[MUD-Dev] Re: CGDC, a summary

J C Lawrence claw at under.engr.sgi.com
Wed Jun 3 12:19:38 New Zealand Standard Time 1998


On Mon, 1 Jun 1998 13:15:14 +0100 (BST) 
Marian Griffith<gryphon at iaehv.nl> wrote:

> On Tue 26 May, J C Lawrence wrote:
>> On Thu, 21 May 1998 17:40:10 +0100 (BST) Marian
>> Griffith<gryphon at iaehv.nl> wrote:

>> > In <URL:/archives/meow?group+local.muddev> on Mon 18 May, Travis S. Casey wrote:
>> >> On Friday, 15 May 98, Chris wrote:

>>>>>> The nominative value of progress is also significant.  "I know
>>>>>> I'm playing well becasue I keep gaining score and advancing in
>>>>>> levels!".  Its a self-referencing, self-supporting, and
>>>>>> self-defining closed system which defines both the goals and
>>>>>> the expected accomplishments of players.  Its actually almost
>>>>>> definitionally impossible to have an open-ended or user-defined
>>>>>> game with this sort of system in place as point rewards are
>>>>>> tied to activities, and thus implicitly devalue and thus damn
>>>>>> all non-rewarded activities.  ("Thou shalt have fun killing
>>>>>> monsters 'coz you get points for that, but thou shalt NEVER
>>>>>> poison the water supply to kill all the monsters, or build
>>>>>> towns, or RP, or set up wineries and blacksmith shops because
>>>>>> you get no points for that.")

>>> So the challenge of a roleplaying oriented mud is to provide an 
>>> alternative system for levelling that does not bias towards one 
>>> particular type of behaviour.

>> Any system is going to be biased.  You can't avoid it.  You can
>> provide multiple biases, but that's merely multiplexing the
>> problem, not solving it.  The nasty bit comes when sufficient
>> multiplexing is deemed "good enough".  When does that occur?

> *grin* Actually I was trying to say -ONE- particular type of
> behaviour.  I could be happy in a game where you could actually play
> in a different style without messing up the game goals.  Like a game
> where a healer is expected to do just that and not to go out and
> kill the monsters.  

> I do not understand what you mean by
> multiplexing though and why it is such a bad thing.

Any system is going to be biased towards the machanics which lead
towards its goals.  It will encourage actions which it rewards and
sicourage or penalise other actions.  If you have a single goal
(hopefully), rewarded actions will lead to progress toward the goal,
and unrewarded or penalised actions lead away from the goal.

If you have multiple goals, then you are going to have multiple sets
of these relationships. of actions that are rewarded because they
progress towards one of the goals, and of actions unrewarded or
penalised because they don't progress towards any (or enough) goals.

The latter case is the same as the first, just more of it.  If you
think about it in terms of a river channel crossing a plain, where the 
water flows down the channel because that's the quickest way downhill, 
then in the latter case we have multiple river channels which cross
and recross as they weave about the plain.

The point is that in neither case have we removed the bias for the
channels.  Water still doesn't want to flow uphill or sideways between
channels except at the crossing points.  We haven't changed the
machinics of the bias, we've just added a lot of instances of bias.
Thus we've multiplexed the problem: we've not solved the problem we've
merely multiplied it in the wan hope that its sheer numbers and
internal interaction will hide the fact of the remaining underlieing
bias.

Not necessarily a Bad Thing.  You just need to be aware that you
haven't solved the problem, you've merely masked it.

>> So, instead you build in enough goals, and enough reward systems,
>> with enough cross-pollination and cross-currency among them that
>> the total set of valuable (per player viewpoint) permutations is
>> "large".

> <--- imagine totally blank expression

Okay, in the simplest model you can go out, kill monsters, gain XP,
gain levels, and eventually max out.  There is a single goal, a single 
progression to get there, and a single method (set of actions) to move 
from any point to the next in terms of advancement.

Now take a more complex model.  We have two goals: Be a super XP/level 
dude, or be a super healer type.  There are now two seperate
progressions to achieve each goal.  Killing monsters doesn't do
anything for yuor healing ability, and healing wounded warriors
doesn't do anything for your killing-monsters ability.  However, as a
healer you need people to go out and fight monsters so that they will
be injured and you can heal them.  Equally as a fighter you need
healers so you can return to the fray more quickly and kill more
monsters and thus progress more quickly.

This is only two goals, but we're already getting interesting
cross-mechanics between players attempting to achieve each goal.  

Now imagine that we have a whole slew of different goals, perhaps ten
or twenty major ones, along with long and short term goals (eg short
term goal: kill Tiamat, medium term goal: save the kingdom from the
Orcs so I can get the King to give me a super-nifty sword, longer term
goal: gain ownership of a chain of stores that provides me with a
residual income for my adventuring so that I can outfit and support
entire parties, long term goal: become King).  That plain we talked
about before is now criss crossed with channels going every which way.
At any one point or location the number of pssible channels to follow,
or to side-step into with reasonable effort becomes large.  Players
also start to select __sets__ of goals, rather than a single goal with
its simple mechanics that they pursue in ignorance of all else.

A new player entering the world no longer has a simple goal list
presented to him: Become a super monster killer, or become a super
monster killer or a super healer, but a large range of possible
variations on sets of goals, with each goal having its own
sub-mechanics.  

Sort of like real life, not everybody ends up a Batman-style
super-hero, and not everyone ends up a fireman or a gun-toting army
dude.  Instead there are many possible goals, and many flavours and
routes to each goal.

When does this great flexibility and complexity kick in?  It depends
on the viewpoint of the players.  The number of possibilities they
perceive is the range of permutations they are aware of (likely a
subset of the total range).  However, once that number of
possibilities becomes "large" in their eyes they largely cease looking
for the boundaries to the range of possibilities and instead
concentrate on selecting the set of possible actions they wish to
pursue.

>> Yup, carving anorexic tree god statues from ogre brains may very
>> well not be rewarded by the game, and will thus be selected
>> against.  Ditto perhaps for excellant deep sea fisherman skills, or
>> being able to live an entire life balanced atop a hermit's pillar.
>> Ya gotta draw the line beyond which you just don't care any more.
>> Its an arbitrary.

> I see no problem with that, as long as the things that you -can- do
> are sufficiently varied.  


__Exactly__.  See above.

> I am not speaking for UOL as I never played that, but for most muds
> that I am familiar with there is only one thing to do and nothing
> else matters.

Bingo.

>> We enter the question of the definition of a game.  Is a "game"
>> which has no goal (ultimate, sequential, or otherwise), still a
>> game?  Arguably, no.  This is the crux of Bartle's dismissal of the
>> entire Tiny-* clan in his MUD survey.  While they're not games per
>> se, he never really defines either what they actually are, or what
>> drove (and drives) their success.

> There is a difference between having no goal and allowing the
> players to define their own goals.  I think that what makes mushes
> intereting to play is that they do not have a goal.  They are a
> stage where you have a limited control over your environment and
> they allow you to do what you like.  Within the limitations of the
> theme.

I'll counter with the idea that as soon as a player (of a MUSH or
otherwise) defines a goal for himself in that world, then it becomes a
game for him.

--
J C Lawrence                               Internet: claw at null.net
(Contractor)                               Internet: coder at ibm.net
---------(*)                     Internet: claw at under.engr.sgi.com
...Honourary Member of Clan McFud -- Teamer's Avenging Monolith...




More information about the MUD-Dev mailing list