First,

I want to clarify a Doctrine is a quirck of a civilization while a national idea is a fundament it.

Mounted Armored Amphibious Knight Supremacy doctrine can't be called a national idea.

Second,

How to implement it ?

exemple :

For troops the game analyses your 50 - 100 least units produced, makes some stats with it and see what those stats trigger as doctrines.

If you produced +80% of foot soldiers you may get a message asking if you want the Foot soldiers doctrine focus ?

If you produced +75% of very armoured units you may get a message asking if you want the armoured soldiers doctrine focus ?

and so on ...

Every doctrine give you bonus and minus it is up to player to chose whatever he want to apply on or not.

Triggers will only roll a probability on whatever the opportunity to get a given doctrine is offered. The wierder the doctrine and less probable you it proposed to you.

Armoured Foot Soldiers Focus.(plausible)

Armoured Spearman Focus.(hard)

Armoured Ranger Spearman Focus.(very hard)

In addition you will get proposed one military doctrine at time even if you meet the requirement for a bunch of them and upper-tier doctrine will require to already apply the related lower tier doctrines.

Wait, didn't we briefly discuss this on the IRC today?

I'm just going to say that I do like the idea of player-influenced "civics" (or Doctrines, whatever we choose to call them) but I just don't see them being realized properly. I hate to be the nay-sayer, but I just don't want to wake up one day in the game, and wonder "Why the hell am I set to "Giant Space Hamster Doctrines"?

I'm very much more in favour of a Civ4/SMAC/X-approach, where I choose my civics as I see fit, in a way I feel my nation should be roleplayed, rather than (risk of running) the prospect of wondering "What the hell is this AI thinking that I'm trying to do?". With that comes the problem of possibly having to re-gear your entire nation for doing something else; Maybe I suddently want to re-route all my research from magic research to technological research, but because of my long, long time of researching magic, I'm hampered?

This idea is a problem for the same reason(s) it's a problem in games that apply a good/evil paradigm based on my actions; It can never know my intentions. I may not even know them myself, at the time.

I do like the final idea in the second post, however. A simple analysis on the majority of the your units produced (over the course of ~100 turns at the time, adjusted by, say, a civlizational creativity stat?), that give you the chance to choose a military doctrine of some sort, or even better in my opinion; the opportunity to research an additional, exclusive, technology (magic or otherwise). Sometimes it'd give you a civilization-wide benefit, sometimes a benefit with a penalty, sometimes an extra item that your soldiers could equip, etc.

The idea of some sort of cloaking "device" comes to mind, after fielding a great many scouts, assassins or such.

I think we might have a chicken vs egg question here.  Let's try few different ideas that I think might be interesting to contemplate.

1.  Seven Kingdoms had an interesting model where each city had a limited population.  This population could be tasked to one of several different tasks, they would gain experience in those tasks making them more capable.  As an aside, espionage meant sending "immigrants" to the opposing village and having them become part of their population.  Amusingly, appointing a spy to a general or a governor could turn over control of the fort or the city. 

The practice would be to model several different skills for each "Charismatic Leader".  Each city has a few leaders that "learn" skills.  I can build a unit of soldiers with one leader.  As they train at facilities, the leader and his troops becomes better skilled.  (I'd think of them as the captains of a company.)  If he trains with Infantry, his troops get better with fighting as infantry.  If he is trained as a archer, his company will become better at fighting from a distance.  History and fiction are full of sending out barely trained people to die on the battlefield.  The inherent balance is that you need to have time and food to feed troops. 

As an aside, you set up skills for:  Farming, to give boosts to food production for that city/town.  Trade, to give boosts to economy.  Woodworking/stoneworking to give boosts to construction or durability.  Espionage/ Counterespionage would end up being a cat/mouse game.  Do you trust your new immigrant with maintaining your walls without a keeper?

Max skill for anything could be a result of tech, or maybe mana.

One interesting factor for this mechanic leads to skilled labor sending caravans of products, rather than materials, to less developed 'Frontier' states.  Instead of sending metal, your now sending swords and armor that can be raided and amusingly used against you.  You can setup assassinations and CRIPPLE a city or productions by killing the smiths and bowwrights.

2.  The Sims MMO had an interesting balance mechanic.  Basically, no one could master everything.  Ypu had 10 (or 15) points to spend on skills.  After you used up your points, trying to improve one stat caused something else to degrade, usually your best skill. 

The mechanics behind this idea is that you set 100 (or 1000) points of focus on a set of sliders reflecting different social agenda such as infantry focus or trade.  You can set your sliders to however you want in the first turn but, each turn afterwards can only alter a particular focus by 1 percent.  Figure weekly turns and you're now trying to anticipate what you'll need months down the road.  To possibly complicate matters slightly, once you get past a certain amount of focus, you might unlock "hidden" sliders.  For example, if you improve your focus on infantry past 50%, you can now dedicate yourself to specializing specific troop types such as light infantry, heavy infantry, or Ninjas. 

Note that this idea is global and applies a constant pressure on the troops in the field.  All Infantry will become less effective as the focus moves away from them to Archers.  This tends to model the idea that individual troops require more than a single sword, or pair of boots. 

3.  Since each faction will have unique trees and spells, simply make different troops require or benefit from different raw resources on the map.  To produce troops for the Fallen Juggernaughts, they REQUIRE a source of Bronze.  To produce Human Britons, you need lots of wood.  Give Britons a food bonus from Forests and they'll be spreading trees across the map.  This is the MOM approach. 

@Luckmann

The UI always offer you to apply or not a Doctrine. You can always refuse it.

@lwarmonger

When you embrass a doctrine, you are in for quite a lot of turn.

In fact the cheap way to leave a doctrine is to change the stats : go below the 80% of foot soldiers produced to lost the Foot soldiers focus. Another way to leave a doctrine is to pay a few turn of unrest : "You fired you generals because you think that Focusing on Armoured Ranger Spearman Focus is now obscelete".

You can also that apply the Doctrine/Quirk to cilization building policy seeing how you split your yearly budget between:

Defense Buildings

Ressouces gatherings

Research related

Entertainment building

From here you go for 2nd tier buiding Doctrine with this exemple :

Defense Buildings Focus => Fortification Focus or Defense Turret Focus

About the Buiding Quirk we can even go down to the level of each city. For a city covered with farms that city should be recalled the farmers city of Hobbitburg and so on.

----------------------------------------------------------------------------------------

About the various Malus/Bonus of Doctrine.

Bonus 1 : Reduced production cost, research cost

Bonus 2 : Increase of effectiveness of the buiding, troops, research

Bonus 3 : Unlock some niche research, buiding, spell

Malus 1 : Increase production cost, reasearch cost

Malus 2 : Deacrest of effectiveness of the troops, spells

Malus 3 : Unavailable Unit, Building for production. Locked research.

McFungos,

I thought about the action-based Doctrine/Social-Engineering idea overnight and it seems more feasible now.  It would be a pretty significant chunk of work, but as Spartan said there are other requests being made that tie into the same thing.

Still, there are simpler models like SMAC or EU/HOI that might serve better in terms of the devs actually being willing to implement them

Anyway, the piece common to all 3 models:

A number of doctrine "types" kinda like:

<DoctrineTypeCollection>
    <DoctrineType name="Quantity vs. Quality" ID="DT01" />
        <DoctrineTierCollection>
            <DoctrineTier threshold = "-100">
                <DoctrineTierModifierCollection>
                    <DoctrineTierModifier modifierType="MilitaryMoralePercent" value="-20" />
                    <DoctrineTierModifier modifierType="MilitaryUnitCostPercent" value="-20" />
                </DoctrineTierModifierCollection>
            </DoctrineTier>
            <DoctrineTier threshold = "-90">
                <DoctrineTierModifierCollection>
                    <DoctrineTierModifier modifierType="MilitaryMoralePercent" value="-18" />
                    <DoctrineTierModifier modifierType="MilitaryUnitCostPercent" value="-18" />
                </DoctrineTierModifierCollection>
            </DoctrineTier>
            ...
            (not including all of it)
            ...
            <DoctrineTier threshold = "100">
                <DoctrineTierModifierCollection>
                    <DoctrineTierModifier modifierType="MilitaryMoralePercent" value="20" />
                    <DoctrineTierModifier modifierType="MilitaryUnitCostPercent" value="20" />
                </DoctrineTierModifierCollection>
            </DoctrineTier>
        </DoctrineTierCollection>
    </DoctrineType>
    <DoctrineType name="Control of population vs. Freedom of population" ID="DT02" />
        <DoctrineTierCollection>
            <DoctrineTier threshold = "-100">
                <DoctrineTierModifierCollection>
                    <DoctrineTierModifier modifierType="EffectOfPoliceUnitPercent" value="50" />
                    <DoctrineTierModifier modifierType="CivilMoralePercent" value="-25" />
                </DoctrineTierModifierCollection>
            </DoctrineTier>
            ...
            (not including all of it)
            ...
            <DoctrineTier threshold = "100">
                <DoctrineTierModifierCollection>
                    <DoctrineTierModifier modifierType="EffectOfPoliceUnitPercent" value="-50" />
                    <DoctrineTierModifier modifierType="CivilMoralePercent" value="25" />
                </DoctrineTierModifierCollection>
            </DoctrineTier>
        </DoctrineTierCollection>
    </DoctrineType>   
    ...
    (more doctrine types)
</DoctrineTypeCollection>

These aren't the best examples and assume those modifier types being implemented, but other features will need to apply civ-wide modifiers anyway.

Also, game state for each player would need something like:

<DoctrineValueCollection playerID="PL001">
    <DoctrineValue ID="DT01" value="-45" />
    <DoctrineValue ID="DT02" value="22" />
    ...
    <DoctrineValue ID="DT14" value="67" />
</DoctrineValueCollection>

The difference between the three models is what modifies a player's doctrine values (and in some models you wouldn't necessarily store the raw numbers, but rather the pieces of the calculation).

In SMAC, you chose various social engineering choices which gave you plusses and minuses to the "doctrine" values, so you'd have something like this.  In SMAC the "doctrines" were almost always the higher the better, and negatives were penalties.

In Europa Universalis, you had the ability to do a plus or minus (equal to about 10 points on the above scale) every 10 game years, with the cost of 1 point of stability (on a scale that went from -3 to +3).  Random events (most of which gave the player a choice between two sets of modifiers) could also change the values, sometimes quite a bit.  For each category going positive on the scale would generally give bonuses to X, Y, and Z, but penalties to A, B, and C, and going negative on the scale would do the opposite.

McFungos's suggestion, if I understand it correctly, is that the game computes the doctrine modifiers from your last 50-100 turns of actions, allowing the player to "veto" any tier-change but not allowing the player to artificially cause a tier-change. 

For implement-ability's sake I'd suggest not trying to have it do those computations in retrospect as it would be a pain data model and processing wise to keep the kind of detailed action log necessary for this purpose.  Instead have each action give a + or - to one or more doctrines (might need to increase the scale to -1000 to +1000, or wider, to allow for enough granularity). 

The difficulty here is that it would require logic to be added to the code of every single action (or at least doctrine-influencing actions), even though that logic could be fairly simple. 

Another problem is simply that of identifying each action that should influence doctrine and in what way, according to what parameters.  For example, destroying a city should shake things up a bit; but does it depend on how many people were in the city, which faction/race it was, whether you were at war before the attack, how long you were at war, how many of your cities they had taken or destroyed, what magic types the enemy channeler uses, whether the city was surrounded by your territory... etc?

Anyway, you could at least do something simple along these lines by having something like:

<ActionTypeCollection>
    <ActionType ID="AT01" name="Finish Building Infantry">
        <DoctrineModifierCollection>
            <DoctrineModifier doctrineID="DT06" value="1" />
        </DoctrineModifierCollection>
    </ActionType>
    <ActionType ID="AT02" name="Finish Building Cavalry">
        <DoctrineModifierCollection>
            <DoctrineModifier doctrineID="DT06" value="-1" />
        </DoctrineModifierCollection>
    </ActionType>
    ...
</ActionTypeCollection>

And at the end of the function that processes completed unit construction you'd need to do something like:

if(unitJustBuilt.Category == Infantry)
{
    ProcessDoctrineModificationsForAction("AT01");
}
else if(unitJustBuilt.Category == Cavalry)
{
    ProcessDoctrineModificationsForAction("AT02");
}

This would allow modding of the effects through modding of the xml ActionTypes, rather than embedding the logic in code (you could even store the "AT01" on the unit type record for infantry, I suppose).  But I don't know how I'd put all the necessary logic in xml for the doctrine change of destroying a city.

Question:
- all these models assume each doctrine has only two directions (plus or minus); what about a doctrine where your faction favors either infantry, cavalry, or archers?  What about something with more than 3 options?  Don't have to have these, just thought I'd bring it up.

Anyway, that's enough from me for now.  Sorry for the length, I just need to think concretely to get a handle on how much it would take to do something.

Thanks for starting this discussion,
Keith

And why not some "magic" doctrines?

Eye of Nowhere (use your magic to get better information)

Magic technology (your buildings are speed up thanks to your mana)

The force, luke. The force (you can use mana to improve relations with other players)

Mass enlightenment. Each town with a mage tower will speed the process of making your people more fanatics

@Keith

You also notice that the doctrine/quirk tier things has some serious weakness. The worst being the near no end of the number of combinaison of troops possible leading to the same number of doctrines/quircks. So to my chagrin the idea to have doctrines in different tier can't be done.

However found a way to go around, keeping mostly the doctrine/quirks concept.

How to pull the stunt ?

I will have a kind of Troops building buffer containing the Traits of the least 50 - 75 least built units. When this buffer is full, it will start overwriting the oldest data with the newest

With a clock timed event like in Heart of Iron or Europa Universalis, the game will ake the stats of those units how many cavalry, beast, footer, ranged, magic, blunt, pierce, etc....

It will pass the results to the triggers test.

Afterward the UI will notice the player with the List of preferences, the players having the choice of which to take and which to discard.

From there the game will etablish a prefered units traits list and a shuned units traits.

In the end every unit will get a doctrinal accordance rating +1 for every in traits in the prefered list and -1 for traits in the shuned list.

Higher the rating for a given more bonus it has lower and crapiest is this unit.

On positive side of the scale :

0 No change

1 -1 turn to train

2 +Atk

3 +Def +Hp

4 +Mov +Moral

5 ++Atk

and so on.

On positive side of the scale :

-1 +1 Turn to train

-2 -Atk

-3 -Def -Moral

-4 -Hp +Increase unkeep

-5 --Atk

and so on...

The Militaries Doctrines is reviewed every 40 - 60 turns, if the player want to change them he will have to start building others kind of unit or paying with unrest to revoke some or the whole set of doctrines.

That way we can even handle unit with +6 Traits without exploding the combinaison of coding.

Some good stuff here. I would love for some input from SD on this idea as well as the others in the ideas forum.