Outline about the social mmo-game in my mind1. We just make stage, players play their own story 2. Real person is the most important resource 3. The more the guild have, the harder the guild want to keep 4. A highly simulanted battle field makes diplomatic/social work a key factor Basically, I believe there will be a revolution of structure of current mmo-game shortly, future online game will be more social than ever before. Players may not play what we design but rather enjoy their own activity and the consequent result. That means players will compete or cooperate with another groups of players. Battlefield in World of Warcraft is just a good but too simple instance. People are complicated, so we have politics and all kinds of competition in the real society, these together makes real world amazing. But current mmo-game doesn’t have enough depth for people to create a social structure, so we need a change. I used to play World of Warcraft, but give up after 2 years. I found it’s quite hard for a solo gamer to have fun in WOW. I never experiencing the high level instance (never join a guild), never participate in battlefield (too many people waiting). The only reason I stay there is because my friends playing. As the level disparity increase, I feel myself valueless and boring for solo, finally I quit. From my point of view, the most serious problem in current mmo-game is the huge gap between hard core and light user. So are there any possibilities to establish a bridge between those different types of gamers? Suppose there are something hard core want to gain, but as humanity, they have finite energy, so the most efficient way is to let others – light users as example, help them. This would be a solution. People are always wanted to be accepted from others (so they will be active), contrarily, everyone has some kind of bulwark in their hearts (so they will be negative). If more guild (yes guild but not single user) need other players especially the light user, I think single player are more willing to do social activity. So we should create demand let guild need light user. Different from single user, guild or a group of player are steadier, more pressure-resistant. A group of people they come together for their common benefit, they have the same goal, but might different on mind and attitude. The ambitious leader wants to lead the group become most powerful guild of all, in this case, they probably willing to pay expensive cost for their amour-propre. That is to say, we should provide ultimate honour for them. For instants, in one paper work of mine, theoretically a guild can occupy all cities in the game (There are more than 50 cities in the game) or they can even beat the NPC emperor down and rule the whole world. But they must pay quite a lot of work force for this target, nevertheless, much higher if they want to maintain their govern. I hope this design could create more demands and by this a guild need an enormous quantity of real person (not the character). If this works, we shall have a solid foundation to make light user valuable for hard core gamers. They are the work force which guilds need. Another problem is to make more conflict opportunities. A common phenomenon in mmo-game is too much emphasizing on absolute equity. Weak guy can never beat down powerful one. Suppose there are some new rules at siege war(just a model): 1. A guild can have 50 people on maximum 2. A city can be defend/attacked by 20 people on maximum 3. There are 10 cities in the game and you’ve already take 5 of them 4. You don’t know when & where your enemy will launch their attack on your city The defender is powerful, every member of the defender is an elite. But by those rules, if 3 weaker guilds launch its attack in the same time on the different city, one of the attacker will have more chance to occupy it (20 attackers vs. 10 defenders). Of course we can provide user some adjustment method to rebalance, such as defender can hires some players joining the battle temporarily (for guild, they are work force ). By this change, war in the game may represent more complex, and more diplomatic /social activities are required if players want to be more efficient. Fundamentally, in game nature resource distribution can be used to drive conflicts. ------------------------------------------------------------------------------- ok, it's a bit long and apologized for my bad english. :P how you look about this topic? Hope we could share some ideas.

We've seen elements of this appear in many games - Sandbox non-instanced world, freeform character development, content driven by player interaction, players shaping the world, player scripted content, etc.

The best MMO would be one that combines all these facets into one in a polished high production value way, but no one with the money to pull something like that off seems up to the task or brave enough to risk the money. The last big game to try was SWG, and they ripped all the best parts out of the game in a blind panick to arrest subscriptions by being more like WoW, instead only hastening their demise.
Bioware is pretending like the princibles of the first SWG were a failure and has dutifully replicated the WoW paradigm, except merely with the addition of even more highly scripted quest content.

Bethesda would be our last hope for a major MMO to focus on open ended gameplay and player driven content, given their design philosophy behind Oblivion and their commitment to the mod community, but I wouldn't bet on it. They said they are open to the idea, and they want to do something no one has ever really done before - Well, that's it. They've got enough wealth as an independant developer to be our only hope of ever seeing such a game. Because we're guaranteed to get more of the same if they rely too much on publisher funding, who will demand that they make the game like WoW in order to safeguard profits (Even though we all know that would only ensure their failure, publishers aren't that smart).

I always thought that simply allowing only one avatar per player would do a lot to make the social aspect more... social. That way, your in-game avatar is really YOU - it isn't just one of your generic assets you can use and discard at will.

I also believe the social quality of online games is increasingly becoming the games most important aspect. It seems these types of ideas are getting popular.

Your design is very similar to a design model that I'm working on.Do you have more of your design doc available?

ZeroShift, ya, generally I agree with you. But Fallout3 makes me a little bit worry about Bethesda. Actually it doesn’t have much difference with Oblivion.

Talin, personally I think we could do better than just allow only one avatar per user. This is a paradox, if users have to choose a race/job/etc, they would like to try all different option. If they can do everything (no race/job/etc), I’m afraid they will feel confused because of don’t know how to start the journey.

GninjaGnome, the only problem is it’s in Chinese…
Actually it’s hard to see everything clearly in a detailed design doc coz the idea has been disassembled into everywhere.

But if you like, it’s happy to discuss the detail with you.

Personally I hate it when mmos try to force me to play with other players, like activities which can only be done by a party of 5 people or worse must be started by a guild leader and only guild members can join.

If one wanted to make an MMO more social, I think the best way to do it is to include good message board software, both public and private message boards where the private ones can be organized around a roleplaying topic, ages verification which allows verified 18 and older players to do adult-rated roleplay without fear of being banned, and all pick-up group activities in the game should be really easy to 'teleport to' (for free) or access from a menu instead of needing to hike there inside the game. Dungeon instances should be scaled so they are equally difficult and rewarding for soloing, groups of two, and groups of three. On top of that I would add some kind of personal ad system into the game, because that's what people really need help with, meeting other people who have actual common interests besides playing the game.

Interesting, the OP and sunandshadow use a very different approach to the same problem and I think both can succeed in their way. Personally I prefer the topic starters way of encouraging role- and teamplay by creating more of an ingame sociaty but the other way might work as well.
There is an old joke "WoW is the most expensive chat program" and it is true in a way. You could expand on that by creating better ways of ingame communication.

Still I prefer giving players and guilds power over the world by using other players of the same factions for their goals. Create several layers of government, occupied by NPCs at the beginning but possibly taken over by players. Have those players create Quests. So you could effect the economy by giving out lots of "gather items" quests or attack another faction by giving quests to raid a specific village at the boarder.
Of cause there should be NPCs the people talk to for getting the quests - or a blackboard or something like that ...

I hate been forced too :P

For the operator, the want to make players know each other so that they can stay in the game for long time. This is true, but guys like us will feel uncomfortable.

But in case many guys and guild need you, they beg you for help; probably you would like to help them. A example is what I said about the siege war, they can hire some players to help them, and by those mercenaries, they might restore the situation.

In my mind, work force is important for guild in doing anything. You can govern the whole world, but if you want to, make sure there are enough friends stand on your back. :P

An evil thought is the more people you interact on, the more chance involve in a conflict you have.

Quote:

4. A highly simulanted battle field makes diplomatic/social work a key factor

I'll start with this as it is probably the thing I most disagree with.

Simulated battle fields do not make for diplomacy. They make for complex/complicated battles.

If you want social interaction, then you have to design for it. If you want battles, then design for it.

What you need to do, if you want social interaction, is to make social interaction more meaningful.

I am currently working on a design for a social interaction framework for these kinds of games (as I am not in a commercial company, I can share it).

The Social Interaction system operates around the concept of Agents. Every NPC you interact with is an Agent. That is they operate independently from each other and make their own decisions.

This is nothing new.

What is new (and important to my Agents) is how they make the decisions and how they interact with each other.

Players also count as agents, just ones not run by the CPU. However, all Agents (whether Player or CPU controlled) have to be able to take the same actions (in this instance, this only applies to the social system).

The basic Actions are:
- Setting Affinity
- Communication (note that this is not limited to conversation trees)
- Navigating a Social Network

Ok, the first one "Setting Affinity" is pretty simple (but it can be as complex as you like). At its more simple it is just working out if an action an Agent took is either Good, Bad or Neutral and adjusting a particular value to reflect that.

More complex ones can have multiple axes, or even discreet values for various criteria. But, we'll just keep it simple for now.

"Communication" is more tricky because you ahve to know what is being communicated. Looking at human behaviour, around 90% of communication is actually gossip.

This can be simulated by allowing Agents that interact to trade information about other Agents. Mostly this will be about someone they both know (each agent keeps a limited list of who they have interacted with), they will more readily talk about other Agents that they have interacted with recently and they will also talk about other Agents that have effected them (Negatively or Positively) greatly.

In effect we will ahve 3 lists
1) Combined list of Agents shared on a list of who they know.
2) Agents that the Agent has interacted with recently (but not matched against the Agents the other ones know)
3) Agents that ahve had the greatest Positive or Negative influence on their Affinity with them.

An Agent can appear on all 3 lists simultaneously (so if an agent is know by both parties, has interacted with them recently and had a big influence on their affinity levels, then they will be very likely to be selected).

Each Agents combines all 3 of these lists into their own "Gossip List". From this list an agent is selected for transmission to the other interacting Agent. This transmission is then handled like the selected Agent has interacted with them, and gets inserted into the lists and can change their affinity in the receiving agent (most likely to a lesser degree).

Gossiping will also allow the Agents to construct a "Map" of the social Network (who know who). This can be a permanent Map stored in each Agent, or it might be a temporary one as they try to Navigate the Social Network (see below) and is discarded (to save space on the server) when finished with.

When dealing with players, this agent might offer to talk about the selected agent(s) as part of a conversation tree, with appropriate Positive or Negative comments. They can also offer to make a "Social Network" connection (see below).

The "Social Network" is about the relationships between the various Agents. Using the two above aspects (Affinity and Communication) you provide the basic tools needed to interact with a Social Network. However, to properly navigate the network you need 2 extra tools:

1 - Social Structure
2 - Introductions

Social Structure is is the organisation of the Social Network. The easiest way to implement it is to make it so that "Strangers" rarely interact unless they hold a social position that requires them to (like a Shop Keeper, Tax Gather, etc).

The second part is "Introductions". This aspect allows one Agent to introduce another as a "Friend" and to make them no longer a Stranger.

Combined, these tools allow groups to form in your world, and then for an Agent to navigate through the Network over time.

If you restrict Introductions to Agents that have a level of trust (their Affinity), then it will require the Agent to gain the trust of one Agent so as they can be granted an introduction to new Agents.

For computer controlled Agents, this is basically a path finding algorithm (with a few twists needed) through a network, pretty simple to start with.

The twists come from not knowing the "terrain" of the path before hand. Terrain is basically the Connections between Agents, and the ability to gain Affinity with the Agents that make up the path.

The facility of Agents to navigate the Social Network can be controlled by how many "jumps" in the network they are allowed to do before giving up on a particular path (2 would be about average).

The path finding algorithm attempts to find an Agent in common between them and the target Agent with a positive Affinity for the Target Agent. To do this they start path finding from both them and the Target Agents and stop when they find one in common,but only using the Agents they know, or they know the Target Agent knows (through gossiping).

That is the basic framework, and I am designing it to be expanded upon. It is also still a work in progress and is not complete (but getting there - I have the skeleton of it worked out as you can see). However it is good enough that it can be used, but you will still ahve a bit of work getting it to a point where it can be used in a project.

The idea is that whatever interaction method you use, if it can allow these actions to take place, then you will (with the rest of the system operational behind the scenes) be able to have a complex and rich social system for an RPG (be it MMO or SP) and can be tuned how you like it.

You won't necessarily have to have every NPC in your game world operating within this Social System, but you will need a minimum critical number (as they have to have enough interactions often enough).

However, you can get away with a certain number of them as aspects of a single Agent, and although there might be multiple NPCs in the world, they might share the Same data and Agent "soul" between them.

For instance, you might have a "Farmer" Agent, and then several NPC farmers in the world are just projections of this same Agent into the world and any interaction with them is the same as interacting between any of them.

This technique would not be good for important people in the social network, and you certainly would not want too many of your NPCs to be like this (even if they would otherwise be similar or in similar positions). It is just a technique that can be used to give a perception of a larger population without needing too much server memory/hard drive space

A lot of the feeling of richness of the interactions is dependent on the number of actions you allow the Agents to take. These must be meaningful interactions, not trivial interactions (being able to buy and loaf of bread, or buy a fish - or buy a sword - are not multiple interactions, they are the same interaction).

However, if you had actions like: Buy a sword, insult his ancestors or gossip about the Fish seller. Then these would be multiple interactions.

In terms of action these would be:
Buy a sword = Trade
Insult his ancestors = Negative Affinity action
Gossip about the Fish seller = Communication/Gossip

So you have to think of the actions in abstract terms. Actions that have the same abstraction are the same action.

However, when you implement these action for the Agent to choose, then you might package several abstract actions into a Package Action. These Package Actions will add a level of complexity into the system and make for interesting options for the Agent to choose from (especially players), but the downside is that the necessary decision making code for a CPU Agent will become more complex.

Quote:

Original post by ZeroShift Bethesda would be our last hope for a major MMO to focus on open ended gameplay and player driven content, given their design philosophy behind Oblivion and their commitment to the mod community, but I wouldn't bet on it. They said they are open to the idea, and they want to do something no one has ever really done before - Well, that's it. They've got enough wealth as an independant developer to be our only hope of ever seeing such a game. Because we're guaranteed to get more of the same if they rely too much on publisher funding, who will demand that they make the game like WoW in order to safeguard profits (Even though we all know that would only ensure their failure, publishers aren't that smart).

Yeah, because CCP have proven time and time again that they hate open ended, player driven gameplay. :P

That's what I'm waiting to see, how successful their next game will be.