by dictatorships.

Then we'll have failed to come up with a 'realistic' model.

I disagree strongly, I believe that all other things being equal, freedom produces happiness and productivity. When Germany did well after WWI (I assume you mean), it was a free country, a democracy. Japan never matched the other nations in production or creativity, when they were an Empire. China is experiencing a boom because they have instituted free market reforms. And Russia, in spite of pretending to be 'free', still is clearly in the Iron grip of the same powerful people. Putin's secret police still storm into Media offices with no cause (other than being unhappy about the coverage!).

Yes, a player could still run a happy, productive Dictatorship. It's just not going to be the most productive possible society in the game.

* * *

Richard:

But even including the 'riots' model, there will still be no 'uncertainty of outcome' once you know the rules. If you do discriminate against people, they'll get mad. If they get too mad, they send a note. If you still ignore them, they rebel.

It's designed to work like an equation, not a game.

I assume you've played Caesar 3. Or Sim City. In those games, it's a matter of learning the perfect order to do things in. Or memorizing the perfect order of moves to accomplish a task. You do this, that always happens.

A seasoned player will know exactly what the outcome of every move will be. The only 'game' part is in the combat/competition with other players/civs. If an action can never fail, the only 'uncertainty' is in the outcome of the chain of events. Once you learn the chain, the outcome is no longer uncertain.

My GUI has exactly the same randomness as your system. This is because it is based on your and Rodrigo's systems. It has no new rules, aside from automatic escalation of foul play.

There is no way to predict what actions the goon squad will have to take. Their actions are as random in my GUI as they are in your system.

Here is the only difference:

In your system, you order a bribe. If it works, the measure passes, and if it doesn't work the measure fails.

Results: X% chance pass, Y% chance fail

In my interface, you order a set of covert actions. If the bribe works, the measure passes and people are not mad. If the bribe does not work, your people order more extreme measures, the measure passes, and people are unhappy.

Results: X% chance pass without unrest, Y% chance pass with unrest.

Note that X and Y are the same in both cases. That is because my GUI uses exactly the same equations as your system. Your system and Rodrogo's system are the things that are powering the interface I am proposing. So the GUI has exactly the same randomness as your equations.

And players will have the option to prevent escalation to the more extreme measures. So your current system is a special case of the GUI.

Actually, the 'automation' of the 'foul play' is removing a large part of the 'randomness'.

First, you won't know if the bribe 'fails' until after the vote, after the turn. The bribee will take the money, and promise to support many times. Whether the bribee actually *does* support you is not predictable, it'll be random, based upon honesty levels, the amount of the bribe, etc. So if a bribe fails, the solution might not be escalation to more serious foul play. Maybe just try a bigger bribe next time. Or try to bribe a different leader. Or a different group entirely. Or maybe you want to put the hit on the guy for taking your money.

But it *has* to require a decision to be made *by the player*. It *can not* be automated. If it's automated, it's a puzzle, instead of a game.

If a player's action fails, the player must be in charge of the decision on how to proceed. There are choices to be made that should not be automatic.

We really have different ideas of how to make a game.

Ya'll seem to be after a 'sim' game. My idea of a game goes beyond that. Fortunately, we can put both sets of ideas in this game, if ya'll don't mind.

Again, I assume you're a veteran of Sim City, Caesar 3, etc. Have you played Railroad Tycoon 2?

Those are 'puzzle' games. What ya'll are focused on building here is another 'puzzle' game. The 'game' is played by learning the complex 'if-this-then-that' interactions of the system. Once you 'figure out' the puzzle, the game gets repetitive quickly. It's a matter of 'build orders'. If you play a scenario building exactly the same things at the same times in the same places, then the outcome will be the same.

I think its a pretty clear historical fact that free societies do better in the long run. Dictatorships can be better in the short run during wartimes, or when 'catching up' economically. However we could argue this forever...

The point I would make in terms of gameplay. Dictatorships clearly offer the player more scope for action. If there is no penalty for this, which is what LGJ seems to be suggesting, then everyone will want to be a dictator. Is That a fun game? Why would a player Ever want to put up with all the crap involved in being a democratic leader if there were no advantage? So I think, along with F_Smith and probably many others, that we should include penalties for dictatorships and authoritarian governments in tech advancement and many other areas. The size of these penalties should of course be determined by playtesting.

Randomness vs Determinism:

If the player can always predict what can happen then the model is a failure IMO. But I think the deterministic riots model is so complicated it may defy prediction anyway! I don't think determinism is such a sin in a complicated model, because the chance of Ever being under the exact same conditions is vanishingly small. If it is predictable, then I would vote strongly for putting randomness in it. But that is easily done if we decide its needed.

If the government model is so twitchy that the player constantly needs to have their hand on it, then its against our whole design philosophy. We want to reduce the need for micromanagement while enabling the player to drill down deep into each area if they like. The colored slider system is just one way to handle the reduced-micromanagement part of the scale.

Of course if the AI can handle the twitchiness ok, then there's no problem. But a very nonlinear system like the 51% one is Very difficult to do good AI for. Besides, the 51% system is deterministic Also. So I don't understand completely the claims that one is predictable and the other isn't.

Even in a complicated model, determinism creates predictability. Because after you play the game a dozen times, you have seen all the combinations.

I'm sure you've played Caesar 3 . . . very complicated model. Very predictable, repetitive gameplay. It was a fun puzzle, for about a month, tho. Then 'build order memorization' sets in.

Railroad Tycoon 2, same story. StarCraft, Age Of Empires, and on, and on. It isn't the complexity of the model, it's the absolute *need* for randomness.

That's what I've been trying to get here. The current models are so predictable, in fact, we are going to use colored sliders to tell the player what the outcome will be before they ever make a choice. We're actually able to pre-compute the exact outcome.

The Politics game is *not* deterministic at all. You make the exact same choices, they will turn out differently. The bribe will work sometimes, but not others. Sometimes arresting a leader will suppress his supporters, sometimes just the opposite will happen. Randomly (within certain parameters).

That's what Richard so objects to, in fact. He wants the player to be able to know the exact effects, to know his tax increase will *always* pass.

How many times do I have to say that the sliders are exactly the same thing that you proposed? They do not make a new system. They simply put an interface on your equations. They use your system of computing things. The sliders do not tell exactly what will happen. They simply show how much support something has. Green means that a certain size of coalition supports the thing. Orange means that a certain percent of the power supports the thing. They do not predict what will happen. They simply give the player information and serve as a quick interface.

Every time you are criticising the sliders, you are criticising your own system. The sliders are based on nothing other than what you described.

The thing you seem to object to is the automatic change in foul play. Let me try to explain why this is not a new or different thing:

If the player really needs something, that player will not tolerate failure. Suppose that I am playing a game and need to gat a lot of taxes to support an emergency military buildup. I will want to raise the most taxes possible while minimizing the amount of unrest that results.

If I did not have the automatic escalation, I would do the following:

1) Save the game
2) Set the tax increase and try the least severe action.
3) If that didn't work, reload the game and try a different action.
4) If that didn't work, reload the game and try a more severe action.
5) If that didn't work, go to step 3.

Players WILL do this. If the need for the proposal is vital and the country is in danger, they will keep trying and reloading until they get the increase with the minimum possible consequences.

The automatic increase in severity is simply an automation of this process. It is exactly the same thing that the player would do anyway. It turns an annoting and time consuming process into a quick calculation. It does exactly the same thing that your proposed interface does, but it does it in a way that causes the player less grief.

You have a good point about not knowing the outcome until after the vote. But why must there be only one vote per turn? The automatic escalation assumes that you can call many votes over a period of one turn. If the bribe in the first vote didn't work, you can submit the proposal again, pretend it is different, use different underhanded measures, and vote again.

I am only disagreeing with the automation of 'foul play' escalation. As I said above, the sliders would work for just setting a policy, if you want them.

The player will *have* to tolerate failure. The possibility of failure is a *must*, if this is going to be a 'game'.

Yes, in single player games, some players will 'cheat', when they couldn't get what they wanted. When they fail to get a tax increase thru at a critical time, because they messed up and have to have it now, they will reload. That's their right. They lost -- but they can cheat instead of starting over. If someone wants to cheat the game rules because no one is looking, so be it. And if someone playing Chess against themselves wants to move pieces around, that is *their* right. If someone want's to cheat while playing solitaire, who am I to complain?

They couldn't do that during muliplayer games, tho.

And they can't say they 'won' the game.

Besides, the increase in severity is *not* automatic. It can *not* be what the player would do, because you don't know what the player would do. Or rather, each player would do something different. The same player might do different things, depending on mood.

The player *must* first find out if the first action was successful. That requires a turn ending, unless we're going to begin allowing multiple actions in a turn, which is a possibility. Then the player *must* select from several possible choices -- live with the results, repeat the bribe hoping things turn out differently, bribe using more money, arrest the fool, kill him, etc. There is no way to know which to 'automatically' do.

That is my idea of a game.

A choice must be made by the player. You wait until the turn is over to see if that choice worked, then you make other choices.

If you automate making those choices, you remove the 'game' part.

1) F_Smith wrote (29Aug/09:36):

quote: Yeah, I believe Axi is talking about the 'econ' model. For the govt-system's purposes, 'social classes' don't matter, except as a provider of input values. They're just 'variables', so to speak. I'm just trying to understand what page he's on, so I can move in that direction soon. But no, none of this is getting coded yet.

2) About the "Politics" procedure:
Richard wrote (29Aug 14:44):

quote: I have an idea for the interface. Hopefully it can combine F_Smith's and Rodrigo's systems so everyone is satisfied: Initially, the allowable range on the slider is any policy that can get support from any combination of groups with at least 51% of the power. .... The previous part of the proposal used F_Smith's ideas, and while being more accurate it can lead to thorny problems. The second part of the system is as follows: The player can ask for negotiations. This would most likely be done when the other classes' ranges do not overlap. It would have to be done if no proposal is supported by 51% of the people. ....

quote: ...Of course the PC ideology doesn't dominate all the time. There are the following possibilities: a) PC Ideology > 50%. Then the player can choose between inforcing his policy as is and finding collaborators to increase the consensus. b) Another Ideology > 50%. Two cases: i) Dii) D>Dc. The distance between the two ideologies is unbridgeable, so the PC loses the struggle. The dominant ideology is the Govt policy. c) No ideology > 50%. Then the two ideologies that share the smallest D collaborate. If they do not exceed 50%, the next bigger D is checked and either a third ideology is added to the group, or another couple is formed. This formation of groups goes on until: i) Dc is reached without any group exceeding 50%. Then a Universal Consensus Government is formed, with the participation of all Ideologies. ii) A group exceeds 50%. Two cases: 1) If the PC ideology is in the dominant group, the Govt policy is the weighted sum for this group. 2) If the PC ideology is not in the dominant group, the case is similar to case b).

F_Smith wrote (31Aug 00:04):

quote: Also, you actually bring up a great point, about being able to vary the % needed to pass a law. That is a definite enhancement to this politics system that should be included. A 'Political Body' -- 'Council', 'Senate', whatever -- can give a civ the power to control it's govt in that manner. Each govt who discovers the 'Political Body' tech and built a working 'Council' can control that % needed to pass laws to some degree, so some govts could be very hard to change policy in. The base 51%+ is clearly just a simplification, based on pure 'simple' power politics, as exists in most human interaction. As it was played in human societies for most of our past. I like it...

quote: What a negotiation procedure of any kind is always lacking, is what I have mentioned before as "concentrisation". I'm not sure if the term is valid, but I'm referring to the kind of political system where some sort of "majority rule" applies. It is like the various power groups are gambling their share of power, in hope of gaining it all. As it is generally advertised, this is due to the fact that it is impossible to make realistic policies through negotiation, because of the existant ideologic distances. In reality, this is just another way for the mighty to steal the power of the weak. This doesn't ness need to be the "absolute majority" (50%), but also the "relative majority" (biggest %) and all variations where the "critical percentage", or "Consensus" is different than 50% (f.e. in the Greek Parliament, to pass a law you need 50%, to elect the president you need 60% and to revise the constitution 75%, while in the general elections, the govt usually needs about 35% of the people's votes to obtain 50% of the parliamentary seats). So I feel that the election-type system should be enhanced by a DNP called "Consensus", that will provide us with a basic "critical percentage", upon which the consensus for each individual policy should be based (f.e. it should be multiplied by a factor >1 for some DNPs like SL, ED, RD). This and the critical ideologic distance should be a measure of how cooperative the political system is and the should be derived from cultural elements. Of course, if the needed consensus is 100%, then it totally reverts into a negotiation procedure. For big numbers (like 75%), the policies will be mostly negotiated. This way, we can allow for mixed (election-negotiation) systems. I figured this out as an enhancement to my election system, if F_Smith wants to proceed into coding both ways.

I feel that with these new elements (active formation of coalitions and varying consensus) the "Politics" model is mature and antagonistic of the "Negotiation" model in it's new form.

3) LGJ wrote (31Aug 10:40):

quote: There is what is known as the "minority rule" in deciding what passes/fails. Although i can't think of any country that has it, it has been used and has had important consiquences. Remember Tianemen Square (hope i didn't really screw spelling of that up). Anyway a small minority of the students wanted to stay, but most wanted to leave when they group of chinese students had their vote. They used the afore mentioned "minority rule" and so everyone stayed.

Almost similar is the case of the student political organisations I participate, in the university. There is no way one can force a political line upon a minority that disagrees with it, so in the cases that we need to show up with a common line, there are fervent discussions, the outcome of which is always rather shallow politically, because it mainly consists of our common ground. We could not do things anyway else; we are not like the mainstraem political parties, which force a line unquestionably (take it or leave it) to their members. Something like this would also happen to the Clash Project if there was no Game Lead or Model Leads to take critical decisions when needed. If this is good or bad, it depends on the purpose of the body; if it wants to struggle for educational and social liberty or if it wants to perform a certain task on time.

4) Richard wrote (31Aug 19:11):

quote: My view is that if a player really wants something to happen, that player should be able to get the thing to happen. For example, suppose that I really needed to raise taxes to support a last ditch defense of a valuable province. I wouldn't want to micromanage a lot of petty politics and run the risk of my vital tax increase being rejected. I would want to set foul play to unlimited and run the tax rate up into the red zone. I would want to make sure that I get my money.

5) F_Smith wrote (31Aug 09:50):

quote: The ruler can indeed fight against giving power to the people. The player can keep the 51% in his hands. He'll see his civ outpaced by more 'free' countries, in production and research, but if he'd rather keep the reigns of govt for war purposes so be it.

quote: Well i don't believe in arbitrarily punishing the ruler for keeping a 51% control or whatnot. More freedom doesn't ness. mean more advancement. ..... So with that said, as long as the player makes wise choices he can still control enough of the power under your system to do whatever the hell he wants and still keep up with the rest of the world (cept maybe socially).

Btw, we are going to have to close this thread soon, due to the 150 posts length limitation. I feel that the initiation of a new thread calls for a new summary of the govt model. Should I do this, or would you like to do it yourself Rodrigo?

Since the limit is up, let's just continue any 'off topic', non-default govt discussions in that other thread I started. Then let's keep the new govt thread on the topic of Rodrigo and Axi's govt model.

I am interested in discussing this at length. I enjoy hearing criticisms of an idea, it only makes the idea stronger in the long run.

THIS THREAD IS THEN OFFICIALLY CLOSED!

A new one will be opened.