[MUD-Dev] NPC Goals ( Was: RE: Dynamic Timelines)

[Travis Casey]

Monday, December 11, 2000, 1:52:30 PM, Lord Ashon <ashon at wsunix.wsu.edu> wrote:

>> - Making NPCs behave in a consistent way.  This requires the system
>>   to know things about the NPCs.  E.g., if Bubba has a goal "marry
>>   Buffy", then if Bubba's first scheme to do that is thwarted, it
>>   makes sense that he'll try another one later.  If Bubba goes after
>>   a different princess every few months, things will seem less
>>   consistent.

> This statement I agree completely.  The Mobile's _must_ know a couple of
> things:
 
>  - They Must Know their Own Goals
>  - They Must know the 'shape' of the World
>  - They Must know how to get things done.

Question: What do you mean by "the 'shape' of the world"?  My guess
would be that they need to have a way to check on current world-state,
but I'm not sure that that's what you mean.

>> Attitudes tell how an NPC feels about things.  Some attitudes may
>> stand alone, while others require an object.  For example,
>> "violent" can stand on its own, but "desires" needs to have an
>> object for that desire.  So, Bubba the Evil Warrior might have
>> these attitudes:

>> - desires Princess Buffy
>> - dislikes Buffy's Father
>> - violent
>> - dishonorable
>> - desires money
>> - desires followers
>> - hates paladins

> What if we start at even a more basic level.  Erwin S. Andreasen,
> and Brandon Downey came up with a Bartle-Quotient test.
> (http://www.andreasen.org/bartle/) To determine what 'suit' players
> fit into on a mud.  It seems logical to me that we should do the
> same to for the NPC's.  We can then use these to determine what
> type's of activities and the NPC is most likely to participate in.

Bartle's four divisions seem overly simplistic to me, especially when
dealing with NPCs.  Take Achievers, for example; they desire power
(higher levels, more goodies, etc) -- but people in the real world
don't usually want power simply to have power, they want it for a
*purpose*.  You could class an NPC as an achiever, but that doesn't
give you much by itself -- what's that NPC going to *do* with the
power they gain?

To me, it seems that Bartle's divisions are answers to the question
"How will X go about achieving their goals", but they don't help us
decide what those goals are.  (Well, to some extent, they do, but it's
a very vague guide at best.)

> Passions

> This is perhaps one of the most innovative ideas that I have heard
> of in dealing with NPCs If we use the Bartle Suits to give 'weight'
> to a set of passions, and then randomize it to give it a little
> variability, we can match or 'shift' the direction of our npc's If
> we have to many killer NPC's/Players we can give weight to the
> algorithm to shift more of the npc's to an explorer or achiever
> goal.

Instead of thinking of these as goals, I think it might make more
sense to think of them as methods.

For example, a character might have a goal of "defend my village from
the orcs".  An achiever might try to gain enough personal power to
defend the village.  An explorer might seek a new place for the
village to move to.  A socialiser might try to talk to the orcs, or to
organize a defense.  A killer might go on the offensive, trying to
destroy the orcs.

>> Major and minor characters: Characters who have many strong
>> personality traits and the capabilities to do something with them
>> would be more likely to be have scenarios centered around them than
>> other characters.  This creates a natural division of NPCS into
>> major and minor characters.  Allowing for scenarios to change an
>> NPC's personality traits and/or capabilities can allow the status
>> of an NPC to change over time.

> To make the system work and feel dynamic, this would be a necessity.
> We would have to be able to have minor characters who can rise to
> major character.  But, you could and shouldn't ever do it in
> reverse.  A Major Character that plot is canceled and they are
> thrown off their seats of power, they can still work on re-building
> their previous position and hunting down the Player that did.

I think that "major" and "minor" status is a dynamic thing.  A "major"
character who loses their status might temporarily become a "minor"
character, but that's not necessarily a permanent condition -- if/when
that character gains power again, it can become major again.

To me, "major" and "minor" for NPCs are more designations of what
paper RPGers sometimes call "spotlight time" than anything else -- how
much of the "focus" does this character get?  This can naturally
change over time.

For that matter, allowing major characters to slip to minor status and
then recover can add additional interest and color to the game world.
Let's take our old friend Bubba the Evil Warrior as an example:

Bubba goes after Princess Buffy by starting a war.  Bubba's army gets
trounced, and, worse yet, the fact that he's doing this for love of
Buffy gets out.  The evil hordes don't want to follow him any more,
and he loses his army.  Without the hordes to defend it, Bubba rapidly
loses his castle and treasury, but escapes with his life.

So, what does Bubba do now?  There are lots of possibilities.  He
could sign on with another evil character as a henchman.  If he's been
disgraced badly enough, he might be relegated to a low-status position
as a guard for some other evil character.  He might become depressed
and sit in a tavern and write bad love poetry about Buffy for a while.

In any of these cases, he's no longer a "major" character right now,
but he has the possibility of becoming one again later.  Personally, I
think this sort of thing adds life to a game world -- it reinforces
the idea that these are *people*, not just playing pieces for the
admins.

>> PC involvement: Plots could be generated with PCs as objects --
>> e.g., Bubba might put a price on the head of a PC who thwarted his
>> plans, causing the system to search for NPC mercenaries who might
>> decide to go after that PC.  I'm not sure how good an idea this
>> would be -- it allows for "natural" solutions to certain problems
>> (like paladins who are misbehaving), and can reward PCs who take an
>> active interest in doing things on the mud by giving them more
>> attention.  On the minus side, however, it may focus too much
>> attention on those characters.

> Players need to be an integral part of the Plots.  If Bubba is
> trying to destroy a certain temple, He should know how powerful it
> is (How many players attend there) and know that it will make it
> more difficult.  So he will go about trying to frame others for the
> job.  Or even to go so far as to hire players to be a part of his
> plot.  Bubba hires a group to raid a guildhall, and bring him back a
> specific Item, which he will use to destroy the temple.  or
> whatever.

I'm not sure that I'd accept this as true of all plots.  First off,
there are always ways for players to become involved -- they can kill
one of the principle characters in the plot, try to prevent a plot
from succeeding, etc.  Secondly, I think there is value in having a
few plots that aren't necessarily of direct interest to the players --
not a lot of them, just a few.  For example, a good knight might fall
in love with Buffy and ask for her hand in marriage, and Buffy accept.
This may not involve any players in any direct way, but the simple
fact that some things happen without player involvement will go a long
way towards making a world seem more real.

--
       |\      _,,,---,,_    Travis S. Casey  <efindel at earthlink.net>
 ZZzz  /,`.-'`'    -.  ;-;;,_   No one agrees with me.  Not even me.
      |,4-  ) )-,_..;\ (  `'-'
     '---''(_/--'  `-'\_)   


_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev