[MUD-Dev] Re: Group mechanics/Leadership [was: Room descriptions]

John Bertoglio alexb at internetcds.com
Mon Apr 27 21:57:04 New Zealand Standard Time 1998

-----Original Message-----
From: Shawn Halpenny <malachai at iname.com>
To: mud-dev at kanga.nu <mud-dev at kanga.nu>
Date: Monday, April 27, 1998 7:59 AM
Subject: [MUD-Dev] Group mechanics/Leadership [was: Room descriptions]

>On Sun, Apr 26, 1998 at 02:40:54PM -0700, John Bertoglio wrote:
>[ some cut & some reformatting ]
>> -----Original Message-----
>> From: Richard Woolcock <KaVir at dial.pipex.com>
>> To: mud-dev at kanga.nu <mud-dev at kanga.nu>
>> Date: Sunday, April 26, 1998 9:52 AM
>> Subject: [MUD-Dev] Re: Room descriptions, was Re: Roleplaying
>> >John Bertoglio wrote:
>> >
>> >> As the party (or members of the party) performed actions, the same
>> >> mechanism which improves individual skills would work (at a
>> >> reduced level) on the party stats. Take the ambush skill. Since
>> >> the ambush is assumed to being organized by the person in the
>> >> group with the highest ambush skill,
>> >
>> >But it could only be organised by the person in the group with the
>> >highest leadership skill ;)  You might well have a great group, but
>> >with no clear leader you're not going to get anywhere.
>> >
>> A small disagreement. The strength of the leadership skill would
>> control the size and availablility of skill bonuses. Without a great
>> leader, great bonuses do not happen. But anyone can organize a party
>> for any purpose. A person without leadership skill could not name the
>> party (the game would provide a random description based on the
>> composition of the group) or derive any party bonuses, but could
>> still carry a boat or a log. The party would be volatile and
>> disappear from the world when it disbanded. There would be a chance
>> (small chance) a leader of such a party would gain a basis in the
>> Leadership skill and be able to form non-volatile parties. I would
>> also think a party leader could designate someone in the party as
>> second-in-command and increase the chance this player would gain the
>> leadership skill or increase the existing skill. I intend to make
>> Leadership a very expensive skill, one that is probably only
>> *affordable* during character creation by using a "skill bundle"
>> which is associated with a certain life history. Aquiring it during
>> game play will be very difficult.
>How would the code determine how good a leader someone is?  What is it
>that determines whether or not the leader's Leadership (LS) skill
>increases or decreases?

At some point I will post my full skill system. But here is a summary:
Skills have a difficulty of 1 to 6 which controls their advancement rate.
The have a "buy-in" which cost development points. These points are
available during character creation and are earned through play. DP's can
be traded for stat increases or skills during creation and/or reserved for
*redemption* later in the game. An important tactical decision during
character creation is how many DP's to reserve for later. (Our basic advice
will to be to reserve some as who knows what you will want once you learn
of the *wonders* of the world.

Skills are rated on a traditional 1 to 1000 (actually .1 to 100.0) with the
ability to increase over 1000 creating special advantages for those
qualified. Improvement is infinite but diminishing returns affects both
advancement and the benefits of advancement. (Note we use 1000 to avoid
having to use decimals and therefore numeric data fields. Integers are
faster and smaller.)

Leadership is rated on a 1-1000 basis. It has a very high buy-in during
character creation because of its special powers. Improvement is governed
by a simple DO CASE structure like this psuedo-code: Note: <> changed to []
to fool HTML reading email readers.

Chance of improvement on success = 35           ; 3.5%
Amount improve on success        = 7            ; skill points
Chance of improvement on failure = 10           ; 1.0%
Amount improve on failure        = 4            ; skill points
Critical success on              = 50
Critical failure on              = 950


Create a random number
Skill Roll: MOD it into a number between 1-1000

IF Skill roll <= CharacterLeadership skill
    IF Skill roll <= Critical success #
        Create HTML for critical success
        SQL UPDATE various data tables
        Create HTML for success
        SQL UPDATE various data tables
    IF Skill roll > Critical failure #
        Create HTML for critical failure
        SQL UPDATE various data tables
        Create HTML for failure
        SQL UPDATE various data tables

Create a new random number
Success Roll: MOD it into a number between 1-1000

IF Successful Skill Roll
        CASE Successful: New random number if <= 50
        Send random Success message: ;critical success
            '[LI]Your leadership skill has improved dramatically!'
             SQL UPDATE skill record skill = skill + 5 * skillimprove
        CASE Successful: New random number if <= SuccessImprovement
            Send random Success message:
            '[LI]Your leadership skill has improved.
             SQL UPDATE skill record skill = skill + skillfailurimprove

IF Failure Skill Roll
    Successful: New random number <= FailureImprovement
        Send Failure Improves skill message:
        '<LI>You learn from your mistakes. Your leadership skill has been
        SQL UPDATE skill record skill = skill + skillimprove

There are additional elements like skills over 1000 increase crit hit
potential and decrease crit miss among other things.

This not real elegant right now, but it feels right on other skills. Note
that the only time raw numbers are exposed to the user is during character
creation.  Many people like to play with numbers during this process. We
have a STATS ON command which exposes all numbers for diagnostic and
testing purposes. It will be disabled during the commercial (?) release. We
use a large number of bar graphs to show various stats but these are
relative and do not reveal exact numeric values. One major problem with
muds is people talking IC about numeric values. It is much better to say,
"I am considered a master swordsman by many", than "I have a 63 longsword
skill with is upped to 73 by my trick sword".

>Is leadership transferrable if someone with a
>higher LS skill joins the group?

I think not.  (this is not coded yet) We are trying to build as number free
a game as possible. Acknowlegement of leadership prowess is something the
characters will learn through reputation, IC sources like handbills and
board postings, and by potential leaders *telling* about their skills.

In fact, Fearless Leader (our current leader) would have to take Grand
Poubah's word for it that he should lead the party. Fearless Leader would
have to disolve the party. Then Grand Pubah would form a party aroung
himself. Success will bring its own rewards and characters will quickly
learn who "is brag" and who "is fact" (to quote a very old western).

>If so, does the transfer happen
>automatically or does the newcomer have to run a sort of gauntlet to
>earn the trust of the group?  (Bubba can tell Boffo that he has
>dazzling LS skill, but what way does Boffo have of knowing Bubba's on
>the level?)

Doesn't know. Trust backed with deed will give Bubba a chance to lead.

>Could the leadership of the group be forcefully overtaken
>even by a group member with less LS skill than the current leader?

Again, I think not. All the leader wanabe has to do is form a new group and
say "Follow me".
Characters follow or they don't.

>the leader leaves the group, what mechanics does the group assume?
>What if the second-in-command dies?

This is a good point. The party is just a "container" in the form of a
character. When a character ride a horse, a mounted flag is thrown and his
location_ID is change to the Item_ID of the horse. The player then gains
control of the horse and moves it through the world. In a party the
characters have the in party flag thrown and they are "in" the party
character container. There are several possiblities on how to do this:

(1) If the leader is killed or gorked, command devolves to a designated
(2) If successor is unready (or unalive) the party could:
    (a) Disolve and dump everyone back into the world
    (b) Leadership could track based on the succession order as entered on
        the "party management screen"
    (c) This kind of "battlefield promotion" would carry an increased
        of improving existing leadership skill or getting an initial level.

Remember, anyone can lead a party. If the skilled leader loses control, the
party is
reconstituted around the new leader and the bonuses (or lack of same) are
created. The party
standard still exists but the party can only be rebuilt by the original
leader (who owns the standard). The ersatz leader can continue to lead and
carry the standard but as players log off they are removed from the party
and no new members can be added. When the brevet leader logs off or the
party is disbanded, the standard is returned to the leader's base of

>I think it would be possible for anyone in the group to follow anyone
>else in the group.  Perhaps that would shift the change rate on the
>leader's LS skill:  the more followers, the more your LS goes up.

This is a good idea. Will have to imp it.

>This can lead to the above Bubba joining Boffo's group and the group
>internally splitting as some followed Bubba and others followed Boffo.

This is interesting as it points to the concept (you mentioned elsewhere)
of sub leaders within an overall military organization. Like nested
containers, this could work. However, in my current model, Bubba could
agitate for leadership. His statement: "I'm going north to the abandoned
mine, who want to follow a real leader and join the Pork Rind Patrol?.". He
leaves the party and waits for followers. His chat window tells him "Barney
joins the PRP", "Boffo joins the PRP" etc.

>Should this be mirrored by creation of a new meta-character and then
>leave two groups in its wake?  If the factions remained in the same
>group, with gradually more followers forsaking Bubba and joining Boffo,
>Boffo's LS begins to drop.
Sure. Make sense.

>There is more to leadership, though, than just having people follow
>you.  How does code determine if Boffo's LS goes up or down?

Pretty mechanical at this point, see above.

>So Boffo collects a bunch of people to go kill the dragon, and they do.
>group success at any action sufficient cause to raise Boffo's LS?

Any action for which is affected by leadership skill bonuses. If Boffo is
the best cook in the bunch, the command  >P COOK RABBIT, would result in
his skill being used. However, Bubba the leader would not get a shot at a
leadership increase because the skill does not impact on the COOKING skill.
Killing dragons, for sure. Two chances for increases. Bubba's personal
leadership skill could increase and the party bonuses for the various
weapon and other combat skills in use would have a chance to increase.

>What about the entire group going into the bakery to haggle for bread?

>I don't think Boffo's LS should go up if he happens to get a good price
>on a bulk buy.

You are right. Here we just take the skill of the best bargainer. I have
always found it tedious in CRPG games to have to move the person with the
best skill into the leadership position to do anything.

> Killing a dragon can take teamwork, and perhaps that's
>the distinguishing characteristic.  Should there be a way to attach a
>teamwork value to actions?

Yes. Every skill archtype has a number of flags like:

    (1) Averaging....party skill is the average of member skills plus bonus
(if applicable).
        This would be for skills which require all the party to
participate. It would
        also apply to general party events like NPC reactions, Aura checks
and the like.
    (2) Maximum......party skill is based on highest value of the group.
This would
        be skills which only require one person, like tracking, or
    (3) Minimum......party skill is based on the lowest value. Skill like
rock climbing
        where the weakest link is the limiting factor.

    (4) Learn by doing...a skill which can be improved by use
    (5) Learn by failing...a skill which can be improved by a failure roll
    (6) Trainable...a skill which can be improved by formal training
    (7) Must train...a skill which can ONLY be improved by formal training
    (8) Party skill.... Can skill be attached to a leadership standard to
        adds during that action?
    (9) Leadership skill, does success trigger a leadership improvement
    (10) Advancement profile. How is the skill improved. Can it go over
1000 points, etc.

Player skill tables add additional skills
    (1) Mastery skill...have development points been spent to improve the
advancement rate
        in this skill.
    (2) Hedge skill....is character improvement limited because of informal
    (3) Core skill....is this skill part of a persons background (character
        and can therefore show greater improvement.

>Is it possible that not _all_ the group members participate?  What if
>Piddle wets himself every time he gets within smelling distance of a

Leadership cannot cure all evils. Our boy Piddel could still flee during
combat, leave the party, faint dead away, anything. The leadership bonus
only reduces the chance for automatic cowardice. It does not eliminate it.
Further, if I hang back (in our ranked combat system) and swing at empty
air, I will not take any damage. This might be noticed, but the leader
cannot make me attack. He can kick me out of the party (or even execute me)
which might reduce my survial chances in the Badlands. (Lone characters are
more tempting targets to random encounter monsters that powerful groups and
I would not have access to party resources, pack animals, group lifing
power, etc.)

>And there's no amount of leadership that will get him to
>thrust a sword at one...but the group needs him to help them creep into
>the dragon's lair because he's the stealthiest?
You bet. The cowardly thief is an important factor in heroic fiction. "I
got you here, now you heros do the dirty work" is perfectly reasonable
attitude to RP. (And as it turns out, RL ;)

Balancing the party system is going to be tricky as my goal is to use
parties to eliminate tedium. Once a party gets somewhere, I would want the
group to split up and go about their merry ways even if they stay in the
same area. Therfore P commands (standard commands which are modified to
work for the party will only be a subset of the full command structure.

John Bertoglio

>Shawn Halpenny
>MUD-Dev: Advancing an unrealised future.

MUD-Dev: Advancing an unrealised future.

More information about the MUD-Dev mailing list