Plot versus Interactivity Solved!

(Inspired by Robin’s post.)

There is currently talk about shutting down the dormant Idrama mailing list. From recent postings, no small amount of frustration is shared between the participants on the list.

Chris Crawford, as ever, remains certain of victory sometime in the future.

I remain absolutely certain that interactive storytelling can and will be achieved. Many of the arguments I witness on the topic no longer excite my attention, as I have long answered most of those questions to my own satisfaction. First among these is the “plot versus interactivity” debate. I solved that problem 15 years ago, published the solution, and nobody seems to have noticed it. Fine. They’ll figure it out someday. There remain serious problems to be solved, but I no longer consider any of them to be killer problems. They are what physicists like to call “engineering details”.

Taking one step back, I think the basic issue with “interactive drama” or “interactive storytelling” is that as headers they need to be qualified. Here are a few options:
1) Is it “narrative” – the presentation of a sequence of events?
2) Is it “story” – a fixed sequence of events?
3) Is it a question of content – human interaction and such?
4) Is it the symbolic coherence and economy of narratives (if a gun hangs on the wall in act one, it must be fired by the final act)?
5) Is it creating believable AI characters?
And the academic version:
6) Is it redefining our terms so that the problem goes away?

It’s certainly hard to solve a problem until you know what the problem is … And since the overall heading of “interactive drama” does not refer to any specific problem, there isn’t going to be any specific solution.
(Btw, I think Facade is aiming at 3, 4, and 5.)

In the quote above, Crawford claims to have solved the “plot versus interactivity” problem 15 years ago. I think he is referring to this piece:

Is not West Side Story a rehash of Romeo and Juliet? Sure, the sequence of events is different, but isn’t it the same story? Think of how many stories have been told and retold a thousand different times, each time with different details of wording, while preserving the basic story.

What’s important are the interpersonal dynamics going on, not the actual events. The events serve to reveal those interpersonal dynamics, but they are windows on the story, not the substance of the story itself.

Our program must start with the interpersonal dynamic and generate events in response to the player’s actions. The precise sequence of events will be variable, but the underlying story will always be the same.

Crawford seems to be making two very different core claims:
A) The “heart” of any story is not tied to a specific setting but can be transferred between environments. While this is at least partially true, I am not sure it actually reflects on the overall problem, unless we want an “interactive story” where the user’s input consists of moving a core story between different settings (could be fun). It’s not unlike some of the improv theater I have seen, but is this what we were asking for?
B) A very different claim, that the heart (“meat”) of a story is not the actual events, but the interpersonal dynamics. From a user perspective, I am not sure how to make a clear distinction between events and interpersonal dynamics. This interestingly rules out centering a story around chance encounters and fate, and I do think there is a close relation between storytelling and a feeling of fate and inevitability (try implementing a Paul Auster story without it). If we are talking about implementing an interesting open world (3-4-5), sure, but is it still Romeo and Juliet if they never happen to meet (actual events)?

Not very convinced. Also not very convinced because as much respect I have for Crawford’s work and writings, it is not clear what problem he claims to have solved.

37 thoughts on “Plot versus Interactivity Solved!”

  1. 6) Is it redefining our terms so that the problem goes away?

    I’m not sure this isn’t such a bad idea. I find that story has changed so dramatically in its presentation over the last fifty years that it might be time for a bit of a re-tooling. Our audiences have changed, not to mention the coping skills they need stories to teach them, that the traditional definitions of story may be broken.

    If we make the decision (as individuals) to do that, however, I think we should do so carefully.

  2. Perhaps it isn’t specifically the interpersonal dynamics that stays the same. Maybe it is the interaction of specific element with respect to other elements that are of a similar value? As in, “Romeo only cares about Juliet’s death because he cared about her in life.” Thus, we must create a situation that includes something or someone that the player cares about, so we can take it away?
    Okay, so that example is a bit vague, but what I’m basically suggesting is putting the player in the position of a character in the plot, and then giving him motivation to fulfill the events of the plot. Thus, he needn’t follow a script, events and story developments can happen anywhere, because character relationships are qualitative rather than scripted? This is basically delving into using A.I. and individualized object motivations. Like, how Half Life 2 has physics puzzles, only the elements of the puzzle are characters, and the rule system is what causes everything to fall into place. Make sense? No? Okay.

  3. I was just flipping through my recently purchased, not yet read copy of “Chris Crawford on Interactive Storytelling”, and in the section on plot vs. interactivity he mentions something about rules-driven rather than events-driven plots, or “metaplots”.

    My own uninformed opinion on the matter is that we definitely need to be treating interpersonal dynamics as a seperate research interest, as not only is it independent of the plot/story structuring interest, but I imagine that plot/story structuring must ultimately rely on advanced interpersonal dynamics modelling to work very well in the first place.

    In that spirit, I definitely prefer the term “interactive drama” over “interactive storytelling”.

    But again, I have no idea what I’m talking about.

  4. Also, on the plot/story structuring side, it would be interesting if someone were to look into the prospects of psychological profiling the player, both before and during gameplay.

  5. I notice Jesper said :”Taking one step back, I think the basic issue with ?interactive drama? or ?interactive storytelling? is that as headers they need to be qualified. Here are a few options:
    1) Is it ?narrative? – the presentation of a sequence of events?”

    What happened to “drama”? My problem with interactive drama is that we cannot be both participant and observer and being a participant in a drama destroys the etic nature of drama: drama shows us character development through struggling against fate and circumstance. I do not see how we can enjoy (aesthetically) that struggle and character development when the character is us.

  6. I do not see how we can enjoy (aesthetically) that struggle and character development when the character is us.

    I would argue that when playing a pre-defined character, such as in Beyond Good & Evil, you can. Even though you directly control Jade throughout the game, she undergoes her own emotional development through cut scenes. Because of that, I’d argue that BG&E qualifies as an interactive drama, and an effective one at that.

    I got to thinking about my initial comment last night and I’d like to retract a bit. I think it’s important to define which elements of the game relate to the story and which relate to the medium. I think if we can successfully do that, then much of the story translation issues are put to rest.

  7. I guess when I think of drama, I’m just thinking about an interpersonal, emotional playing out, rather than something strictly performed on one side and strictly observed on another dealing with the same. It’s just emotional turbulence. When someone says “that was dramatic”, they often just mean that it was emotionally turbulent, rather than “that was characteristic of the art form known as ‘drama'”.

    Corvus, while I love BG&E and certainly think it one of the more emotionally stirring games out there, I think it may be needlessly stretching the term “interactive drama” to call BG&E one. I think that we want to reserve the term for games in which the “drama” itself is, well, interactive. The interpersonal, emotional playing out must be affectable by you, and *include* you.

  8. But then isn’t it just better to talk about making emotionally turbulent or “stirring” games than trying to make “interactive drama” or “interactive storytelling”?
    It also has the advantage that you are describing the aeshetic goals of your design (in MDA speak) from the outset – what kind of experience is the user supposed to have, rather than trying to square the circle.

    A linear single-player game like BGE is not really an “interactive story” since the story is predefined, but that doesn’t prevent it from being emotionally interesting.

  9. Good points, Walter and Jesper (that probably goes without saying, but I’m just getting warmed up this morning).

    It seems to me that almost everyone who is involved in this discussion (and not just here) is arguing from different strata of the problem. For example, when I first read Walter’s reference to “interactive drama” I read, “a drama that you have interaction with.” What Walter seems to have meant (jump in if I’m wrong, please!) is a medium in which you directly influence the direction of the drama and provide further dramatic elements based on the types of decisions the player has made (I’m extrapolating from the profiling mention in the next post). Walter was referring to a deeper level of interaction with the drama than I originally concluded (due, in no small part I imagine, to my recent arrival to this long term discussion).

    I wonder if it would be beneficial to determine which elements of a game’s structure represent story, which drama, and which are merely elements of the medium? For example, Jesper refers to the problem of translating the potentially endless existents of a game to the screen (in Games Telling Stories?), but I would consider existents of that nature to be an element of the game medium, not of the narrative. Perhaps discreet events and characters are not necessarily elements of the story. If the story consists of the moments along the journey where the primary character(s) hit certain developmental points, or gains new insights, then the details by which he arrives there are the trappings of the medium, whether it be novelization, movies, traditional storytelling, interactive storytelling (my strength), or games.

  10. A linear single-player game like BGE is not really an “interactive story” since the story is predefined, but that doesn’t prevent it from being emotionally interesting.— Jesper

    To help me understand how we’re using terms here, could you provide some example of a story that isn’t, in most respects, predefined? Are we talking the difference between reading/playing a story vs. generating one, where Half-Life might be a slightly better example of the former, and the Sims a slightly better example of the latter?

    The only true “interactive story” I can think of that doesn’t involve (or require) predefined events is one that you *write*, rather than one that you read/play. I agree with you that linear single-player games (BG&E, Prince of Persia: SoT, etc.) can be emotionally interesting (and I’d be inclined to argue that they probably are pretty strong examples of “interactive story”), but I’d like to hear less about “story” or “drama” and more about “interactivity” – what makes something like Prince of Persia:Sands of Time less “interactive” than this idealized iDrama entity, or more interactive than reading Flannery O’Connor, and in what ways? I suspect that’s the nut to crack, because we’re making assumptions about the levels of “interactivity” available in drama and stories as though those acted or written genres are the benchmark, rather than a digital game, or this hoped-for computer-mediated drama.

    As to the Romeo and Juliet analogy – it’s an interesting problem, because if we tossed players (for lack of a better term) into the mess without much direction, the play would probably be called Romeo and Rosaline (and I imagine would suffer for it). As Willa Cather wrote: “There are only two or three human stories, and they go on repeating themselves as fiercely as if they had never happened before.” If we took all two or three of them (or however many), plopped them in a database, and let the story emerge from there, I guess it might work, but I rather suspect it might be more akin to monkey’s tapping out Shakespeare – a whole lot of disappointment for one or two potentially great moments.

    So, I’m probably in the “let’s talk about ‘stirring games'” camp…

  11. Then I think we should stop talking about ‘drama’ or even ‘dramatic events’. It sounds like the issue is perceived interactivity, and actual interactivity. How to afford it via predefined narrative in interesting and engaging ways, and whether the issue how to engender a feeling that we are inside a story (narrative immersion?), or that we are being “authored to”, as it were.
    I apologise that I used the term ‘drama’ as a form, I fully understand it has 3-4 meanings at least, but it seems to be used in the formalist sense in http://www.gamestudies.org/0101/bringsjord/-although he says “dramatically compelling” this seems to be based on a notion of external observerable characters.

  12. I read the Chris Crawford linked essay,
    http://www.erasmatazz.com/library/Lilan/plot.html
    and I wonder, if the story is the essence the beating heart behind a plot, can we have that generic story without embedding and enriching it via plot?

    Crawford seems to think we can.
    ” I can fiddle around with the plot and still preserve the story.”
    ..”The trick, of course, is to understand the concepts behind the story well enough to permit variations that remain true to the story. ”

    But the above quote identifies linked features that suggest they share the same basic “story” core but it does not create a story independently of events, it does not tell us what the story actually is. He has deduced there is a meta-pattern from shared features, without saying (or being able to say) what that meta-pattern is?
    Secondly, he wrote:
    “What’s important are the interpersonal dynamics going on, not the actual events”
    I am not sure of the relation of these dynamics to story. So you have to have “interpersonal dynamics going on” to have story? Or are they the result of a generic story (meta-pattern) bedded down into a specific context?

  13. I should probably stop talking as I’m not at all an expert on the interactive drama/story subject, and I have no idea how representative (or misguided) my thoughts are on it. Regardless…

    Jesper, isn’t any game trivially capable of being emotionally turbulent/stirring? “Interactive drama” as I like to think of it makes the emotional turbulence accessible on a mechanical level. Say I’m (virtually) coming home to the family after a bad day at work and I start yelling at the kids and wife. The kids start crying and the wife starts yelling back at me. (Assume that the audio/visual representation is good enough so I can take these events seriously.) That’s the sort of experience you’d want out of interactive drama.

    My own concept of interactive drama (which is probably not representative of what other people think) doesn’t require any active structuring on the game’s part. There’s no need for a Three Act Structure Manager or a Well Formed Story Manager or anything of that sort. Just take one or two NPCs with reasonably advanced AI and dump them in a room with the player.

    That’s why I prefer the term “interactive drama”, where “drama” is a genre rather than a form, as it is for movies. I feel this carries less conceptual baggage than “interactive story”, which implies to me the presence of a Well Formed Story Manager, which is where most of the problems *seem* to be coming from. I think that if you have reasonably advanced NPC AI and player input capabilities, dramatic situations will invariably occur, and these will be interesting enough on their own without having to resort to a Well Formed Story Manager or something of that sort. Which isn’t to say that a Well Formed Story Manager isn’t a valuable thing to shoot for.

    Anyway, that’s my take, but please don’t take it as representative of anything said so far in the interactive drama/story debate. :)

  14. The kids start crying and the wife starts yelling back at me … That?s the sort of experience you?d want out of interactive drama. — Walter

    Maybe that’s the sort of experience you’d want out of interactive drama … I’ll take mine on a beach with a cold beer (and no screaming), thanks ;-)

    Maybe that’s inactive drama…

  15. Walter: isn?t any game trivially capable of being emotionally turbulent/stirring? ?Interactive drama? as I like to think of it makes the emotional turbulence accessible on a mechanical level.
    I think you point to a basic problem of wanting to do “interactive drama/storytelling” – players have lots of emotional experiences with games – some people even get so agitated over MMOs that they kill each other! But “interactive storytelling” suggests that the emotion should be in the game, and not be something that the players make up between themselves?

  16. Jesper: But “interactive storytelling” suggests that the emotion should be in the game, and not be something that the players make up between themselves?

    Games cannot contain emotions any more than books, movies, or plays. Authors of the media can draw upon their own human experience to portray events intended to evoke certian emotions in the audience, but they have very little control over the actual result. Poorly written or performed dramas may be interpreted as comedies. Darkly comedic works may be interpreted as thrillers by certain audiences.

    MMOGs to date seem to have made no attempt to evoke specific emotions and focus instead on the time worn “kill them and take their stuff” motif of the early days of D&D. Of course the resulting emotional drama of those games leads to real world conflict.

    That, in and of itself, should be proof that games are already “interactive dramas”. That is, the fact that emotions are stirred by the games indicates that their audience is looking to them to fill the same role stories do (or once did) in their lives. This perception causes the game to become a story, whether the technology or intent supports it.

    Perhaps instead of focusing on the lack of drama, or story, or narrative in games, we should focus our efforts on how to use their inherent and existing capabilities to create constructive emotional responses, which can only lead to positive effects on the world, the industry, and the technology.

  17. No more lurking for me!

    I believe I agree more with Walter’s characterization of *interactive* drama. Although I would agree that most games out today are dramatically flavorless. Drawing on the metaphor of Half Life 2’s physics system again, when me and my colleagues were looking at Hammer as a platform, we were quite surprised to discover that almost none of the major action events in the game were scripted. Rather, they were nearly all a result of the attributes of the objects (hammer has a detailed system for making each and every item in the game have its own properties, weight, breakability, etc.) and how the objects had been arranged with regard to each other. Essentially, what I was trying to earlier, I forsee a similar structure for interactive dramatic arcs. That plots are driven by actual characters with motivations, fears, needs, etc. interacting with each other. Thus, I see how this whole structure could be arranged in such a way as a coherent plot will emerge, regardless of what the player does. But the player will be able insinuate himself into the plot, or if given an existant persona, will be sought out by other characters. This sort of model is already duplicated by the quest-explanation system, so I can’t see why we won’t make those silly “go get the red sword *because*” explanations a tangible reality. It would essentially be creating an emergent system with a controlled outcome.

    Although such a system would require time and effort to produce, and would require a great deal of knowledge about interpersonal interactions and social psychology, and maybe even another generation of processing power, I do think it could work, and work well.

  18. That’s exactly the sort of thing I’m working on, Blue. I’m also trying to scale it so that it’ll run on today’s hardware. Of course, by the time I’m releasing a game, we’ll be running the day after tomorrow’s hardware, so that’ll be all right then.

  19. Yikes! Looks like most of us are doing that, since that’s our approach also. We’re either all on to something, or all gorfing from the same petrol can.

  20. Jesper: I think you point to a basic problem of wanting to do ?interactive drama/storytelling? – players have lots of emotional experiences with games – some people even get so agitated over MMOs that they kill each other! But ?interactive storytelling? suggests that the emotion should be in the game, and not be something that the players make up between themselves?

    Yeah, in an interactive drama/story, emotions should be procedurally simulated for NPCs (along with motivations and intentions and so on), and the player should be able to affect those emotions. It can’t just be about the player having any emotional reaction to the game, since that always happens anyway.

  21. Blue, Corvus, and Aubrey: But that’s the deal with the whole thing. Once you decide not to do a branching narrative but a detailed simulated world with human motivations and intentions, you rather face the problem of having to create complete and robust AIs far beyond anything that has ever been made! And pure increases in processing power will not solve it.
    So we are stuck between a rock and a hard place.

    Walter: The devil’s advocate says: – But if the emotions are supposed to be simulated on the screen and the actual emotions of the player do not count, this begs the question … why? Why make interactive drama/story if they do not add to the player experience?

  22. Jesper: Or, perhaps, we just need to reposition our branches into something less linear? My problem with existing branching narratives is they still follow a linear path and therefore must hit choke points that restrict the players’ progress, forcing them into completely predetermined paths, regardless of player approach.

    I don’t think the “Ai” needed to fix this has to be as granular as it would seem at first glance. Of course, I’m still at the drawing board (allegorically speaking) so I may find that I’m wrong…

  23. Jesper: The player’s emotions definitely do count, and I really do believe that interactive drama/story can and will add to the player experience. It’s just that talking about player emotions isn’t sufficient by itself. It would be like writing stories where only the protagonist has emotions and everybody else was merely a scripted facsimile of a person. Sure, the protagonist could have important emotional experiences. But unless someone wrote an astronomically large number of scripts, his experiences would be limited, just as they are for players in today’s games.

  24. I agree with Walter and Corvus with regard to emotion in games. Contrarily, though I think that once we figure out *how* to create individualized elements that form an emergent system, it will become easier and easier to do. Essentially, by replacing decision trees and those choke points with character motivations placed within a fully fleshed out environment, characters will be increasingly able to solve problems and interact with each other on their own, without complex plot maps to guide them. Plot trees, after all, work best when you can’t tell there is a plot tree, right? So why not go the extra length and eliminate the plot tree altogether!

    Really, the only big humps that I see is deciphering the mire that is human social psychology and solving the problem of limited computational resources. Regarding the problem of computational resources, what about parallel processing, and 64 bit operating systems?

  25. from the comments I’ve read from Crawford on the games studies mailing list, I get the impression that he doesn’t really want games as we know them, but rather plays. That’s fine, a play can be fun, but there are also times where pointless shoot ’em ups are just as desirable. While there certainly CAN be interactive narratives, I don’t feel that there HAS to be. There is so much to study about what games ARE that I find myself much to busy to wonder about what they SHOULD be.

  26. There is so much to study about what games ARE that I find myself much to[sic] busy to wonder about what they SHOULD be.

    That’s fine, but someone has to look to the future or we wouldn’t have any progress at all, would we?

  27. A discussion within the same scope is going on at http://www.intelligent-artifice.com/2005/03/chris_crawford_.html.

    There?s a completely craziness going on with AI. It seems that game/story salvation is there. We just need to develop “emergence” in it. . Well, I normally disagree with the academic corpus anti-narrative view of Juul, but here I agree 100% with him. There?s no room for that :)

    Going for “emergence” is going for a realm even more complex than developing an interactive story. For a light flavour, look at “Emergence” by Steven Johnson (http://www.amazon.com/exec/obidos/tg/detail/-/0684868768/qid=1112749309/sr=2-1/102-5831172-0137744?v=glance&s=books ), for a deeper understanding of it and truly amazing view of its foundations go for “G?del, Escher, Bach” by Douglas R. Hofstadter (http://www.amazon.com/exec/obidos/tg/detail/-/0465026567/qid=1112749268/sr=8-1/ref=pd_csp_1/102-5831172-0137744?v=glance&s=books&n=507846 )

    Believing in achieving “emergence” to arrive at Interactive Stories, is nothing more than believing that you can build a “mind” and inject it in a robot, making him like any human. We?re not talking here about a regular robot, that could perform tasks we as humans find boring, but about a robot guided by moral values and with free arbiter. I believe that people are just dreaming with Sonny from “I, robot”.

    But even thinking that we could arrive there. What would be the interest of that? If a robot or a game character could be so intrinsically human as any human, I don?t see the difference of having writers or actors behind the digital surface controlling the nowadays NPCs, and responding in real time to player choices. (something like we?ve been hearing in the last days about matrix online – http://pc.gamespy.com/pc/the-matrix-online/598441p1.html?fromint=1 ). The only big difference I see here is automation and costs reduction, nothing else. Even here, robots with moral values understanding would not like to be treated as storytellers slaves :-)

    I believe that Interactive Storytelling, is not looking for that. I believe also that it?s not looking for Interactive Story. As we can see they are different labels.

    I agree with Jason here,
    “”The only true “interactive story” I can think of that doesn?t involve (or require) predefined events is one that you *write*, rather than one that you read/play.

    and Juul,

    “A linear single-player game like BGE is not really an “interactive story” since the story is predefined, but that doesn?t prevent it from being emotionally interesting.”

    Sure, but it can be defined as an “interactive storytelling”. Storytelling means, the “act of telling” a story. So the interactivity occurs through the telling choices and not through the story choices outcomes.

    I agree with ErikC,

    “How to afford it via predefined narrative in interesting and engaging ways, and whether the issue how to engender a feeling that we are inside a story (narrative immersion?), or that we are being “authored to”, as it were.”

    If we look at HL2, this is an Interactive Storytelling artefact in my perspective. We play the role of Dr. Freeman and so we participate in the telling. If we, players, don?t help the telling progressing, we?ll never find out the outcome. So the system needs the player to exist. Without the player, there is no storytelling. The way people play the game, the way the story is told in the screen, is different. The storytelling experience created by the player who made the recording of the playing HL2 in 3 hours, is a completely different experience of the same story, because the telling was different. It was made without any think or relaxation pauses, with no time to admire the view, no time to talk with NPCs and understanding they feelings, no time to look at tired faces or at our dead troops, no time to be emotionally surprised by bodies falling near you, no time to enjoy the beauty of Alyx, no time to throw garbage at guards, etc, etc.

    The story is pre-authored, but the telling is not. So albeit the story is writer dependent, the telling is player dependent, and this creates a shared experience. What we need is more interactivity in these shared telling moments, to engage in situations with a more diversified emotion spectre.

  28. Who said we needed realistic AI to create I.S.? Our interface into a game is entirely crippled by a tiny amount of bandwidth, and a limited vocabulary of abilities. It’s easily for us to believe that each AI actor ALSO only has that range of expression within the game world.

    People seem to forget that ultimately, we’re not trying to simulate reality in creating I.S. Percievably conventional stories can be created from far, faaar more cut-back notions of the real world, and frankly, players benefit from systems that are seemingly simple – an overcomplex system creating perceivably random crap dumps on players’ ideas of causality, taking them out of the game.

  29. You are refencing the crap fest that is “I, Robot”? The one with Will Smith?

    You can’t just sidestep the whole conversation by declaring airily that “Interactive Storytelling is not looking for that.”

    And I read Steven Johnson’s book. Give us a little credit, for god’s sake.

    I’m wondering if YOU read it. Emergent systems are defined by how individual elements LIMIT other individual elements, causing them to behave in particular ways. Thus, any game that has even a predefined setting is rigged, and is not a true Interactive Story by the definition you accuse us of having.

    Newsflash – Half Life 2 is a form of emergent design. There is no scripting system in the game. All objects are arranged in such a way as to interact with each other on a individual level, following the laws of physics and such. The player simply adds his interaction to the already complex web of forces that exists within the game, and it is designed in such an elegant way as to allow individual and interactive gaming experiences that are balanced in pace and emotional tenor.

    ALL stories told by humans are based on human experiences, so the natural destination for ANY storytelling medium is more realistic and\or vibrant characters, plots, and settings. Thus, it is only natural that we begin to push characters to be more psychologically similar to human beings (I mean, they are already shaped like us on the outside, aren’t they?). Saying that all games need to be implausible and unrealistic representations of reality (not to trash abstraction, which is a simplification, or streamlining of reality), is like saying that all movies have to have 50 ninja fights, and villains must be incontrovertibly evil. It is an ignorant view, and damaging to the medium.

    And to think that Emergently designed personas, A.I.’s (which, as has been stated repeatedly, are a long way off) with moral codes and such would somehow resent being used as “slaves” is so ridiculous that I’m not even sure it warrants a response. FYI, they only become slaves, dissatisfied with their lot in life, if they have an extension into PHYSICAL REALITY. Otherwise, they will be completely unknowing of the nature of the world that they live in. The whole crutch of the matrix was that people could somehow sense that they were in a world not their own, and could “snap out of it”.

    Ok, /rant off

  30. And with regard to your comment on having writers and actors simply play parts (this is all directed at Nelson, by the way), what do you think live action roleplaying is?

  31. Blue: You are refencing the crap fest that is “I, Robot”?

    Yes. “I, Robot”, (2004), by Alex Proyas, suggested by Isaac Asimov’s book

    Blue: You can?t just sidestep the whole conversation by declaring airily that “Interactive Storytelling is not looking for that.”

    Don?t take out of context my arguments, please. I didn?t declare that. I said ? “I believe?”

    Blue: And I read Steven Johnson?s book. Give us a little credit, for god?s sake.

    I didn?t assume that you or anybody else hasn?t read it. I was only passing information about the theme to the forum and not to anyone in particular. The point was sharing information. I don?t know everything, and so I?m also willing to get new bibliographies, new information, new perspectives, and new visions, are it from industry or from academy.
    “Emergence”, is an interesting book that tries to describe a very complex concept through a cool and well communicative writing. Steve Johnson is a journalist not a scientist (at least in the way he writes his books). So, if you?re really into try a deeper understanding of what he is “trying” to say, you must look for other sources.

    I?m no expert in emergence. And being this a forum where we are discussing game/storytelling theory, I assume that most of the people are also not experts in that area. So it is a bit annoying, seeing deep and very complex genetic, mathematic or physic?s concepts as “Emergence”, “Chaos Theory” or “Heisenberg Uncertainty Principle” being treated with light weight in discussions of new media or videogames.

    Blue: Newsflash – Half Life 2 is a form of emergent design…

    According to Steven Johnson, yes.

    Blue: ALL stories told by humans are based on human experiences, so the natural destination for ANY storytelling medium is more realistic and\or vibrant characters, plots, and settings.

    For realism and then anthropomorphism, just think about Disney or Ken Perlin works.

    Blue: Saying that all games need to be implausible and unrealistic representations of reality? is like saying that all movies have to have 50 ninja fights, and villains must be incontrovertibly evil. It is an ignorant view, and damaging to the medium.

    What you are calling for reality in a movie, is nothing more than film conventions developed through decades, accepted by masses as true expressions of reality and that you learnt to understand since your infancy, seeing TV and movies. I could give you interminable references for this, but once you?re considering me as an ignorant, I believe that you wouldn?t be interested in any of them.

    Blue: And to think that Emergently designed personas ? resent being used as “slaves” is so ridiculous?. they will be completely unknowing of the nature of the world that they live in?

    It was a joke, didn?t you get it? Only taking into account an old Cartesian dualism mind/body, we could think about these myths and fairy-tales. If we assume a more recent scientific perspective of no possible separation between mind and body, there is no possibility for mind transferences over networks, or even for artificial mind creations, because we “humans” only exists as a “mind + body” entities. Your thoughts, your reality view depends not only of your mind interpretation and meaning construction but also of your body responses that helps you in the process of reasoning through emotions.

    Blue: And with regard to your comment on having writers and actors simply play parts (this is all directed at Nelson, by the way), what do you think live action roleplaying is?

    I don?t know, what “live action roleplaying” is. Is it a mix of theatre drama plus paintball socializing? What is the success of it as a storytelling form, has it any type of popular wide acceptance as movies, books or even theatre?

    I?m not chasing a type of interactive storytelling experience achieved through multiplayer or massive multiplayer systems, because these are completely different forms of the IS single experience. The first one uses the computer/game as a mediator between persons, helping in creating complex and social relations, and so can maybe be an easier path to arrive at a certain type of IS. The second one is an object, which must interact alone with the person, it has not the advantage of having another complex being in the other side of the court to help rising the complexity of actions, and so to rise the moral controlled uncertainty of the system. I believe that I?m giving preference to this relation, because I?m looking at the game as an artefact not as a medium. So I?m going for the relation game-man as the relation movie-man or book-man, which is completely different from the relation man-machine-man.

  32. Nelson: Blue: ALL stories told by humans are based on human experiences, so the natural destination for ANY storytelling medium is more realistic and\or vibrant characters, plots, and settings.

    For realism and then anthropomorphism, just think about Disney or Ken Perlin works

    Me: Way to take my quote me out of context. I made a specific comment about abstraction and you ignored it. Anthropomorphism is a form of abstraction, as is impressionism. And, as we all know, impressionism is a development of realism.

    Nelson: What you are calling for reality in a movie, is nothing more than film conventions developed through decades, accepted by masses as true expressions of reality and that you learnt to understand since your infancy, seeing TV and movies. I could give you interminable references for this, but once you?re considering me as an ignorant, I believe that you wouldn?t be interested in any of them.

    Me: Don’t make judgements about my upbringing or disposition and then accuse me of being closed minded. And with regard to realism and film conventions, I have a thorough background in film study and I am fully aware of those conventions. I am not talking about the realism of a situation, I am talking about realism of experience. Like, when say, a cartoon character smiles, it helps if the smile behaves as the audience might expect to behave. Or that arms and legs bend where they are supposed to. Or when characters have actual motives for the things they do (although this can be subverted to dramatic effect).

    Nelson: Blue: Newsflash – Half Life 2 is a form of emergent design?

    According to Steven Johnson, yes.

    Me: As well as every biology and psychology class I’ve ever taken.

    Nelson: t was a joke, didn?t you get it? Only taking into account an old Cartesian dualism mind/body, we could think about these myths and fairy-tales. If we assume a more recent scientific perspective of no possible separation between mind and body, there is no possibility for mind transferences over networks, or even for artificial mind creations, because we ?humans? only exists as a ?mind + body? entities. Your thoughts, your reality view depends not only of your mind interpretation and meaning construction but also of your body responses that helps you in the process of reasoning through emotions.

    Me: I don’t disagree with you here too much, just a few nitpicks. Emergent intelligence simulations are possible if you simulate body functions with a programmed system, a la the Matrix. Also, emotion itself is a body response because all thought is simply a function of chemistry in the brain, a physical realm itself.

    Nelson: I don?t know, what ?live action roleplaying? is. Is it a mix of theatre drama plus paintball socializing? What is the success of it as a storytelling form, has it any type of popular wide acceptance as movies, books or even theatre?

    Me: Live action roleplaying is like playing D&D, except you act out the parts. It has wide acceptance in old soviet bloc countries and Israel. Also, it is fairly widespread, though not as much as in the aformentioned areas, in the USA.

    Nelson: I believe that I?m giving preference to this relation, because I?m looking at the game as an artefact not as a medium. So I?m going for the relation game-man as the relation movie-man or book-man, which is completely different from the relation man-machine-man.

    Me: I don’t disagree with this, (I don’t think Game-man is the same as movie-man, though. And really, I’m not sure why you brought that up in the first place) I and I don’t think all games should be played out by live human characters. But I do think that if you simulate human characters, give them motivations and build a setting specifically tailored to these motivations, more could be accomplished with this style of system (it is more robust and dynamic) than with a traditional plot-tree or quest system. Like how events based on emergently designed physics are better and more robust than events based on scripting.

    There. :P

  33. Yikes! Looks like most of us are doing that, since that?s our approach also. We?re either all on to something, or all gorfing from the same petrol can. – Aubrey

    I believe it’s the former, and I also believe we’re all just doing the obvious. As a rough formulation: “Interactivity is being able to always tell the What and the Why”. If that could be the ground were engineers and storytellers congregate, I would see that as a major step forward in our ability to communicate.

    Once you decide not to do a branching narrative but a detailed simulated world with human motivations and intentions, you rather face the problem of having to create complete and robust AIs far beyond anything that has ever been made! And pure increases in processing power will not solve it. So we are stuck between a rock and a hard place.
    – Jesper

    An alternative is to face the problem of simulating a suitable AI, in a piecemeal fashion – an AI that is not based on a huge body of “common sense information” (whatever that is), but on what Aubrey has, in that other recent discussion revolving aroung the same theme, called the “Edge Metaphor”: “a percievably logical reason why the player can’t leave the designated world space”. To me, half of what makes stories useful in the context of interactivity is that they provide those edge metaphors. The other half is that they provide structure and coherence for what’s inside of these edges.

    Our interface into a game is entirely crippled by a tiny amount of bandwidth, and a limited vocabulary of abilities. It?s easily for us to believe that each AI actor ALSO only has that range of expression within the game world. – Aubrey

    Being human, I regard my interface into the world as being “crippled” like that. No offense meant, but Jesper’s dictum “What goes on in a game is considered ‘unreal’; has another status than the rest of the world” just does not work for me. If my experience of “what goes on” when I’m playing a game, reading a book, watching a movie, really had “another status than the rest of the world”, I don’t think that I would bother doing it. Any medium only ever works for me if I can recognize myself in “what goes on”, i.e. if it’s compatible with the way in which I construct my world. That’s why it’s so difficult for me to appreciate stories and games that come from cultures I’m unfamiliar with – I need to expand my bandwidth and vocabulary before I can comprehend the range of expression available to the characters that I encounter therein, and realize how they map to my own.

    So, no, I don’t believe that there is “a completely craziness going on with AI”, as Nelson suggests there is. To me, it’s just that both engineers and storytellers, having attacked the same problem from opposite sides for some time, now have worked through to the realization that 1. it is possible to build what Blue calls “an emergent system with a controlled outcome”, and 2. to do that successfully involves painstakingly defining and motivating all of the objects and object relationships in that system, in a process that seems to be, in essence, non-automatable: TANSTAAFL. As long as AIs are incapable of understanding the What and the Why of their actions, human storytellers just will have to simulate them doing it. Doing so lots of work (I like to compare it to cell animation, and believe that, regarding available tools and procedures, we’re in the Steamboat Willie phase of this emerging [sic] craft), but some of us are actually doing this now, and yes, I actually think that we’re quite sane, thank you.

  34. I wonder if it would be beneficial to determine which elements of a game?s structure represent story, which drama, and which are merely elements of the medium? – Corvus Elrod

    Yes, this can be helpful. I use Espen Aarseth?s distinction between texton and scripton, but with a twist: since I’m dealing with natural language, I have input scriptons and output scriptons, with a system of textons “in between”, and a traversal function that takes an input scripton and generates an output scripton from it (an interpreter, basically). The code is organised so that input scriptons correlate with “drama” – the PC is the character in (a varying degree of) conflict with the NPC, and her input is the way she expresses that -, the output scriptons correlate with “story” – the sequence of output scriptons that are communicating the results computed from any sequence of input scriptons -, and the textons and the traversal function afforded by them correlate with “medium”.

  35. The contributions to this discussion by Blue, Corvus, and Aubrey, which indicate that more people than I was aware of have hit upon the same ideas for AI modeling that I use, prompted me to do some more searching, to see whether more examples for this could be found. Indeed, they can. Naysayers might, for instance, have a look at “Creating an AI modeling application for designers and developers” (8-page PDF), in particular at the section about “Polymorphic Indexing”, as it describes a general mechanism for implementing context-dependent AI behavior. This is the same principle as I apply, for instance, when figuring out what the input “Why?” might mean to my system with regards to any system state. The example given in the paper describes the implementation of context-dependent “attack” behaviors for games like Half-Life and Civilization, and works in just the same way.

    I feel much less lonely already :-)

  36. Hi Dirk,

    I’ve seen your intriguing contributions on GTxA and a couple other places, and I’m happy to see someone attacking these problems from the writer’s camp, instead of the Comp. Science camp. My own meager efforts are over for the time being, but I managed to get a PhD. thesis out of it, which is up here – http://www.thealmosfunkband.com/backup/SGOPIATE.pdf

    I think my approach is at least sipping from the petrol can (which, by the way, I think Chris Crawford originally trekked all the way to the petrol station to fill)

    Hope you’re even less lonely now ;)

Comments are closed.