Autumn is the time of year when I spend a great deal of time raking leaves, it also offers another perspective on the "Persistent versus Instantiated Spaces" discussion.
I am the victim of a long rambling garden beneath a dense over-growth of oak and maple... Furthermore, my compost pile is wedged at the tippy-top of a little hill, in a deeper woods: a treadmill of another sort. Now it doesn't seem to be all that long ago when I thought more of my neighborhood did their own raking. But this seems to have gone way of lawn service companies. One could meet a great deal more neighbors before. Yet, admittedly, too much socializing can distract from the autumn and a lot of good leaf raking...
Looking to GDC 2005, one attraction will be "Persistent Versus Instantiated Spaces: The Great Online Game Debate" - featuring Starr Long (Producer, NCsoft), Raph Koster (Creative Director SOE), Jack Emmert (Creative Director Cryptic Studios), Jeff Strain (Producer Arena.net). To my way of thinking, this issue is one of the great conundrums of our time, in this space. And it all comes down to (to quote from the session abstract):
Persistent spaces offer an open play space in which players can interact spontaneously with the entire player base of that game, but fail to provide the much needed guidance that help players understand how to succeed and grow within that game. Instantiated spaces offer a more directed/guided experience for players, but require much more work to create and can potentially isolate players from opportunities to discover new friends within the player base.
If indeed this is the true debate - are we having it because players are somehow giddy (especially the casual ones) and are unwilling to invest the time, acrue the bankable experience and craft the necessary social networks? Or is this debate a proxy argument. Is instancing a short-cut, a hack, indulged because the persistance vision of MMOGs hasn't yet figured out the right sort of treadmills to provide directed missions and purpose and goals and heroic deeds while still belonging to a large sprawling persistent place?
If an idealized persistent world is a place of perfect harvest - one that can transcend autumn peepers, casual leaf rakers and neighborhood socialites, the question then is, is instancing a solution on the way - or just a fix of the wrong sort. It has been argued that our MMOs barely persist enough of their collective experience to qualify as virtual worlds - diminish it further, less a world? Can it still be autumn if it is not shared?
-nathan
Gathering Leaves
Spades take up leaves
No better than spoons,
And bags full of leaves
Are light as balloons.
I make a great noise
Of rustling all day
Like rabbit and deer
Running away.
But the mountains I raise
Elude my embrace,
Flowing over my arms
And into my face.
I may load and unload
Again and again
Till I fill the whole shed,
And what have I then?
Next to nothing for weight,
And since they grew duller
From contact with earth,
Next to nothing for color.
Next to nothing for use.
But a crop is a crop,
And who's to say where
The harvest shall stop?
"It has been argued that our MMOs barely persist enough of their collective experience to qualify as virtual worlds - diminish it further, less a world? Can it still be autumn if it is not shared?"
What if autumn happened only after it was shared ?
What if instantiated spaces actually affected the overearching shared space in some way ?
Maybe instantiated spaces could be used as gates/roads to more widely shared spaces in the random-encounter fashion familiar to PnP and single-player CRPG players (obviously depends on the universe mythos/concept).
Maybe, conversely some instantiated spaced could be enabled only if special conditions were met in the larger shared space, entailing more P2P interactions ?
Emergent behavior is easier to manage in sets limited in complexity: why not use instancing to generate new areas that could froze as they are after the fact, more or less permanently, and remain connected to the world at large ?
/rambling.
Posted by: Yaka St.Aise | Nov 06, 2004 at 21:42
They mistake who assert that the Yankee has few amusements, because he has not so many public holidays, and men and boys do not play so many games as they do in England, for here the more primitive but solitary amusements of hunting, fishing, and the like have not yet given place to the former. Almost every New England boy among my contemporaries shouldered a fowling-piece between the ages of ten and fourteen; and his hunting and fishing grounds were not limited, like the preserves of an English nobleman, but were more boundless even than those of a savage. No wonder, then, that he did not oftener stay to play on the common.
-- Henry David Thoreau
Posted by: dmyers | Nov 07, 2004 at 13:35
I would probably argue that 'persistant vs instanced' is only looking at symptoms, and not the root causes.
Off hand, the problems I can think of to solve with instancing are :
1. griefing
Depending on how the instancing is handled, it can help avoid problems with some types of griefing, and I might agree that is worthwhile here, assuming we can't manage the community well enough to discourage it otherwise.
2. access to limited static (and usually perceived as required) content
This is probably the more important problem, in part because it is one of the things that leads to the griefing in the first place.
Instancing makes it so small groups can't monopolize the content, but does nothing to address the 'static' or 'required' aspects, and arguably causes problems of its own, from decreased community to less control over item rarity (assuming that anyone who wants to can spawn a new copy of rare_item_dungeon_x at will).
A truly dynamic world could have less meaningless camping*, and less opportunity to kill monster X at known location Y to get item Z. Large scale AI, with actual invasions and such, probably with high level direction from GMs, randomly generated quests, etc. The level treadmill would probably have to be cleaned up a bit as well, to avoid the need to find (as well as the existance of) an optimal place to hunt...
I suppose to some extent, it comes back to the old simulation vs narrative debate...If the goal is to allow every player to be the hero of a specific, predefined narrative, then possibly instancing is the best solution, and we try to move the closer to a single player game. To the extent that we want to go simulationist, however, I would argue that instead of instancing we should work harder on the simulation.
*: where meaningful camping would involve something that actually matters to the larger game world: defending a pass, pushing back the invading orcs(who actually stay back for a while after being pushed), etc.
Posted by: 3b | Nov 07, 2004 at 14:32
Well, instanced areas have several uses, for example:
- "load balance" virtual content so that it can be consumed by multiple persons or group simultaneously; if only the first group at the Black Mountain could kill the Big Black Dragon, the late comers are left out in the cold
- create private areas for adventuring parties where they are protected from disturbances of other player controlled explorers (the focus here lies on "private" - you and your friends)
For example, a developer would choose persistent content at places where she wants to incite socializing (taverns, inns, public meeting places), and would use instanced content where the disadvantages of persistent content would outweigh the benefits, or at places where most of the players do not intend to socialize but enjoy the game "as is" (raids, content where players have to complete a task step-by-step in a serial process).
Persistent and instanced content are nice instruments for the developer to control the flow of a MMORPG and to design the world in such a way that the players can experience the most fun; socialize at times and places where socializing is fun, enjoy the game without disturbance (griefers, gankers, player killers) at others.
Of course, this is only a part of the big picture.
Posted by: Elegantly | Nov 07, 2004 at 14:44
A few months ago, when time was more free, I had a thought exercise about designs that could integrate public vs private, instanced vs shared.
My thoughts wondered towards Dante's Inferno and the Buddhist's path towards Nirvana. From an odd perspective, instanced space could be seen as a virtual "hell," redoing the scenario over and over again until an collectively acceptable outcome results.
So, why not make this the objective of the share game/experience?
Now here come's the Experience Temporal War reference: just as we have shards and instanced spaces, the idea that everyone is engaged in a Temporal War to find a collectively acceptable result sounds like an interesting idea worthy of a game, yet have the ability to integrate the stories of the different shards and instanced spaces to the collective whole.
We can have autumn, spring, summer and winter all as an facet of the same share reality.
Posted by: magicback | Nov 08, 2004 at 08:18
I agree with Yaka in the comment:
"What if autumn happened only after it was shared?"
Instead of being afraid of instanced areas cutting off possible social extensions, have it require that you bring one or more people into said area with you in order to enter. The in game story mechanics are wide and varied, such as a guard tells you you're not strong enough alone and must have a companion before he'll let you through or some other thing involving magic, multiple puzzles or perhaps as simple as a two lever drawbridge.
Perhaps even you could have a "rating" between characters, where you can only bring characters that have a certain positive "friendly ratio" to you. This rating could be based on how many times you talk to a person, how many times you grouped with them, traded with them, etc. In this fashion you could create instanced areas that not only could encourage elders helping out newbies, but also areas that required people to have old timer friends whom they have an extended relationship with.
As you can probably tell, I'm all for pro-instances :) I don't believe there's any really bad ideas, just bad implementations :D
Posted by: Lee Delarm | Nov 08, 2004 at 08:31
I wish I could manage such optimism. ;)
My - more skeptic - motto is:
"There's no such thing as a good implementation of a bad idea, but bad implentation *will* kill a good idea."
So much so than a couple instances of bad implementations of otherwise valid ideas/features will give those a bad enough name that they'll be magically cursed to become officially, intrinsically bad for at least a generation.
PD and impacting fatigue/food in VWs are prime examples of this.
Posted by: Yaka St.Aise | Nov 08, 2004 at 10:56
I like instancing. The core issue imho is how to leverage technological resources to overcome the paradox of memorials: if one person's grave has a monument, he is remembered, but if everyone has one, no one is. In a society of thousands of people not every one can be a hero. And this is the state of affairs on Earth. But with gaming technology, anyeone can be placed in a situation that results in heroic actions. All that's missing is the externali validation that the actions were indeed heroic. I think all you need for the latter is to place the actions in a social context, one that values the actions as intrinsically heroic.
practically speaking, I'm saying that an Instance can be used to allow every player to be a hero, solo. But by having that instance take place in a virtual world, the feeling of being a hero is deepened.
Posted by: Edward Castronova | Nov 08, 2004 at 12:24
Ed>
I'm saying that an Instance can be used to allow every player to be a hero, solo. But by having that instance take place in a virtual world, the feeling of being a hero is deepened.
Per the thread/discussion cited above on "Where are the Heroes" - it seems that "heroics" and what constitutes it varies, especially when translated into a game/vw context.
However, its still worth asking, if one can be a hero if the outcome does not impact (or perceived to impact) the larger world/society at large... (different question than whether a player can be a hero solo).
In an instanced world, if one thought of it as a black box.. a player goes in and a player comes out with a grin and says "boy that was great"... I suppose that could be one sort of "heroic act" if everyone understood that most who went into the black box didn't make it... but still seems diminished by the lack of broader impact.
Now perhaps, everyone can get a crack at a shareable experience in an instance - e.g. everyone take turns at tackling some dragon, and if the dragon were a baddy that ate half the playerbase every day... well, folks might viscerally relate more...
Posted by: Nathan Combs | Nov 09, 2004 at 09:55
Skipping the ideals of fantasy heorics, I and many feel heroic when at a video arcades we get to sign and leave our initials upon attaining a top 10 score.
While not heroic, I love passing an eager kid waiting in line of a rollercoaster and egging the kid on with encouraging phrases like "that was scary!" or "oh my god, oh my god, that was crazy!"
Lastly, every soldier in IRAQ is a hero. Each street patrol they go on feels like an instanced hell, like repeating the same hellish counterstrike scenario over and over again....
Posted by: magicback | Nov 09, 2004 at 10:45
Except the mobs die faster than they respawn, and scores will eventually get maxed out.
Some buggy AI out there obviously conspires to frustrate heroes from rightfully leaving their initials on the top 10.
Bad designer, no cookie.
Suggested fix: play it the America's Army way, everyone plays the good guys, whoever dies' a wuss.
BTW, how much 'Hero points' for strafing an elementary school ?
/sarcasm
Posted by: Yaka St.Aise | Nov 09, 2004 at 12:37
My first knee-jerk reaction to the idea of instancing in MMOGs is as a clan leader: There goes one staple means for recruitment. There is, after all, nothing better for clan image than having your group do a run through a nasty dangerous territory in front of god and everyone to witness your pooled skill (or lack thereof). There's a lot of interesting beyond-clan interaction at key spots (such as boss spawn sites, etc. -- the first to be instanced, I imagine) that serves as great fodder for within-clan shared history and story.
That said, I'll also admit that, in game, there are many times when I wish to heck that I didn't have to wait in line to do x.
Possible options, off the top of my head:
* Post a 'high scores' list outside the instancing spot (discussed above). Gives bragging rights. Hell, run the instancing the way they run coliseum battles in Lineage I, where folks pay a minimal fee to watch the big battle inside.
* Let the content inside the instance be instanced as well, such that a troupe of mages encounter a different sort of boss than a troup of dmg dealers might. I mean, wouldn't it be grand to have a team of nothing but staffs take on something themselves rather than always healing/buffing the rest of us?
* Jigsaw it: Jigsaw is a classic instructional method where each person in a group specializes in their own topic (unique to the group). Let's say I pick x while others in my group pick y z p d & q. I first meet with other x specialists from other groups to tool up on it, but then later I go back to their original diverse group of x y z p d q to use that expertise in x toward something collaborative. What would this look like? Well, say each clan/group gets their own instanced version of the content (say, a boss) but then the next stage in the process requires folks who encountered different versions/instances of the content to pool their findings/drops toward something else.
People play their own ideosyncratic versions of MMOGs already. Instancing just instantiates that practical fact into the hard-coded design. My 2-cents would just be: Don't leave that individual experience hanging there so that my own MMOG experience becomes nothing more than a series of individual instanced experiences. Hell, if you want that, play a single player game. Rather, fold it back into something shared beyond the safe group I hand-pick to run the gauntlet with. I guess I'm just saying, let that instanced experience be unique to my class/level/whatever, but then make that experience useful and meaningful for social and material contexts beyond.
Posted by: Constance Steinkuehler | Nov 15, 2004 at 10:23