[MUD-Dev] Alignment (very long)
Paul Schwanz - Enterprise Services
Paul.Schwanz at east.sun.com
Thu Apr 20 16:28:44 CEST 2000
Richard Ross said:
> Has anyone else tried
> non-standard alignment systems, and how have they worked out?
I've only played around in my head with this system I'm about to share, so in
the real world, there could be much that would go wrong.
Most of the simple binary alignment systems leave me wanting more in the way of
interesting moral decisions. I thought that the system that Christopher Allen
posted from DartMUD was fascinating and will look into it in more depth in the
next week, but it is the first I've seen that even attempts to explore some
deeper moral issues.
For the most part, moral consequences, such as guilt experienced when one's
actions are contrary to one's values, are not modeled in the MMORPGs I've seen.
I think that this is an oversight, since such a system may well lend itself to
curbing PK and other anti-social (when abused) actions by implementing natural
moral consequences to actions.
I see character morality revolving around what a character values. Consider
your run-of-the-mill, goodie-two-shoes, treehugging character. He walks around
ant hills and raises bunnies in his back yard. He places a _very_ high value on
life. We might _think_ that he could never kill anyone, but what should we do
if he picks up a knife and goes to stab someone? I see three possibilities. 1)
We do absolutely nothing. 2) We don't allow his plan to be actualized. Or, 3)
we give him the freedom to act, but present his character with natural
consequences that (hopefully) are appropriate. Perhaps he is horrified by his
own action. Maybe, he is hounded by guilt and is unable to function as well as
he did previously. Perhaps for the near future, he is reduced to a helpless
emotional blob at the very sight of a knife.
(NOTE: The fact that there may be moral consequences to an action in no way
denies the existence of additional consequences imposed by justice systems,
lynch-mobs, etc.)
While the last option (3) will be the most complicated and hardest to program, I
think that the game will have more depth _and_ harassment of goodie-two-shoes
will be curbed to some degree by the knowledge that, if they ever decide they
are willing to suffer the moral consequences, they may at any time attempt to
take the life (or something less drastic) of the harasser. Someone once said
that a suprising amount of courtesy is rooted in the knowledge that someone else
can punch you in the mouth if you become too rude.
To bring this concept down to something that might be programmable, I have
identified four basic things that are valued in an MMORPG. They are health,
wealth, information, and power. (The issue of time has been addressed
elsewhere.) Significant actions in the game can be seen as either giving or
taking in one of these areas. Some actions give and take to a greater degree.
This concept could be charted to look like this:
................GIVING(+).................TAKING(-)
(H)EALTH........Nurturing(H+).............Conquest(H-)
(W)EALTH........Generosity(W+)............Greed(W-)
(I)NFORMATION...Teaching(I+)..............Secrecy(I-)
(P)OWER.........Loyalty(P+)...............Betrayal(P-)
If a character values life or nurturing, then he is health-giving(H+). If he
values subversion or betrayal, then he is power-taking(P-). At character
creation, the player is presented with questions regarding the values that his
character will posses. Does he prefer to heal and grow things or to overpower
and subjegate? He will also be asked questions to find out what priority he
would give to each of the values he holds. The character will end up with a
values profile that might look like this:
Bubba: W-4, H+3, P+2, I-1
This character places the highest value on gaining wealth. Next, he values
nurturing and healing living things. To a lesser extent, he is loyal. Finally,
in most cases he is more likely to hoard secret knowledge over teaching wisdom.
Perhaps someone with these values would make a good merchant. It probably would
not be wise for Bubba to attempt to make military endeavors the focus of his
life.
Once we give every character a values profile, the next step is to give each
significant action in the game an appropriate profile as well. Knocking another
character unconscious and taking a small amount of gold from them might result
in a values profile of H-1, W-1. Killing someone from your own town and
removing from their body a priceless gem might result in H-5, W-4, P-3. To find
out how well an action lines up with a characters stated value, you could simply
multiply the value categories of the person and the action and sum the results.
If Bubba the merchant, with the values given in the example, were to kill
someone from his own town and remove from their body a priceless gem, the math
would look like this:
Character....Action..........
Profile......Profile...Result
W-4.....*....W-4.......+16
H+3.....*....H-5.......-15
P+2.....*....P-3.......-06
...............Total = -05
(NOTE: -I- is irrelevant to this particular action.)
In general, this action goes against what the merchant character values, so he
will face some moral consequences. The -05 result can be used to determine the
extent of those consequences. At the most simple level, whatever skill or
experience is tied to the actions can be multiplied by the -05 so that Bubba the
merchant will actually _lose_ skill and experience as a result of his acitons.
(Morally speaking, he is troubled by his act and will not function as well,
especially in the skill areas that are associated with his action.)
That's the basic idea, but there are numerous opportunities to use these values
in other interesting ways. A person's values profile could be dynamic so that
cursed items would encourage the recipient to role-play the curse. (i.e. as
Frodo wears the ring, his H+4 might move toward H-4 over time. Or, in Jordan's
Wheel of Time, Fain's dagger could make Mat move toward W-3.) If Frodo or Mat
"resists" the curse (by not playing in accordance with altered values), they may
suffer moral consequences which result in degraded skills. Also, I think it
would be a good idea to have values normalize over a long period of time so that
the character's values would shift to fit the play-style of the gamer. Perhaps
it would be beneficial as well as interesting to allow quicker changes to values
to mirror different types of "conversion" experiences. However, I think that
such experiences should only be the result of a complete "moral break-down" such
that most skills become negligible and must be built back up.
In addition, reputations can be built on the foundation of values. If
significant actions have values asigned to them, then those characters who
"observe" the action can make assumptions about what the "actor" values. As the
observer then moves about and interacts with others, his observation can be
shared to other characters. After a while, a pretty realistic picture of what
others value can be pieced together to generate reputations. This system could
be expanded even farther by having a character's evaluation of an action (or
action observed and reported by another) filtered through that character's own
value system in a way that amplifies or diminishes his own reaction. Also, it
might be interesting to experiment with making the opinions or actions of VIPs
(i.e. high level characters) carry more weight when calculating reputations.
Even further custumization might result in thresholds being set so that
character's automatically "talk" only about the characters that they feel most
strongly about, dependant upon social setting. But if asked about a particular
character, they could give the specific information requested (or perhaps not,
if they decide they don't like you).
Each player could also be given the limited ability to affect the reputation of
others by entering up to five names on their "short list" whom they talk either
positively or negatively about when interacting with others. In addition to
using this as a way for players to give some input into evaluating other players
(which could help cover up some of the loopholes which might crop up in a
completely algorithm-dependant moral/reputation system), slander and politics
could take on new meaning in the game.
When you stop to think about it, alignment is simply your reputation with a
particular deity. This system lends itself quite naturally to a (more western)
system of one God and one anti-god. Simply assume that "it is more blessed to
give than to receive," and it becomes straightforward to calculate alignment
based on the giving and taking values and priorities. Simply take the values
profile of the character (X+/-4, X+/-3, X+/-2, X+/-1) and sum, adding positive
values and subtracting negative values. If you then divide by two, you end up
with a nice +5 to -5 morality/alignment scale. In order that the gamer may
_care_ about this, tie the magical resources of "inner light" or "inner
darkness" to this scale such that the extremes have exponentially more
deity-granted power than the middle.
There are many interesting moral subtleties that I am still trying to work my
way through. One is the fact that two people can have a high regard for life,
yet act very differently upon their values. For instance, I think the Amish
place extremely high value on life, thus their pacifism. However, the patriot
might place a similar value on life, but choose to fight to protect it. Perhaps
one values life in a passive way and the other in a more active way. Or maybe
one values life-through-peace and the other values life-through-chaos. Perhaps
a lawful/chaotic modifier will help resolve this.
The other issue is the issue of motive. If an assassain does a bunch of really
good deeds so that he can gain the trust of his mark, is he really doing good?
Should the system be changed to handle this, or should the "good" actions
detract from the assassain's "evil-ness" and adversely affect skills?
Anyway, I apologize for the length of this. I'm sure you were looking for
something a lot simpler. :-) A majority of my time spent thinking about how to
make MMORPGs better has revolved around these issues, so I am happy to share
them with developers and would _very_ much appreciate any suggestions for
improvements or even attempts to break the system. I'd also love to see them
tried in a MUD and would be very interested in the results.
--Phinehas
-----------------------------------------------------------------
"All things are permissable,
but not all things are expedient."
-----------------------------------------------------------------
_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
http://www.kanga.nu/lists/listinfo/mud-dev
More information about the mud-dev-archive
mailing list