Traits, stories, and holy grails

 Posted by (Visited 12080 times)  Game talk
Jan 052006

Mirjam Eladhari has an interesting talk up on her blog that considers what she calls “the Holy Grail” for MMORPGs: making stories meaningful for thousands of players.

The talk isn’t about events, it’s about shaping a given player’s path through the game in meaningful ways, to reveal true character. The eye-catching idea is to use personality assessment early in the character creation process, and then set the traits invisibly on the avatar. Then the traits can be used as inputs into the game systems.

Leaving aside the critiques of personality trait systems that are out there, it’s still an interesting idea that echoes stuff like the Gypsy in the Ultima series. That merely defined a class, but a class itself is also a strong shaping of a journey though a gamespace, so there’s clearly something to that notion. Interestingly, I think players these days would be very resistant to the Gypsy, because class (like character race and gender) is more of a Rorchasch test for playstyle these days.

When working on the Love Story Challenge, I was trying hard to capture actual story, as opposed to plot coupons. The algorithm for the romance novel generator is therefore dependent on ascribing personality characteristics to the players. I used a heck of a lot more than 5 traits, too:

  • Not ready to settle down
  • Overly dependent
  • Too proud to accept help
  • A social snob
  • Thinks they want adventure
  • Scared of the world
  • Rigidly conservative
  • …and dozens more

Each of these was then broken out into a typical character arc, so that each character was round, rather than flat. These arcs were based on basic storytelling theory patterns: some stage-setting, an epiphany, a reversal, a denouement, etc.

  • Not ready to settle down
    • Evidence of past refusals to settle
    • Realization of need to settle
    • Attempted settling but flees
    • Settles
  • Dependent
    • Evidence of past codependency
    • Anger at self over inability to take action independently
    • Fail miserably at trying to take independent action
    • Forced to take independent action and it works

The rest of the alrogithm involved interleaving story incidents that fit each character with the skeleton plot of a typical romance novel. The algorithm proved surprisingly good: you could get it to spit out entirely plausible romance novel outlines with arbitrary numbers of characters.

But as a tool for generating quests in MMORPGs it wasn’t very useful, because you can’t assign traits to players as they enter the game. Rather, you have to detect traits they are expressing.

Perhaps the idea of sticking a personality assessment in quests in the newbie portion of the game is a partial answer to this. Perhaps it could be made part of a character creation process. (“Tell us about your character’s childhood. Was she afraid of spiders, the dark, and slimy things?”)

Either way, if, as Bartle argues in Designing Virtual Worlds, virtual worlds are about a player’s exploration of self, then we ought to be finding ways to aid in that design goal. The games these days don’t seem to take that as a design goal at all, replacing it instead with other forms of self-actualization (power, as discussed previously, but also group identity). The narrative, including the personal narrative, isn’t explicitly supported very much outside of a few badge systems, an optional character description, and maybe a roleplay forum or two.

And yet, the prevalence of alt-play suggests that multiple narratives of self is something that players deeply value (hardly surprising, it’s something that people tend to crave in the real world as well).

There’s also market pressures that push against having things like reversals in our player’s narratives. Reversals, after all, are unpleasant…

So, is this the Holy Grail? How many players want to learn about themselves as they play, versus the players who would rather not think at all, and instead take a break from the real world?

  14 Responses to “Traits, stories, and holy grails”

  1. Blogroll Joel on SoftwareRaph Koster Sunny Walker Thoughts for Now Sex, Lies and Advertising

  2. I can certainly agree that great story without gameplay wouldn’t be a great attraction for me. Perhaps this is the reason Façade quickly became boring: sure, there’s story (in the traditional sense), but there’s no game-play. (As an aside, Raph wrote about the potential for systems to drive the kind of story that attracts me, that is, story about me. I won’t dwell on this article now. But it’s an interesting read.) LostGarden, in Ze Story Snobs

  3. Bartle?s 5 most important folks in virtual worlds [IMG] Posted by Raph’s Website [HTML][XML][PERM] on Fri, 20 Jul 2007 03:12:38 +0000

  4. I never fully agreed with Bartle in that virtual worlds are about exploration of self. I view it more as a sense of a desired self. The play of alts furthers my view of that to some extint and somewhat is disregarded in that one has a ‘main’ which they relate to and rest are solely for the game value.

    That said you could be far more creative than a series of questions really. Look at what mobs they target the most (after taking care to be sure there’s a selection of identical stats/placement except for name and visual mobs). See how many tells they send to different people. How many on the friends list. How many ignored over a period of time. Active vs inactive time. You really could embed a ‘personality’ (or behavior depending on your view) test into the entire system and it’d adapt over time to changes in the person (people change, constantly even if they are not fully aware of it).

    All that said, virtual worlds are on the decline if you define world as a highly flexible ‘choose your own adventure’ style universe where almost anything is possible. With SWG’s demise from the world catagory to the badly implimented FPS catagory you’re pretty much left with games that have hints of worlds, UO, maybe Linage (I wince a bit at that but it depends on how flexible you require the character ‘classes’ to be to qualify) and some text muds. Eve may qualify but without more avatarish interface I hesitate to classify it as such.

    The problem I personally see is that there is a world or a game and the two haven’t mixed all that well honestly. Given the market I see the games that are the most successful tend to orient towards everything but self exploration (the you are a hero here’s your heroic task for the day with no chance to botch it for great harm or turn). While from a social perspective the depth probably should be there the market isn’t supporting that view currently. Perhaps we’re a decade too early.

  5. […] Eladhari has an interesting talk up on her blog that considers what she calls “the Holy Grail” for MMORPGs: making stories meaningful for thousands of players. […]

  6. Considering the powerful cultural context these manifestations come from, we very well be merely “a decade too early”. In America, I see the general trend of the public to be “hold my hand and guide me through to the light”. I can’t speak for other countries nearly as well, but the Eastern giants are in the process of gathering themselves into a “jump the lightspeed”, as it were, and Europe, well… I’ll just say I know nothing about Europe and shut up. It’s hard to call this an environment conducive to relaxed and peaceful intellectual exploration of identity. Most people who have that aren’t playing; they’re innovating in the thick of high tech. Despite my interest in games and building worlds, that’s where I am, too.

    In one sense, I strongly agree with Bartle that the path through one’s experience in virtual worlds is a Hero’s Journey. I mostly disagree with him on the particulars or the overall effective meanings. It may be a good idea to set up historical snapshots: “This is who you were a month ago. This is who you were a year ago.” But of course, these snapshots have to be useful. It can’t be “You were a 5th level paladin a year ago, and now you’re a 50th level paladin.” That’s meaningless. You have to inject the nature of identity. Perhaps give them a logbook that they can use, but not show anyone else. “Today, I met my new best friend, Arthas, and we had an amazing time slaughtering critters and dividing loot.” “A year ago, Arthas was my best friend, but now he’s been turned to the dark side, and I’m very sad. I wonder if something happened to him I don’t know about.”

    It surprises me that this idea goes against the grain of “automate everything”, which I’ve always favored, but hey… I’d be the first one to say my hero’s journey isn’t done. =)

  7. In some sense, the contemporary selection of MMORPG, race and class are already a first (tiny) step towards the personality profile. Single-player games like Fable also include a simple good/evil personality profile.

    Personally, I think the more relevent use for the profile is to allow people to meet other people, although customizing content to a profile is certainly important. If I tell the world that I’m LFG, the world should be smart enough to hook me up with other people I’d enjoy hanging out with. Amazon’s book recommendation AI/datamining comes to mind too.

  8. “PlaneScape: Torment” allows a great deal of personality expressiveness, and is responsive to the personality that you present to it.

    “Either way, if, as Bartle argues in Designing Virtual Worlds, virtual worlds are about a player’s exploration of self, then we ought to be finding ways to aid in that design goal.”


    I don’t play myself in virtual worlds. I play characters. I talk about them in the third person. My psyche is divorced from theirs. They frequently behave in ways in which I would not. Even when I am not actively roleplaying, I am, nonetheless, playing a role.

    At the same time, I think that even when people play characters radically different from themselves, they still learn things about themselves, in the process.

    I also think that while class selection is an expression of personality, we need to be careful about what, exactly, it’s expressing. Surveys suggest that many people play rogues in WoW because they are highly effective achiever tools. They do a lot of damage, if you play them right. That’s not me.

    I play a rogue because my greatest passion in virtual worlds is to sneak around in places that are too dangerous for me. I like exploring for the best mineral veins. I like hiding over the vein, timing the mobs, and grabbing the mineral at the exact perfect moment. I like sneaking up behind murlocs, whacking them upside their head, and stealing their special pearls from right behind their backs. I like dropping flashpowder, and running for it, when I get in over my head. In short, none of the things I love about playing a rogue have anything to do with killing.

  9. I think certainly to a degree that MMO characters have the potential to aid in self-exploration, but they do not do so purposefully as they are now. With systems like Allignment in DnD games, or the Light Side/Dark Side dynamic of KOTOR, I don’t feel it’s a representation of who we are necessarily, but perhaps an exploration of possibilities, and a shift in circumstance. With so much invested in our real world identities, most people wouldn’t dare gamble that away by “seeing what happens if…” but they allow us to take that leap, be it playing an evil character in an RPG, being ruthless and cunning in FPSs or action games, or taking completely opposite political viewpoints from our own in strategy and politics games.

    With the possibility of MMO characters being (more) accurate representations of ourselves as well, you fall into a number of pitfalls. First of all is the idea of peer-pressure or conforming to perceived norms and standards. People become (largely) what they feel or think they should become, or need to become to get by. We see this all the time in political and cultural identity: during election campaigns, people are less likely to show their true feelings in polls if they feel they will be judged poorly for their beliefs or decision. The same applies to cultural and/or racial relations, most Americans would say they are not racists, but how do we know for certain whether racism has actually gone down in the US, or if people just cover it up publicly because they have learned to associate those views with “bad” or “wrong”, or at least “innapropriate”. Who is to say that players in a (albeit anonymous) game wouldn’t follow these same trends, of being worried of being treated differently by the perceived norm, if and when they do something that may be “innapropriate”, be it stealing in game, or killing, or what have you.

    The other pitfall I can see encountered is the idea of playing “the best character”. Every MMO so far has fallen into role-specific race/class combos and/or superior skill sets outnumbering “weaker skills”. I can remember when I played the original Asheron’s Call, I loved my Spear-wielding warrior to death, I had a back-story, I role-played, I loved it. No one else loved it though, I was seen as inferior because I wasn’t a l33t UA. People laughed at me and my flaming spear because I chose a character I enjoyed over what was the best for leveling.

    I understand that in the perfect MMO, there would be no superior class, but what about the problem of people being pidgeon-holed? In the original EQ everyone had a class, and every class had a role to play, and if you didn’t do that and get phat exp you got kicked from whatever group you were in. I understand that the classes in EQ were built to work together and the roles were there for a reason, but would a more accurate representation of the human behind the polygons work in this way? Would there possibly be room for characters to meaningfully work together if someone plays say, a cleric who doesn’t want to heal but instead engages in hand-to-hand combat out of a strict moral code against weaponry? What if the game is built as EQ was with assigned roles based on class, and there are classes that no one gets plays as because classes are assigned based on personality of the player, and few people fit the profile for those classes? If these classes were important then people would simply rig the system and start playing as whatever classes were needed. The only way I can see this class system working is maybe with a number of classes, one of which is assigned to the player based on personality, and it’s up to the player to specialize from there, allowing a loose role to be assigned by the system, and specifics to be decided by the player him or herself.

    I want to say it would work much cleaner in a skill based system, but I’m having trouble visualizing it without additional problems. The main problem I see is flexibility, it’s something that a lot of MMOs have not yet offered the player, most end up with the “choose class, here’s your job” type creation system. While this syetm would open up new doors in terms of flexibility on the part of the player, we must also be weary of falling into the same trap of people simply being assigned roles that they have to stick with for 50, 60, 100 levels (again, if it would be a class-level based system, which I really don’t like anyway).

  10. In single player games, like KOTOR, the emphasis is on story, rather than the full gamut of possibility. (What if I did a Corran Horn and decided not to train my Jedi powers, but instead was a Rogue Squadron hotshot pilot?)

    KOTOR focuses on the story, and thus, like a novel, it guides you through the emotional journey of a particular stereotype, or whatnot. As a result, I agree: it’s an exploration of possibility, which may or may not be an exploration of self (“Oh, my. I just can’t do this. All this goody-goody Light side stuff isn’t worth the power I could have. What does that say about me? I value power more than I value Light-side ethics?”), but isn’t necessarily.

    That’s a game thing.

    Virtual worlds have the potential to be, very literally, “Do whatever you want.” In that sense, you might pick up on Mike Rozak’s “Storylines” ideas, and recognize that what’s really happening is that you’re making up your own story and fighting over the emotional turmoils you create for yourself, as framed in the environment created by the developer.

    But right now, virtual worlds are generally very game-like, as noted by countless others, and thus it prescribes paths to follow: stories, really.

    I’m not sure it’s possible to make self-exploration the experience a player has in a particular virtual world; I think that the Hero’s Journey, as Bartle puts it, signifies not a player’s journey through a world, but all virtual worlds, including those created by myth and oral tradition, those created by science fiction authors and fantasy authors, those created by campaign world developers and roleplaying game rulebook writers. It’s a full-blown journey from the first time you encounter self-identification with an external stereotype and meld personalities with it, until the point where you become the Master of Two Worlds and straddle it with the ability to share your newfound revelations with society at large.

    So perhaps it’s not so bad, what VWs are doing right now: they provide explorations via the stories they tell. And if people are inclined to take that deeper look, then they see how they might be transformed by the choices they make. If they don’t, then they don’t go anywhere on the Journey. Le sigh.

  11. Star Wars Galaxies was the best example of this ‘personal and community evolution’.

    Start as a character with 250 credits. Make your way up to a reputable crafter, an Imperial / Rebel Commander, a base destroyer, a member of a music band that makes contracts to perform in player cities, a mayor of a player city.

    Especially the evolution of a Jedi was amazingly ‘an epic story’ in Star Wars Galaxies. After mastering a certain number of professions and unlocking your Jedi, you were forced to live alone and away from your friends in exile to level your Jedi. While doing this, you were forced to play political games with enemy guilds, bribes, threats, Bounty Hunters..After a certain time of disapperance, returning to the world usually as a mighty warrior. Then, climbing the ladder of Prestige; Force Ranking System; to be the most reputable Jedi despite its hard work and broken mods.

    But the player base attracted to this kind of ‘personal evolution’ is unfortunately old and not the target consumer of the market. But they are not very small in numbers. I know a good number of players who paid ingame credits to good story-writer players to write bios for them.

    In games like wow, you reach capped level, solo intances, get your tier 1 sets, do harder instances, get your epics, slow down, wait for the next instance and next epics, rinse, repeat. There’s no time for character background stories or evolution.

    Actually, this concept can be considered with the ‘Do levels suck?’ question. There is a certain development time and fund. Should these be wasted on levelling quests and environment creating an enjoyable lvl 1-lvl max journey or should levelling phase be kept as short as possible (meaning skill-based system instead of levelling) and these resources be channelled to character evolution quests and end-game quests as much as possible? Assuming managing both is a myth of course.

    As for being evil / good; players never see it like that. For an Imperial player, The Empire in Star Wars is not evil. The Empire is ‘trying to restore order and peace to the Galaxy for the cost of Rebel sacrifices’. And Rebels are terrorists causing chaos. Vice verca for a Rebel player.

  12. More randomly deep thoughts on this: Taking a broader view of “shaping the path of the player through the game in meaningful ways”, I’ll point out the following:

    One of the neat things that computers can do is customize the entertainment to meet the player’s needs. Books and movies can’t do this.

    One obvious way of customizing is to let players explicitely make the customization choices… difficulty level, class, race, where they walk, what weapons they use, graphics quality, etc. (Explicit choices also have other uses above and beyond customization, mainly dealing with immersion.)

    Having the game do a personality assessment on the player is merely an implicit way of customizing.

    I suspect there are many more ways that a game can implicitely customize itself; Mirjam Eladhari’s technique is only one way.

  13. We obviously have some different ideas about narratives in MMO’s, and some similarities. Where I think we start to agree is:

    …you can’t assign traits to players as they enter the game. Rather, you have to detect traits they are expressing.

    Here’s something I wrote about WoW a while ago:

    “There is an adage about writing that goes like this: “show don’t tell”. And it’s something that any decent writer needs to learn. I would say that MUD’s/MMO’s are in the process of learning how to show instead of tell. WoW tells with reckless abandon. It tells me that I’m a hero, it tells me that I just saved the day, etc. And all I really care about is the new quest item I got because eventually I realize that this is all that really matters in the context of the virtual world. I.e. what matters is what actually happens gamewise, not storywise, because being told that I’m a great hero for the 15th time through some concocted NPC plot loses its charm awfully quick.”

    I think it’s necessary to emphasize this showing more. And that means far more than giving badges or labeling someone “rigidly conservative” or “rakishly handsome”. Character choices and their story need to transcend racial selection or dialogue selection and overt “telling” manners for pasting real-world meaning into the virtual world and need to become deeper game mechanics that allow players to express meaningful traits that make for a place in a virtual, open-ended story.

    A good example is Fish Banks, an educational game that tells the story of the tragedy of the commons. In the game you and several other teams own fishing companies and try to make money fishing. The game models a realistic fish ecology and of course, eventually, fish populations start to retreat. Despite the fact that players can completely control how much they fish, the outcome is almost always that fish populations are decimated. No one comes into the game picking the “greedy corporation” class. No one earns an “ecologist” badge. And yet a very powerful story is told (shown?). Not through verbiage but entirely through game mechanics. That you are fishing and that it is fish populations which are dying out is merely a detail, and the lesson of the game applies equally to many other contexts.

  14. […] When a well known designer starts linking to and talking about theories that sound vaguely familiar, it makes me feel good. Warm fuzzies, etc. etc. […]

  15. I think there’s a lot of people thinking in much the same way. Detection is critical, and it’s what I’ve been working on.

    The short version is that I simply create “loaded hooks” – game choices which reflect the preferences of the player. As the player makes a choice, you can determine what characteristics he or she has (or is roleplaying) and use those to determine future content. I use the same technique for the “ludic” parts of the game as well as the “narrative” ones.

    If you’re interested in morew detail on the steps I’ve taken on the matter, you can find a very rough four-part intro here: Parts one and two are basic introductory stuff, mostly. Parts three and four are still a bit introductory, but at least they have specifics. I haven’t gotten any further in tutorials, although many of my posts are on the same subject.

    The game I’m making utilizing these tenets can be found here, although it’s got an unflattering front post up.

Sorry, the comment form is closed at this time.