Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[DWeird]

These are not the droids you are looking for.

Mostly because there are no droids at all in this timeframe.

While I appreciate the bump, I'd like an extended discussion on the topic at hand even more.

Any takers?

[Force]

Hey read this and it's a great idea, so I
Just wanted to make a suggestion towards this matter, which would be to establish a Family system that allows a group of families, about 5-6 who are capable of ascending to the throne if the whole royal family is killed, or in the case of a schism or revolt. These Families are determined through a value system that allocates points to the families giving them their importance status. This status could be influenced through:
-   Influence of the family on guilds, someone of the family is elected a guild master then they have control of it, or a possible vassal system
-   How much money they are in possession of (when economy is in place)
-   How long the family has been in the fortress, how old the family is
-   If they hold positions of authority, such as mayor, treasurer, royal guards, etc.
-   How well the family has supported the royal family and country, through diplomacy or fulfilling requests for them (ending wars, winning invasions, defending the city, slaying mega beasts, etc.)
The system should be activated similarly as the economy, through regulations such as:
-   city has to reach a certain size
-   Economy has to be in place
-   Royal family should be in fortress
This would prevent any annoying fighting that could be created in early stages of the fortress and makes the game more interesting in the later stages where it adds some challenges.
The fighting that occurs between the families should not be excessive, because it would be annoying if every few minutes a dwarf dies, especially if they are legendary crafters. Fist fights would be alright, but assassins’ guilds (could be brought into the game) would be a sort of last resource method for the families and could be made dangerous to the family by making them very expensive (a certain percentage of their wealth) so that hiring them affects their position of importance. The main fighting between the families should be to increase their wealth and influence or decrease the wealth and influence of other families, as to get them out of the top 5-6 families that can ascend the throne. (e.g. If at a high enough level get the opposite family member to lead an invasion or defensive battle, in the hope that he will die in the battle, or contesting for a guild by getting a relative from the family to take over the guild (I’m assuming that guilds will have an elective system that can be used if certain requirements are in place, such as backed by the majority of people in the guild) and get more work requests to increase the money they get.
The families should not constantly fight but also be able to ally themselves in order to increase influence and security as top families.
The system for family prestige could be a numeric one, if a condition is fulfilled family gets 10 points (e.g. Becoming royal guard= 10 points; having king in family= 500 points etc.). And the same regarding deductions of points (e.g. Losing guild=-50 points, losing position= -20 points)
Also it would be nice if there was a description for the dwarf saying which house/family it’s in (I think a house name would be more sensible as the people in a family do not have the same surname, so it would not be easy to know who was in which family) as well as a extra side on the main display which shows maybe the top 10 families and their point values as well as the leaders of the houses, a description of the family crest, which guilds they control, their relationships to other families and a list of achievements (how many royal guards have been in family, how many years of kingship, how many mega beasts slain etc.)
The amount of control a player should have on this is limited; he has a certain control over things such as who is appointed as royal or fortress guards, or on who is sent on an invasion as leader. But things such as the mayor or guild masters should be out of their reach.
I don’t know anything on programming so I’m sorry if I’m suggesting impossible things. But the whole thing is just an idea anyway.

[Granite26]

Possibly just measure similarities for group membership.  If you're just 7 random people, you're going to have a 0 affinity for the group, but by the time the first wave shows up, you should have a decent friends affinity for the original 7 (already implimented).

Later on, your affinities become more complicated, with groups, clans, and families.

Measuring differences is important.  While brothers my scuffle and fight, add an outsider, and all that goes away.

Make sense?

[blue sam3]

Perhaps also a clan which likes to trade with other species could be at odds with those that don't, perhaps with cave adaption having something to do with this, where the families of those dwarves who remain deep within your mines pumping out beds for a few generations want to ban trading, whilst those that have been out working your above-ground farms would want more trade, and they could struggle to gain power (possibly some form of election for some nobles (mayor) could be implemented) and if the gap becomes wider, they could actively persecute each other, some form of training for children, or gaining skills (diluted versions) from their parents would also help with this.

[Force]

Ah just noticed I left out the bit about relation systems. Fights between brothers, siblings and separate individuals would affect mainly the individual relation system and if the separate individual involves one of the top families then it affects the family value system. For example if two brothers fight then one might develop a grudge against the other (more than one fight would be necessary), but the family would not be affected. If family members from different families fight with each other then it affects their individual relation and the relation between the families.
As granite26 said. It would need 2 systems to get a good result.
Seems to me it would be really complex to programme but really cool

[Dae]

We had an interesting talk about grudges that quickly went over groups because, well, grudges only start being Fun when several persons are implied.

I think the less specific the system is, the more numerous different situations able to emerge from said system are, which implies more sources of Fun. So I'll try to remain as generic as possible.

A group's influence could be described through three things : its economical power (how much money does it disposes of), its political power (the sum of the power of the politic people among its followers, military power included) and its social influence (how many followers does it count).
Also, so as to reflect alliances, a group would add to its influence a part of allied influence, depending on how tights their bounds are.
I don't really know what such a number would be useful for, though. During worldgen, as a shortcut, or when deciding who to fight or fear, whether victory over some group is possible ?

Goals would be what mainly distinct a group from another. I don't really know how to chose these goals, but I have an idea about how to use them. Basically, each member of a group would have happy thoughts when its group achieves one of his goal, or a step toward it. Of course, when a group fails to achieve a goal or takes too much time to do so, they would have an unhappy thought about it too.

A group could order some laws its members would have to respect (this would be particularly useful for religions and political parties). Someone who does not respect these laws would attract a grudge from this group (so from its members, too). Then there's the idea of punishments, possibly a member being banned from the group (bad thought depending on how much he liked the group).

Now, the part where it all becomes interesting is at the individual level : someone would want to please groups he is a member of. Simplest way to do this would be to try and reach happy thoughts instead of just merely receiving them. This way, you just have to define an occasion of having a happy thought and you're done.
So, members would want their groups to achieve their goals. For this, they would ask themselves "what can I do so my family/guild/clan succeeds ?"
They'd check a list of what they can do, and the repercutions over every of their groups' goals and also over their OWN goals.
Then, we would have to calculate some sort of average/bayesian product and chose what to do depending on the individual personality traits and determination to achieve said goals (which would depend on how close they feel to each group and goal).

Having a difference between personnal goals and groups' goals is crucial here : dilemnas and tension and drama would come out of this. Along with corruption, deception, brother-killing, decision making, not-making-a-decision-because-that's-still-your-daughter-the-elves-have-captured-and-you-can't-order-magma-to-be-released.

Like skill rust, an individual would tend to forget its bounds to a group. They would have to be maintained by trying to act for the group (happy thought for the one who tried) and succeeding in helping the group (happy thoughts for everyone !). This way, a group which fails in everything would quickly lose influence, but someone whose goal matches the group's goal would still be committed.

In short, a member's opinion about something would be influenced by what its groups think of it as well as about his opinion.

Let's take the example of Urist McAverageDwarf. A new Mayor is elected, Dulcim McFamilyKilledDuringElfRush. Dulcim declares war to the elves. What would Urist do ? She doesn't think of anything about elves, nor war. So, mostly "meh, why not". But, she pretty much likes Dulcim. In fact, her family likes Dulcim, her friends like Dulcim, the priests of OdilNom like Dulcim. So, its gonna affect a little her choice, and she's gonna trust Dulcim.

Meanwhile, Dolum is a merchant whose main buyers are elves. As a merchant, he wants to keep having money, so he wants elves to buy him stone mugs. If his fortress declares war to his source of income, he'll no longer have a source of income ! Holly gosh, even if everyone thinks Dulcim is awesome, he doesn't want his family to be poor !
So he'll try to react.

[Force]

I read a thread on religion and conversion and what you said would fit in well with how religious groups would win or lose influence, so that would be interesting.
I'm not sure how big of an influence it should have on the happy/unhappy thoughts of the dwarves because if they're in more than just one group they could easily fall into tantrums when both go down.
How much, were you thinking, would the group have influence on changing the characteristics and thoughts of the individuals? And would they be able to develop such a strong loyalty that they would stay in the group till they die? That could be interesting.

[Dae]

How much, were you thinking, would the group have influence on changing the characteristics and thoughts of the individuals? And would they be able to develop such a strong loyalty that they would stay in the group till they die? That could be interesting.

I think a first problem would be to avoid a sort of resonnance effect. I mean the kinf od situation where the group likes someone, so each member likes him more, so the group likes him more, so each individual likes him more, etc. Perhaps the idea of deciding what the group thinks depending on what the members think isn't so good...
Anyway, fighting until death for something would work this way : for a faction, being at war with a faction means going strongly against their goals, so winning over the opposite faction would be very important (let's say, a factor 10). But staying alive is also a goal, with varying importance (if you have a sense of preservation, if someone depends on you, ...), that would also be influence by the situation (mostly fear coming in).

Let's keep it simple and imply you only have two actions, fighting and fleeing, and two goals, caring for your group and caring for yourself. The goal of the step is to know which action to chose, so you sum up what advantage/disadvantage each action would provide to each goal.

First, fight for your group is pretty important (they'll die if you lose) : 10. BUT fighting implies risking death. The result is (how strong you feel - how strong the enemy looks). Here, there are two armies of equal average strength, so it doesn't matter much, let's say -2. Ideally, for the goal "surviving" you'd calculate how likely you're to survive, but you can take a shortcut. Now the fight option result is (10 * (how much you're committed into that group) + (-2) * (how much you want to live)) = (10 * 5 - 2 * 7) = 36.

Second, fleeing. Fleeing implies not fighting, so the result would be (-(how much useful you'd be in a fight) / (total strength of the army)) = (-25 / 1200) = -0.2 (you're not very important in the fight...). Fleeing also implies not risking death, but doesn't really helps you, so (0 + (difference of power between you and your opponent)) = (0 + 2) = 2. The final would be (0.2 * (how much you're committed into that group) + 2 * (how much you want to live)) = (0.2 * 5 + 2 * 7) = 15.

So yeah, you care for your life, you're not important, but it's very important and you fight for something that counts. So you fight. As time goes by and you are wounded, your usefulness drops and the difference of power between you and your enemy skyrockets, so you'd be likely to want to flee halfway through the fight, but then even if you want you'll have to cut back how difficult it would be (in fact, how likely you are to succeed...)

This mechanism is tightly related to Granite26's Context-based IA, which I think would suit DF particularly well because of its high modularity, and then possibilities of RAWification. I've just more focused about how to check what to choose.
Yet, such an IA would be quite complex to implement. Returning to groups, without really talking of an IA, we could have :

- dwarves' opinion based for a part about what their group think.
- dwarves fighting/not helping someone they really don't like.
- dwarves growing grudges about people who attacked them or someone they like.
- dwarves giving time/work/money to a group or work group-specific jobs.
- special actions done them every once in a while so as to make their group happy (and perhaps not doing it because they're not corrupted enough.). For example : buying presents for friends, lowering the price of mugs so you sell your cousin's stock, oredering someone you don't like for hammering (or assassination).
- So as to keep some coherence in your fortress, a meeting with the mayor or someone to appease grudges.
- grudges and friendship appeasing over time.

[Force]

I guess it would be better to limit the effect the group has on individuals, and instead have it focus on group to group relations. Maybe making it possible that reaching a goal increases their loyalty to the group and happy thoughts. However if someone does sth good for the group it hardly has an effect on the others in the group. It might be a little more difficult for individuals in to different groups.
The calculation thing sounds like a good idea, could also be used for other relation systems, but probably quite complex to programme. Could be worthwhile though  .
I wonder if a sort of aristocratic system could be introduced as well, making wealth become a factor as well, though that's sort of related to my family/house idea further up.