Games, Storytelling, and Breaking the String @ EBR

Greg Costikyan (Manifesto Games and senior IGDA figure) has released a paper titled Games, Storytelling, and Breaking the String at Electronic Book Review. The subject is the "ongoing culture clash between those who view story as perhaps important but tangential to understanding the nature of games, and those who view it as essential":

Dungeons & Dragons, originally created by Dave Arneson and refined by Gary Gygax, was an outgrowth of the Chainmail rules for playing fantasy battles with miniature figures. Chainmail already had rules for special "hero" characters on the battlefield, single individuals equally (or more) powerful than a whole military unit. Arneson took those rules, elaborated them, and set the game, not on a battlefield, but in a "dungeon," an underground domain populated by monsters. In one sense, this was a simple extension of an existing game; but in another, it was a wholly novel form of game.

You played a single character with the ability to grow and gain in power over time; and while (initially) Dungeons & Dragons, as a set of rules, did little to encourage plot complexity, true role-playing, or anything like real storytelling, the mere fact of a character persisting in an imaginary world over multiple sessions of play offered a clear opportunity for a tighter connection between gameplay and story. D&D was innovative in another regard too as it dispensed with the need for miniatures, a board, cards, or other physical game assets. It transpired entirely in the imagination - turning the tightly constrained nature of previous games on its head. If you could imagine it, and the gamemaster was willing to go along, it could happen. This opened an exciting vista of more free-form and flexible games. [...]
The clash between those who viewed games as formal systems and those who viewed them as storytelling media persisted with the rise of digital games; if you view the program of any Game Developers Conference (or before it, the Computer Game Developers Conference), you will find panels or presentations debating the role of stories in games.

Click to expand...

More information.

A story is linear. The events of a story occur in the same order, and in the same way, each time you read (or watch or listen to) it. A story is a controlled experience; the author consciously crafts it, choosing precisely these events, in this order, to create a story with maximum impact. If the events occurred in some other fashion, the impact of the story would be diminished - or if that isn't true, the author isn't doing a good job.

Click to expand...

Rubbish. That's an artifact of publishing, the order is fixed because it can't be changed. We know from fariy tale and folklore varients that stories used to change frequently. If a stories purpose is to achieve 'maximum impact' then all published stroies arn't doing a good job because the author has only limited knowledge of their audience - simplitic example; a story about a hostage situation set in a libary will have more impact on libarians while setting it in an icecream store would have more impact on icecream vendors. The nice thing about c-rpgs (and what the liner story obsessives seem to miss) is that you can check if the audiance is choosing to play a libarian or champion of frosty deserts.

Which leads to one of my pet fustrations:

In principle, it would be possible to implement a game of this type that doesn't conform to the "beads-on-a-string" model; in practice, it makes little sense to do so. Content development is expensive, and if a player is only exposed to part of it in the course of a game, you've wasted development money. And the more branches you have, the less of the overall game a player will see.

Click to expand...

I've seen this repeated several times lately and still don't think much of it. Even if you set aside all the players who will replay the game several times specifically to see all the content there's plenty of evidence at least some players would be happy the game had the content even if they never saw it simply because it represents real alternative choice cf. the minipulation thread.

Interesting site and quite an interesting, if demanding read. I don't find myself drawn to many of his models though, especially his examples of hypertext fiction and narrative and free form "games". The first strikes me as an experience, not at all a game, that may or may not provide a challenge or insight, and the second group as some sort of LARP-a-thon "aren't I important and interesting?' scenario...but I may be missing the gist, and the mere thought of a game that consists of participating in a reality-tv inspired encounter chills my blood. I would definitely call that "breaking the string."

@V7
While I agree with you that if we in story include traditional oral storytelling, the statement that a story is by nature linear is rubbish. I disagree with your implication that computer games can work like oral stories. Everything in a computer game is pre-generated just like the story in a book. The cost of generation is also a lot higher.
While your idea that the game could analyze the players character and fill in parts of a story based on that information is an interesting thought. But there exists a large problem that you are glossing over (or you simply don't know of it), namely that a computers ability to generalize is almost nonexistent.
I wholeheartedly agree with your opinion about game content that can't all be seen in one play through.

@margerette
I have never understood why cRPG people hate and look down on LARPS so much, LARPS are just like PnP RPGs a form of interaktive storytelling. They can be almost anything from improvisational theater about current political events, to juvenile dress up and beat each other with a stick. For me the fact that there are more LARPS of the later type than the former doesn't make me despise the concept.

Damn, that was an interesting article, and not at all what I expected from the quotes and comments here. Strangely enough, it parallels what we were only just discussing here in another thread.

Perhaps there is a storytelling revolution brewing.

We can always hope for a revolution, the big question is of course how will/should the post revolutionary story structure look?

That's the million-dollar question, isn't it?

I think we should split it into two parts:

(1) How does the story structure look from the point of view of the player?
(2) How does the story structure look from the point of view of the writer?

I believe that the answer to (1) we already know -- it'll look more or less like the stories people have been telling each other since they were people. We're wired to appreciate certain types of stories; we expect things like continuity, consistency, characters we can understand and identify with, conflicts, resolutions, and so on. Goldilocks and the Three Bears is a story. This isn't a story; it's just a sequence of nonsense phrases that follows the structure of one (OK, a conference paper rather than a story, but you get the picture.)

However, the answer to (2) is the hard part. To manage the complexity of the next-generation story, we would need a whole new way of thinking about it, and a whole new toolset to make it possible. Any ideas?

While I found the article to be a rather interesting read, I also think it that it has some pretty questionable parts in it.

V7 already mentioned that the author's statement that a story is always linear is a bit misleading. I would think however that the "linearity" of a story pretty much depends on how much of its content is predetermined. Take for example P&P RPGs - a gamesmaster usually has the possibility to predetermine an absolute minimum of content. That leaves a lot of freedom to the players I'd say. Following the author's argumentation that freedom and story are two contradicting forces that would mean in consequence that such a form of storytelling would inevitably lead to a weak story... from my own experience however I'd say that the opposite is the case. Admittedly when it comes to P&P RPGs a lot depends on the gamesmaster.

So my main concern with the article is that it is based on the old "if you want to tell a great story you have to go for linear gameplay" argument. But let's face it - there is absolutely no proof that this theory is right.

When it comes to the models that he offers I also found some strange statements. Take for example the "Embedded Stories" models. According to the author non-linearity is given because players can experience small linear quests in any order. Apart from the fact that this is only partly correct (the quests very much depend on your level for example... so even here we have some limitations)... does no one else think that this makes absolutely no sense? Or better that this model pretty much avoids on touching on the "real" problem?

On the other hand, I found other models quite tempting... it's rather intersting that he mentioned game books, I think. There is one game book in particular which always inspired me when it comes to story building - "The Seven Serpents" by Steve Jackson. It's the 3rd book of a 4 book series and it's the pretty usual fantasy stuff. The background is that you are the hero and on the way to the hideout of the villain to get back an artifact that he has stolen. The seven serpents are agents of the villain and on their way to inform their master that you're coming for him. What I always found great was that you do not have to kill all the snakes to end the book. In fact you don't even have to kill one, but the more of the agents you kill the easier book 4 becomes. And I mean, this is a game book with very limited technical possibilities... just think about what you can do in a CRPG if you put some effort in it.

I firmly believe that nowadays we have all means to create a non-linear CRPG with a deep and compelling storyline. Programmers just have to make an effort.

Ionstormsucks said:

I firmly believe that nowadays we have all means to create a non-linear CRPG with a deep and compelling storyline. Programmers just have to make an effort.

Click to expand...

Any thoughts on the specifics?

<pet peeve rant mode>
Please don't use programmer as a stand in for any game developer. Programmers have very little to do with the story of a game. The only real connection being that they write the tools and systems that the designers/writers use to create and deliver the story to the player. That being said yes improvements in the tools probably can help in creating better storys, but only indirectly.
<rant mode off>

One thought that hit me while reading the article was that it should be possible to use hypertext style to describe the flow of a games narrative to the designers in a very rough way. This might help in plotting more complex networks of player choices, but I guess it wouldn't really change what can be done just make a good prototyping tool.

Prime Junta said:

Any thoughts on the specifics?

Click to expand...

I guess that a lot depends on the construction of the story itself. Very often a story can be split up in different parts (and several games already did that in the past in form of chapters... for example "The Witcher" or "Baldur's Gate"). The trick (or difficult part) is of course to construct the story in a way that you do not have to play these chapters in a particular order and then adjust the rest of the game accordingly. From a technical point of view that's no problem. Almost every engine is able to adjust certain areas of a game while the player is actually playing the game (to a certain extent "The Witcher" for example does exactly that). It's also not a big problem to program that the decisions of one chapter carry over to the next. Moreover you can apply a certain randomness to npc behavior or certain events which also creates a feeling of non-linearity.

For me this is not a technical problem but a problem deeply rooted in design and logistic. Not every story can be turned into a non-linear game that faithfully transports the spirit of the original story, but you can create stories that fit the requirements of non-linear gameplay. To be more specific - very often CRPGs start with some kind of quest or mission (go there, do that), but for non-linear gameplay it might be better suited to let the player find out what the mission is or could be.

From a logistic point of view such complex games are probably a nightmare, but hey, I never said it would be easy.

Loof said:

@V7


@margerette
I have never understood why cRPG people hate and look down on LARPS so much, LARPS are just like PnP RPGs a form of interaktive storytelling. They can be almost anything from improvisational theater about current political events, to juvenile dress up and beat each other with a stick. For me the fact that there are more LARPS of the later type than the former doesn't make me despise the concept.

Click to expand...

Apologies, as I didn't mean to slam LARP or imply it was in any way inferior to other forms of role-playing. I know very little about it, and actually admire those who do such things as historical re-enactments in costume and so forth. As a form of theatre, I'm sure it could be a lot of fun.

I mistakenly lumped it in with my aversion to reality TV, and that is just a highly personal sense of distaste for drawing entertainment from playing to the nastier sides of human nature. Again, that's just a matter of personal taste, and others may see it quite differently.

I thought it was a good article, and I'm glad I read it.

Costikyan obviously gave this careful consideration and laid out his analysis in a way that made sense to me, especially his points about the nature of stories and games and how they're set in partial opposition. I found myself agreeing with his conclusion that some modern game styles are attempts to resolve the “tension between the demands of game and the demands of story.”

He mentioned but didn’t spend any time dwelling on what’s been a pivotal point in these discussions here in the past, that developers are unwilling and/or unable to develop content that a player might not use.

If it lacked anything, Costikyan’s analysis stopped short of making two significant distinctions about MMOs, that they employ client-server architecture and that their customers pay on a subscription basis. The first is important because it enables and also limits their ability do the things he describes. The second circumvents the $60 price point that shackles makers of single-player cRPGs.

I imagine many of the people who subscribe to World of Warcraft aren’t interested in the other players as much as they want to experience something like a living game world. It’s the other players, actually, that tend to ruin the experience.

Single-player cRPG makers should strongly consider how they might compete for those customers. If those monthly checks found their way to their bank accounts instead, what kind of games could they afford to make then, and with what kinds of stories?

Ionstormsucks said:

I guess that a lot depends on the construction of the story itself. Very often a story can be split up in different parts (and several games already did that in the past in form of chapters... for example "The Witcher" or "Baldur's Gate"). The trick (or difficult part) is of course to construct the story in a way that you do not have to play these chapters in a particular order and then adjust the rest of the game accordingly. From a technical point of view that's no problem. Almost every engine is able to adjust certain areas of a game while the player is actually playing the game (to a certain extent "The Witcher" for example does exactly that). It's also not a big problem to program that the decisions of one chapter carry over to the next. Moreover you can apply a certain randomness to npc behavior or certain events which also creates a feeling of non-linearity.

Click to expand...

Right, but there's nothing new to this -- it's the "string of pearls" model described in the article. Been done too; KOTOR and KOTOR II, for example, let you play the mid-game planets in any order you like.

For me this is not a technical problem but a problem deeply rooted in design and logistic. Not every story can be turned into a non-linear game that faithfully transports the spirit of the original story, but you can create stories that fit the requirements of non-linear gameplay. To be more specific - very often CRPGs start with some kind of quest or mission (go there, do that), but for non-linear gameplay it might be better suited to let the player find out what the mission is or could be.

From a logistic point of view such complex games are probably a nightmare, but hey, I never said it would be easy.

Click to expand...

More to the point: do you believe that the effort would be better spent on constructing, testing, and debugging a game to make it more non-linear -- or on adding more story, more characterization, more depth, and so on?

In RPG Solo-Adventures (so are they called for DSA), there are several ways, solutions possible, depending on the choice of the player. If you chose answer one, to to page 15, if you chose answer 2, go to page 30. If you chose neither, go to page 100.

Imho *good* RPGs are of this type: Allowing several possibilities and results in the end. But this is difficult to manage because of the many variables.

It might be that with Neural Networlks within each NPC, things might get better, because the reactions can be quite different, then. But, on the other hand, isn't there then the possibility that such a game becomes an simulation instead of an RPG ?

I can imagine cRPGs with multiple versions of quests, sub-plots and even main plots – all to suit the character(s) being played (or even how they’re being played). Like mods, those variations could be automatically downloaded to provide a tailored fit for the individual tastes of the player.

There are only two things I can see that have prevented that from happening already. The first is simply inertia. It’s always been about the current version. That’s all the choice we’ve ever had, that or an old version. That’s how we’re used to seeing it.

But a “current version” is only really necessary for client-server applications like an MMO, because every player needs it. Single-player can be modded. There’s no need for everyone to play the same version.

But the bigger reason is the $60 price point. There’s just not enough money there to afford big development projects. That’s why I think developers should compete with the MMOs for customers who are willing to pay monthly subscription fees. Done correctly, single-player RPG worlds can seem much more alive than the ones in MMOs.

Just for clarification, I’m not talking about simply updating or expanding games, and I’m not talking about development on demand. I’m talking about a way to enable players to have unique experiences every time they play a new character. I’m talking about dramatically better game worlds that provide a lot more variety. And I'm talking about a way for developers to be able to afford to do that.

@margerette
My rant about cRPG players bashing LARPs wasn't really a rant against you, you just set it of. And yes reality tv shows suck.

@Alrik
While Neural nets might be a use full subpart of a solution for more interactive and proactive NPCs, they aren't in and of themselves a magic tool that can solve the problem. As far as I know scientists still haven't managed to even model a brain as advaced as a bees brain with neural net.
But I do agree that the key to making more engaging stories in games is to make the NPCs more active so they can react to events in the world instead of just waiting around for the player to reach their placement in the story. But I think some of you already discussed this in another thread.
I would go one further and say that they shouldn't just react to the players actions, they should also react to each others actions. But figuring out how to implement such a system in a way that doesn't break down and/or cost millions to develop is not trivial.

And for this to even be a consideration developers need to get over their block about content that cant all be seen in a single play through (maybe it's not developers but the guys paying the bills that have this block, I don't know).

Loof said:

And for this to even be a consideration developers need to get over their block about content that cant all be seen in a single play through (maybe it's not developers but the guys paying the bills that have this block, I don't know).

Click to expand...

Don't forget consumers. If someone sold a game that had 47 completely separate paths through it each of them an hour long there would be people complaining about how short the game was. And in some respects I would have to agree. I think I'd rather play a 47 hour well-crafted story than 47 separate stories.

Attempts at some sort of compromise often are, frankly, lame. It is usually set up so that you do exactly the same missions with the same outcomes with some trivial changes based on your supposed motives.

BillSeurer: But what if those 47 separate paths all spring off, say, 6 different characters, each giving a unique lens and position to affect or observe the same story/set of events? It sounds intriguing to me.

KazikluBey said:

BillSeurer: But what if those 47 separate paths all spring off, say, 6 different characters, each giving a unique lens and position to affect or observe the same story/set of events? It sounds intriguing to me.

Click to expand...

Exactly. You could take that a lot further too.

Imagine a game world with a ten-decade history in twenty different versions where a character could be introduced at any time and in any place. Imagine the myriad of ways that story might be altered by the character you choose, the choices you make and the ways that world might respond.

Now imagine all the money you would need to create something like that.

Finally, imagine all the money being made by World of Warcraft.