Costick on Story/Games

[Mike Holmes]

That was what I was afraid of by linking to the presentation. He's been pretty verbal on this subject before. So check around, and don't read into anything if you can help it.

Does anyone know where we can twig Greg? Again, if he could expand on these ideas, it might help. Does anyone know of an article by him that might explain it better? That would help a lot.

Mike

[Christopher Kubasik]

Hi Mike,

You might have missed my last post on the previous page, so I'm alerting you to it now.

Also, did you catch this http://www.costik.com/weblog/2003_09_01_blogchive.html#106427832498370748
link on the page you linked to. It's his full discussion of MLWM.

He seems to lay it all out very carefully. My last post is my response to it.

Christopher

[Mike Holmes]

That was weird. Missing that last post, that is.

Anyhow, I think that what he may be saying, and I might be wrong, is that MLWM does play "like a game" as opposed to a lot of other RPGs which do not while seeking story. That is, I think he might say that Trollbabe, for instance might not be a game, having lost that game quality in seeking to make stories. If his point can be translated as "Trollbabe is doesn't support gamism" then I'd think that even Ron would agree.

I think he might be saying something akin to, "MLWM supports both gamism and story at the same time." Where story isn't even narrativism, but the post-hoc product that Ron talks about simulationism and gamism producing when the term story is bandied about with those games. Hence why it's Story Now, and no longer just Story.

I have to admit that I find MLWM sorta Gamist myself. I'm always trying to beat everyone else to kill the master. Or, failing being able to do that, ensuring a certain sort of end that seems to fit the character as a secondary objective.

Mike

[John Kim]

Does anyone know where we can twig Greg? Again, if he could expand on these ideas, it might help. Does anyone know of an article by him that might explain it better? That would help a lot.

The most central point is his essay of the same name, "Where Stories End and Games Begin".  You can find that at:

http://www.costik.com/gamnstry.html

He strongly argues the point that gaming is an artform, but he argues that it is one that is fundamentally different from storytelling artforms like movies, plays, and novels.  He likens it to comparing storytelling artforms to non-storytelling arts like painting or music.  

I don't agree with him on that per se, but I do agree that role-playing is fundamentally different from any storytelling, author->audience forms -- including movies, plays, and novels.  Role-playing is not just a different medium or genre for storytelling, because it breaks the author->audience structure.  

It is also why Greg can confidently, and narrowly, state that recounted tales from an RPG session are going to be dull. Well. A lot of the time, yes. When making a story wasn't the focus. But "Moose In the City" and other examples suggest that making Story Now leads to fine stories to recount afterword. Because everyone is on their game to make a story. And voila! A story worth retelling.

Hmmm.  Personally, I think that example actually strengthens Greg's point.  I mean, reading the accounts (from the thread http://www.indie-rpgs.com/viewtopic.php?t=12467">Primetime Adventures: Moose in the City), I hear various people saying "Wow this was really deep for me" and stuff -- but I certainly didn't get that out of just the recounting of the fictional events.  i.e. If you took that thread and stripped out all the player's testimony about how they felt, and only told the story in the game, then I don't think the story is very strong.  The players themselves are constantly saying this, i.e. "I know this doesn't really convey what it was like, but...".  

Now, you can still argue that it is because the retelling was a different medium.  Greg's argument is that it's more than that.  That the enjoyment from the game doesn't come from the story created, but rather from the processes and challenges by which the story is created.

[Christopher Kubasik]

Hi John,

You wrote:

"but I do agree that role-playing is fundamentally different from any storytelling, author->audience forms -- including movies, plays, and novels. Role-playing is not just a different medium or genre for storytelling, because it breaks the author->audience structure."

Cool. I agree with that, too.

But my points in my post had nothing to do with that. In fact, I tried to pull the rug out from under this kind of thinking.  (And apparently failed.)

My point is that that when people compare books and movies to RPGs they ALWAYS think of it at the point of deliver to the audience. Which, as you point out, has nothing to do with the RPG experience.

But I would say that the idea of the creation of a story, with the players being in the same situation of a writer organizing his material and making choices is a fine anology. I'm trying to get people away from the concern about a seperate audience. When a writer writes, he is the first audience and responds, and makes judgements about what's going happen later in the tale, as he sees fit because of how he's reacted so far.

Yes, it's not a 1:1 translation. But neither is it alien to the process of actually creating something.

What do you think? Can we get past this idea that books and screenplays arrive Athena like from writer's head and that there's an actual process to story-making? And that RPGs can provide, in their own way, such a process?

Finally, there is "the story, and the telling of the story.  I would say that the Moose in the City play produced a good story that hasn't yet been told well to anyone who wasn't there.

Translating across media is a tricky thing, but it's common. (Again, as I mentioned in my first post.) Some stories are simply not going to make good stories with almost anything you do with them. Moose in the City is a good story that has yet to be translated out of the RPG creation experience. (Whether in needs to be is another issue.)

People at film studios often by "pitches" -- verbal tellings of a tale that haven't been turned into a screenplay yet. They sometimes pay a high sum for that "story" -- a good story -- that in execution as a screenplay, turns out rather flat. We must remember that few of the actual play posts have anything to do with executing a story well. But that doesn't mean that there aren't stories "pitched" here that couldn't be crafted (playfully, with enjoyment of skill) into solid presentions as movies, novels, epic poems or comic books.

There would be several drafts, most like, of such efforts, with people trying things out on the fly. Just like a Narrativist RPG.

Christopher

[Christopher Kubasik]

Hi Mike,

Oh, I agree. I think he sees the parameters of MLWM comfortable enough for him to call it a game.

I guess I'm at a point of saying, "Okay, fuck it. Sorcerer isn't a game. Fine. Can we play now?"

Christopher

[Gordon C. Landis]

I think Mr. Costikyan (in this slide show, and in general - he's an old-time fave game designer of mine from back in the SPI days, so I've read his stuff over the years) is all over the same ground as GNS, and either disagrees with some underlying principles or hasn't yet grasped just exactly how GNS works.

I'll look at his last three slides - A "Storytelling Medium?", Story and Play are Fundamental, and Play comes First, and talk about 'em in GNS-by-my-understanding terms.  I would interpret "Play" in his useage as being equivalent to Exploration.  So he run's right into the common GNS-confusion that since all gameplay begins with Exploration, "pure" gameplay is therefore all about that (Sim in GNS, Play in his useage).  But the real question (for GNS and RPGs, that is; it may be useless for Mr. Costikyan's electronic gameplay concerns, or not - I just don't know) is what do we do with that Play?  When we "explore the functioning of the world system," what is the observable point of doing so?  It certainly involves "modes of thought and ways of attacking problems", "a sense of satisfaction at mastery", and can be told as a story.  But which of these is the real point of participating in the activity? That is, in a specific instance of the activity.

That's what GNS looks at, and what this slide show doesn't.  He's broken the activity down to the important elements of Nar, Sim, Game and Exploration, but hasn't organized them the way GNS does.  All are present in all play - a Nar player must "explore the functioning of the world system" via "modes of thought and ways of attacking problems", and can certainly have "a sense of satisfaction at mastery."  But if it's Nar play, the point to it all will be learning about the nature of human beings via the mechanism of Story.  That mechanism is also there (in varying degrees and varying levels of importance) in Game and Sim play, but is not the point.  If the point is mastery, that seems a good equivalent to Gamism to me.  His "modes of thought and ways of attacking problems" and "explore the functioning of the world system" seem like pretty good Sim descriptions to me (better than the "celebrate the source material" that seems to be common nowadays, in fact).

So - any of those could be "ultimately [...] what a player takes away from a game."  Or more properly, a player will always take 'em ALL away from the game, and the question is which one is "ultimately" the point in the case at hand.  Can "good gameplay" enhance the Story?  Yup.  Are there compromises involved?  Yup.  Mr. Costikyan (I suspect he'd say "call me Greg", but  . . .) has got a lot of the important stuff down pat.  All that's missing is the organizational structure - which I'm certain makes sense for RPGs, but maybe isn't exactly right for electronic play.

So if the slide presentation is NOT about directly interpersonal RPG play, the organizational gap with GNS might not even be a problem.  But if it is . . . I'm with Christopher that Mr. Costikyan seems to be missing something about Nar play, and maybe Game/Sim as well.

Gordon

[Walt Freitag]

This form of computer-game-focused game criticism, complete with its intractable terminology problems and long tradition of comparing apple cheeks with the color orange, is one of the two worlds other than tabletop RPGs I came out of (the other being LARPs). With the exception of the "Experimental RPGs" slide, the slideshow (along with the related article cited in later links) is a boilerplate overview of a perpetual computer game design topic. Keep that in mind when evaluating Greg's treatment of MLWM: the context here isn't some new thesis or an overview of the newest thinking. It's rather more like Bill Gates presenting an article on "How PCs Work," updated to include newly developed or currently popular hardware. In other words, MLWM is being fit, rightly or wrongly, into an already well-developed conceptural framework (the gameplay/story-quality "continuum") that hasn't changed much in many years. That's why, despite the acknowledgement that MLWM refutes certain notions, the overall conclusions of the more recent slide show haven't changed from those of the 2000 article.

I find myself completely unequal to the task of sorting out the multifaceted muddle that is "the interactive storytelling problem" as understood in computer game design circles. But I'll try to make a start here. The central question is something like this:

Is it possible to construct, out of something other than human brains, a system capable of meaningfully incorporating decisions made by a human participant into an evolving sequence of events such that the outcome is guaranteed to be (or at least is very likely to be) a quality story?

Since at least some humans have the capability described, the task is assumed to be possible given an AI that mimics human mental cabilitites. Also, simple branching text fills the requirements except that either the decisions the player/audience can make are severely limited in number and/or in scope, or the number of segments required is impractically (that is, astronomically) large. So attention focuses on how to shrink the problem from either of those two brute-force extremes, such as by creating libraries of "paragraphs" or other story pieces along with rules for concatenating them (Erasamatron as Crawford describes it, Dramaton [King of Chicago], Tales of the Arabian Nights, and my own in-progress nestable substories approach); by incorporating setting-specific rules and knowledge just sufficient for the task using existing AI technology (Facade, Oz/Zoesis, and the "soap opera plot generator" that every second-year computer science student thinks he can write, but can't); or by somehow grafting "narrative sensibility" or the ability to "guide" a plot onto a cause-and-effect simulator (Erasmatron as it actually is, and many Media Lab installation projects).

The latter approach is analogous to typical Illusionism, and also to no-myth play, in tabletop role playing.

A lot of discussion of the interactive storytelling problem bogs down in two areas: attempts to prove the stated task impossible on general principles (usually, some taken-for-granted ineffable or perfect quality of stories that would necessarily be instantly destroyed by allowing any input whatsoever from any source besides the author) along with attempts to refute such proofs; and voicing doubts about whether such systems would actually be desirable to users (often starting by assuming that "improving" stories with interactivity must be the goal of the exercise, and since there's little evidence that stories need improving, there's no need for interactive stories; this overlooks that the desire for interactive stories in computer games actually arises, most often, from a desire to improve the framing of game play).

Putting human brains into the system changes the whole nature of the "problem." From the "interactive storytelling problem" point of view, MLWM is reasonably comparable to a romance writer's style guide, or to using Dramatica. That is, they all provide a structure that gives the human author(s) specific constraints or goals on what should happen next, while requiring the human(s) to do the hard work of instantiating events that fit the described constraints and/or achieve the goal(s). Even if we disregard, or somehow automate, the creative aesthetics involved, instantiating still requires the author to apply common sense rules of how things work. This is second nature for tabletop role players (though not without the potential for disagreement), but beyond the capabilities of computer programs because general rules ("pets can run away") keep breaking down. (Pet fish...)

It may be hard to fathom,  but a system like MLWM ends up looking like a kind of cheat from the interactive-storytelling-problem point of view. Or perhaps like a solution to some entirely different problem. It's as if you set out to invent an automatic universal meal-cooking machine, and ended up "inventing" a chef in a kitchen.

But, taking that thought too far leads to "system doesn't matter." More specifically, the fallacy is, "I know how role players do it, but I'm trying to do it with a solo system or computer program." Chances are the person saying that doesn't know how role players do it. "Humans do the authoring" isn't an instant automatic solution to the problem of how to make interactive creation of quality stories in role playing (that is, Narrativism) work. It's the beginning of a solution, a fact that makes the goal achievable, but the goal is far from always achieved. Because system does matter.

That means functional Narrativist principles and example systems might be useful to those working on the "interactive storytelling problem" as applied to computer games. Two areas in particular come to mind: the benefits of giving the player more credibility (particularly with regard to "statements" that a player can make through play actions alone, which a game program is more likely to be able to parse), and paying more attention to the kinds of decisions the player is called upon to make.

Yes, the Lumpley Principle does get in the way here, because it appears to vest all credibility in a solo computer RPG with the player, leaving nothing for the program to do. This could either be an indication that despite superficial similarities, tabletop role playing is not usefully related to any solo computer games; or it could be a shortcoming in the LP as currently stated. I believe the latter, but that's another topic.

- Walt

[Mike Holmes]

Thanks Walt,

Given that Costikyan is a tabletop RPG author (amongst other things), why do you claim that this comes from the narrow CRPG perspective? I mean, the presentation does deal with the subject, but not exclusively. Why are you so sure that his argument comes from this perspective? Just your experience with the dialog?

Gordon, Chris, again, what I'm think that he's saying is possibly, effectively, that narrativism is throwing the baby out with the bathwater. That is, in getting to the better story you lose the game. The presentation is all about getting beyond this dichotomy. That is, either he's saying that one can bridge the nar-game gap and play a fully hybrid RPG, or he's saying that one can play gamism, and still get good story as a result - if not story now.

It seems to me that he's either promoting gamism, or gam/nar hybrid. Saying that narrativism is abandoning the game end without having to do so (just as games without story are unneccessary). Does that make sense with your reading of him? Do you agree or disagree?

Oh, and I agree with John, that Story Now does not require good output stories to be created. The process is so personal that any story that happens to have broad appeal will be accidental. Could happen, but it just isn't a goal of the form. The story only has to appeal to the people playing, and only as they create it. Which can be a far cry from what they'd accept from an external author. That is, I think that the same story for the same person might be "good" for them if they create it, and not good enough if they do not. It's like cooking the fish you caught - they always taste the best. But even more personal, because some of you is caught up in the creation.

Mike

[Gordon C. Landis]

Mike,

The sense that I get is that he's not exactly in-synch with GNS, and so it's actually really hard to put what he IS saying into those terms.  By not understanding/accepting the divisions and prioritization inherent in GNS, he does seem to be saying "If you go all story, you lose the game - best you can do is compromise."  When in fact GNS would say "Both can - in fact  [I'd say, anyway], must be present.  The question is, where's your priority?  And then, how do you handle the stuff that might be important (Game, say) but is NOT your priority?"

Gordon

[Mike Holmes]

I agree, Gordon, but that's my point. I agree that he probably doesn't know about GNS or give a hoot about it.

Are you saying that makes him wrong automatically? That is, can't his idea be correct without him knowing about GNS? In fact, can't his idea be correct even if it cannot co-exist with GNS? That is, are you saying that he's wrong because GNS is right?

Mike

[Gordon C. Landis]

Mike,

Hmm.  I'm saying that, for his world of Computer games, I've got no clue.  If GNS applies there at all, I suspect it doesn't apply in the way he SEEMS to be using it, but his conclusions may be useful there anyway.

But if we're trying to apply what he's got  there back to RPGs via GNS/Big Model - I don't think it works.  We have to re-write what he's got so much that it probably ain't his stuff anymore.  Parts of it are wrong, parts of it are right , and I like the way he phrases some of it - but it doesn't look like it holds together in THIS context.  Maybe it does in the computer stuff.

Gordon

[John Kim]

I agree, Gordon, but that's my point. I agree that he probably doesn't know about GNS or give a hoot about it.

Are you saying that makes him wrong automatically? That is, can't his idea be correct without him knowing about GNS? In fact, can't his idea be correct even if it cannot co-exist with GNS? That is, are you saying that he's wrong because GNS is right?

Greg is aware of GNS as well as numerous other models.  Here's blog post of his on the subject:
http://www.costik.com/weblog/public_html/weblog/2004_10_01_blogchive.html#109716795298731530

His conclusion was:

The point is that with a large population, you can slice and dice the data just about any way you want, particularly if your objective is to invent a taxonomy that conforms with your own pre-existing notions (or simply to invent a novel and different one). And while it's undoubtedly true that tastes differ, and different games (and game styles) appeal to different tastes, beyond a certain point it's hard to slot people permanently into particular categories, particularly since a player's motivation may differ a great deal, depending on mood.

So while the effort has some merit, it's also important not to read too much into any set of taxonomies; a different set may well be more useful or important, depending on what you're trying to achieve. Beyond a certain point, your reaction to a new taxonomy has to be more like "Oh, that's interesting" than like "ohmigod! now I understand!."

Naturally, this drew some comments from people like Kirt Dankmyer, Pete Darby, and Zak Arntson.  I personally mostly agree with him, that different taxonomies are useful but you shouldn't make too much of them.  I think the Threefold Model was good for breaking out of the old dichotomy thinking (i.e. story vs game, story vs world, etc.) -- but I think the range of taxonomies is good, and the primary point is recognizing that tastes differ.

[Mike Holmes]

Gordon, again, like my question to Walt, why is his argument local to CRPGs? He addresses the other forms from what I can tell, and he's designed tabletop RPGs like Paranoia. So I'm not seeing this computer bias that's making his arguments irellevant. I think he is talking about "interpersonal play" amongst all of the other possible forms. Games here including just about anything that fits the title.

As John points out, Greg's concepts here are likely not meant to be thought of in terms of our taxonomy. But that doesn't mean that they aren't very relevant to RPGs. The only question is how, and whether or not he's correct.

In any case, when he says "game" and you say that's just exploration, I think your mapping is all wrong. If there is a mapping, I'd say that when he says "game" the best approximation that we could put on it in our dialectic is "gamism." Or rather those same elements in a game that promote gamism and make the game feel like, well, a game.

Mike

[John Kim]

Oh, and I agree with John, that Story Now does not require good output stories to be created. The process is so personal that any story that happens to have broad appeal will be accidental. Could happen, but it just isn't a goal of the form. The story only has to appeal to the people playing, and only as they create it. Which can be a far cry from what they'd accept from an external author. That is, I think that the same story for the same person might be "good" for them if they create it, and not good enough if they do not. It's like cooking the fish you caught - they always taste the best. But even more personal, because some of you is caught up in the creation.

Exactly.  In Greg's terminology (which I think matches many people's usage), it's not the story which they're enjoying.  Now, we use that in the term "Story Now" -- but it is arguably a deceptive label.  There is another quality which is the core: perhaps the creative mode of thought, or perhaps the sense of satisfaction at having created the story (which is different than enjoying the story itself).  I think that's what Greg is trying to get at.