[MUD-Dev] Re: Levelless MUDs

[Matt Chatterley]

-----BEGIN PGP SIGNED MESSAGE-----

On Tue, 9 Jun 1998, Adam Wiggins wrote:
> On Mon, 8 Jun 1998, John Bertoglio wrote:
> > From: Adam Wiggins <adam at angel.com>
> > >On Mon, 8 Jun 1998, Holly Sommer wrote:

I'm going to reply to some heavily quoted comments, since I missed the
original posts in a mail-deleting-frenzy. :P

> > >> The notion of converting to levelless MUDding is floating around
> > >> the MUD I admin, and I am curious as to how this is done elsewhere
> > >> (since I personally have never set foot in a levelless MUD).
> > >
> > >Levelless muds are hard to come by in the GoP field.  Skill-based muds
> > >which have levels that do almost nothing (except for providing players
> > >with easy-to-understand feedback) are much more common.

Hard to come by, but /not/ hard to make. If the definition (or
interpreatation) of GoP is expanded to make power a little more abstract,
then skill-based becomes the logical choice. You can begin to vary from
true HS/GoP gaming at this point of course, but, you can still interest
the same gamers by allowing similar play on a different scale (for
instance, the notion of 'gangs' or 'armies').

> > >Right off the top of my head, the two completely level-less muds I'd check
> > >out would be YaMUD (if it's still around, it goes up and down) and, of
> > >course, Ultima Online.
> > 
> > UO has levels. They are described by prose instead on numbers and are (as
> > of this week) defined by a sophisticated two dimentional matrix...but they
> > are still levels.
> 
> There's two conflicting definitions going on right here.  One is the
> conventional level notation, a single number, usually with a small range
> (ie 1 to 30) which, by itself, describes a given character almost
> completely.  This is the form of 'level' that I assume Holly wanted to get
> rid of.

Yup; perhaps we should refer to these as 'conventional levels' to avoid
confusion when saying 'a high level player' which really means 'a player
who has, by whatever measuring stick is typically used for this particular
mud and in whichever areas are relevant to the context, obtained a level
of ability, knowledge, respect or power in any other sense allowing him to
be very effective in this situation'. Bleh. :P

> The second one, the one you describe, is the definition more commonly used
> on this list.  Since many (most?) of us are working on games without the
> conventional level notation described above, the term "high level" or "low
> level" has come to take on new meaning.  In this case it's just a general
> term to describe how experienced a given character is.  Tacking a 'class'
> onto it, such as 'warrior', narrows down where they have that experience
> invested, which makes conversations about other topics quite a bit easier,
> despite the fact that there is no character->GetClass() == CL_WARRIOR or
> character->GetLevel() < LOW_LEVEL.  The second definition is broad enough
> that you could apply it to almost anything, anywhere.  I might be
> considered a high-level programmer but a low-level football player.
> I doubt Holly really wants to get rid of *any* method of tracking this
> sort of "level", since that would render characters into fairly simple
> shells, which are not suitable if you desire (non-social) character
> advancement to be possible.

All quoted, for posterity. Agreed. I frequently make references to
character types as 'fighter'. This confuses some of my staff - we have no
classes, and our guilds are strictly localised to the towns and regions
that they exist in; they are not unique, and membership is not always
exclusive.

There are (to my mind) four *basic* types of character, into which
everything else can fit: Fighter, Thief, Mage, Cleric. The traditional
four classes. You can argue almost any type of medievally based (or
otherwise) character to belong to one (or more) of these groups; and as
long as you aren't *too* finnickety, it provides a quite useful grouping
tool in that you can summarise with 'and _fighter_ characters should
approach the problem _this_ way, while the _thieves_ go _that_ way, and
_mages or clerics_ approach it from the _other_ angle, _here_.'
 
> In that spirit...
> 
> > >Muds very skill/spell/gold/eq-oriented to the point of rendering levels
> > >almost moot include Legend and Arctic.  On Legend, levels are really only
> > >practice points (one point per level).  On Arctic, levels only affect how
> > >high your skills max out - if it wasn't for that, I'd never bother to gain
> > >levels there.  (As it is, you don't "work" on gaining levels, you "work"
> > >on getting up your skills and learning new skills, and gain levels by
> > >accident.)
> > 
> > This is how levels should function.
> 
> *shrug*...'should' isn't necessarily the word.  IMO if you're going to
> have levels at all (usually an old hang-on from a previous system; Arctic
> is nearly six years old, and at the time they started levels were
> ubiquious in the mud world), this is the way to do it.

'Should' is never appropriate. If you *are* going to use conventional
levels, do it in a convential manner - the only reason to really use them
is to be 'backwards compatible' with some player mindsets. Remember, we as
mud administrators fall head over heels for concepts such as 'skill based
gaming', but not all players agree. If (as at the start of this message)
your own definition of level is different, then of course, you will have a
different implementation. Hmm, Adam said that far more concisely.
 
> > After elementry school, people stop
> > talking about grade *levels*. People who have just received a Phd. do not
> > speak of graduating from the 19th (or whatever) grade, they describe their
> > specific accomplishment.
> 
> True to a certain extent.  People still tend to ignore specifics...for
> instance, caring more about what school someone graduated from than what
> they majored in, or just saying, "She'll know, she's got a PhD after all"
> even when the person in question has a PhD in math and they have a
> question about botany.  Of course, there's a reason for this - generally
> someone that is more experienced in general will have a higher chance of
> knowing the answer to any given problem/question than someone with a lower
> level.

PhD also seems to represent a fairly specific field in many cases; you
can't apply the fact over all areas of maths (although you would have a
high 'backing' skill in it), and certainly not to all other subject areas.
I propose a skill tree diagram, containing a limited list of subjects
(you'll probably disagree with my arrangements, of course). Very limited
list.


		  /--> Foreign languages ..
Linguistics 	-->
		  \--> Native language ..


					  /--> Botany
		  /--> Biology		-->
Sciences	-->			  \-- ..
		  |                       
		  |			  /--> Computational
		  |--> Physics		-->
		  |			  \--> Theoretical
		  |
		  |--> Chemistry	--> ..
		  |
		  \--> Maths		--> ..

		  /--> Visual art ..
Arts		-->
		  \--> Dance & Drama ..

Hmm, very crappy. Anyways, it conveys the general point. Maths belongs to
the science branch (to me, its somewhere amongst the 'base' skills of
science at a low level, and at a high level, a science in its own right -
within physics it's used as a language, and it can be very expressive and
also very confusing ;). To advance to a high level in maths, you would
require basic science. This would make development of other sciences
easier, and likely, and so forth. You could be utterly clueless about
Arts, and Linguistics, but able to converse at a basic level in any
science. Botany would be too specialized, and out of your league without
more specific study. A skill web would probably show this better, but I
can't be bothered drawing one (they probably work better in 3d, anyway).

> When it comes right down to it, people still care about those simplistic
> metrics of your accomplishments, as they serve as general indicators of
> your much more complex and more difficult to define set of abilities.  Ie,
> a company hires an engineer because he graduated from a good school and
> has 'worked as chief engineer at Newtech Systems for 3 years' on his
> resume.  Are they hiring him for those reasons?  No, they are hiring him
> because he's got many skills related to the task at hand.  But those
> skills are much harder to define, display, and support.

Absolutely; you have to get gratification for your achievements from
somewhere, else why bother to achieve, within the game context (this even
applies to the real world to a degree)?

Conventional levels provide a good, simple metric. That was why I
originally had them. I ended up having to hack code to still support
levels, and to make them advance, so in the end I got fed up and threw
them away. We're developing abstract gratification systems now - buying
land, being recognized as nobility, grovelling peasants, singing bards,
setting up things of your own, and making your mark within the game
noticeably (so that a new player could move through and hear of your
exploits).

Levels also provide a measuring stick for 'virtual penis waving' amongst
players (please excuse the male-orientated phrasing, but also note that it
is probably mostly male players who find this necessary). I'm quite happy
for players to fight it out to decide who is tougher. Our unarmed combat
can be happily carried out without high risk of fatality as long as
reasonable controls are set by the players, for instance, it would be a
very bad idea to enter an intended non-fatal fight with the Kung Fu,
potential instant-kill spell 'cobra strike' selected as one of your
standard attacks. Feel free to pile in with less impressive moves, and
some defences, to knock your opponent on his ass and make him feel
foolish, or to beat the crap out of him. Just remember to stop when he
doesn't get up, or the guards will get /really/ pissed. Or don't do it in
a town area, if you don't like being arrested.

But, how about other less risky ways to measure? How about a drinking
contest? Or arm-wrestling?

Is it feasible to change the measuring stick from an abstract one to one
which can be related to far more easily?

> > All games which allow for improvement have levels.
> 
> Again, a simplification I'm ignoring in order to answer Holly's post in
> the spirit in which it was asked.

See my personal definition of level above. If you use numbers to track an
ability and display the result in /any/ form, you have levels. Players may
be aware of them or not. If your game doesn't allow improvement, you can
argue you still have (socially orientated) levels. Not really a useful
point. ;)
 
> > Those designed by people with limited imagination (or a mechanistic POV)
> > use numbers. It is far more satisfiying to look at a character whose war
> > necklace has 4 dragon incisors, a glowing orb, 16 human ears with earrings
> > of famous warrior clans and gold wedding band. This, along with the proper
> > facial and body tatoos would suggest that this bad puppy is a high level
> > character with a bad attitude. He doesn't have to claim to be a level 65
> > fighter...he either is one or is putting on a good show (which is itself
> > interesting from a game POV).
> 
> Absolutely.

Mais Oui. It's this sort of thing which I hope will happen within my game.
There are very few external indications from an OOC point of view as to
how tough someone is. Experienced characters with appropriate skills can
size up an opponent (one with physiological experience might be able to
guess at strength from the set of the body, those with martial arts
experience might be able to size up the threat posed by a current stance
and so forth). Attire and appearance are /important/. If you look like an
easy target, you'll tempt people to find out /how/ easy. Look like a bad
ass, and they'll find someone else to pounce.

Appearance isn't everything of course, and you might want to be able to
back things up. Or alternately, take the other approach - be a wolf in
sheeps clothing, and try to sneak past without trouble. I'd say that any
bullies picking on a Shaolin Monk would be quick to laugh at his strange
attire, and unassuming appearance. Until he put them through a nearby
wall. Reputation counts.
 
> > >As both a player and an admin I despise levels.  I find it far easier to
> > >design, code, and play a game without such kludges.
> > >Be forewarned,
> > >however, that it makes game *balance* far more difficult to tune.
> >
> > Adam: Why would this be? Unless you are talking about things like players
> > can only fight other players within 5 levels or other artifical
> > stuff...those kind of meat axe tuning short cuts are clearly easy to tune
> > but the results are usually a joke.
> 
> Well, in this case I'm speaking from direct experience, but you don't have
> to take my word for it - ask Raph.  Anyhow, a simple glance at the problem
> will reveal why this is so: in one case you have a single number to
> adjust, plus some secondary values that relate directly back to that
> number.  In a skill-based system, stuff is much more distributed.  There
> is a large number of values, all of which affect each other in ways that
> are difficult to predict.  In addition, a more complex representation of
> the characters (and, indirectly, the game world) simply implies more
> dependancies and more subtleties that can elude the designer.  For
> instance, ditching levels means you now need some way to regular skill
> learning.  Okay, so you make skills go up with use.  But, players just sit
> around and spam skills to get them up, so you make skills only go up when
> being used 'against' something of slightly higher skill level as them (ie,
> a 45 climbing skill against a cliff rated at a 50 difficulty has a good
> chance to go up), and regulated by time (say, only once every fifteen
> minutes).  This reasonably small change makes balancing advancement an
> order of magnitude more complicated.  Before you'd just give them less
> "practices" for each "level".  Now you have to go and push and pull
> numbers, and more specifically formulas which can get quite complex if
> you're ambitious.

Adam is to my mind, utterly correct. The more complex that your world
design is, the more complex your world balance must be. 'This is a level
10 sword, the player must be at least level 10 to wield it' is a simple
concept to handle and to enforce. 'This is a quality 37 sword, weighing
15kgs, of type broadsword (standard) with no enhancements.', the player
wields it with 'y=f(x)' efficiency.. .. things get much more complicated.

I recall having to scrap several formulae from game internals, since
differential equations were just too scary to bother with for such things,
and starting over to simplify the situation. Bear this in mind (Holly, and
anyone else) - simplify where relevant. If you're putting together a
formula for damage from falling, you can probably ignore air resistance
damage on the basis that if you fall far and fast enough to get damaged by
air resistance, you aren't going to be in any state to ponder how much
damage it did to you. If you were falling in the USA, you could even be in
several states.

On the other hand, if the player is falling towards a magical crashmat
that will save him, the air resistance damage /does/ matter, since it's
the only really relevant damage.
 
> Not trying to argue against this at all, of course - anyone who's been
> reading the list any time for the last three or so years will know that
> I've always been a strong proponent of the stuff above.  But I can
> understand how someone that wants to have simple character advancement in
> their mud without spending a lot of time working on it would stick with a
> simple metric like levels, giving them time to focus on other areas of the
> game.

Aye.

- -- 
	-Matt Chatterley
	http://user.itl.net/~neddy/
"You may say I'm a dreamer, but I'm not the only one.." -John Lennon (Imagine)

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
Charset: noconv

iQB1AwUBNYPOOEsWGgRW8+MZAQH6PgL/RHEgBtAmKhXg8SjGLVrTSBDSoAZoufkG
prsse6zgydlP38HSTTmvxzeZBQ+gSIMkDZjvUer7krLae+jAfcYsXPDmKqWSB5z8
hWgHdwlZdqN1+v0rOPzWMxRa1ZOkD047
=ApCp
-----END PGP SIGNATURE-----