Approaches to interactive storytelling

[jefferytitan]

Hi guys, I'm interested in what models there are for effective interactive storytelling (i.e. not pick-a-path style). I've thought of a few myself, I'd be interested to know what you know/have thought up, and what you think is the most effective.

- Avalanche. You create something almost unstoppable for the character to run from. They have some choice where they flee to, and particularly enterprising players may oppose the avalanche, but basically there is a forward direction.
- Lockdown. Lock them out of already completed areas. Careful design would be needed to make sure they don't end up in a dead-end.
- Radar. Constant visual reminder of where their next goal is, either by a device or by visual prominence (colour, flashing, biggest thing they can see).
- Companion. An NPC/NPCs that tells them what they need to do. Could get annoying.
- Goal. You tell the player what their goal is and let them figure out how to achieve it. This seems to suffer from the player not knowing "what to do next" or getting lost in an overly large world.
- Baiting. If the player gets lost or bored, send bait characters or a breadcrumb trail of items to lead them to where they need to be.
- Deus Ex Machina. If they're off track, just yank them back to where they should be. Make all portals go to the right place, pick them up with a tornado, or whatever neccessary to get them to the right place.
- Intention guessing. You create a bunch of possible goals, figure out which goal the player is closest to, and egg them on to achieve it, possibly using other methods mentioned.

[jefferytitan]

Actually, two more thoughts:

- Psychology. Figure out what the player is seeking (i.e. challenge, easiest path, NPC interaction) and provide more of that along the path you want them to take.
- Interesting world. Make the world interesting and complex enough, and the player will find something to do. Of course this is difficult.

All of these options offer a ratio of choice to perceived choice. Which do you prefer?

[DFG]

How about goals based on philosophical and/or religious worldview? Game interprets player choices to be strongly aligned with a philosophy or religion and then presents more choices based on the alignment of the character. Kind of a progression on the good, neutral, evil scheme but more advanced.

[jefferytitan]

Okay, that sounds kind of like an extension of the psychology/intent guessing angle. I'm actually quite interested in games with a philosophical basis. I think it shouldn't be in the player's face, but it adds depth for those who want to see it.

One concept I had was a god game from a first person perspective where your goal is simply to prevent Earth from destruction, and anything else is dictated by the player, including how they relate with people, e.g. god/superhero/hidden benevolent force. As the game progresses you get the chance to increase your powers, but those powers themselves make you dangerous to Earth.

Anyway, how important do you think it is to have an overall plot, as opposed to a series of interesting events? With the intent guessing you could actually construct plots for various philosophies, i.e. a kingdom's in civil war, you could (a) take it over (b) turn it into a democracy (c) support one of the parties or (d) just live with it.

BTW, is there any terminology for what I'm talking about? I haven't found any.


How about goals based on philosophical and/or religious worldview? Game interprets player choices to be strongly aligned with a philosophy or religion and then presents more choices based on the alignment of the character. Kind of a progression on the good, neutral, evil scheme but more advanced.

[Greg Squire]

You might want to look into Chris Crawford's approaches to interactive storytelling. He's devoted much of his life to tackling issues surrounding "true" interactive storytelling. This is not the simple branching approach like in "Choose your own Adventure" novels.

Here's his book - Chris Crawford on Interactive Storytelling (http://www.amazon.com/gp/product/0321278909/002-0921482-6535208?v=glance&n=283155)
and here's his current project surrounding his "research" - StoryTron (http://www.storytron.com/)

Also you might want to check out the game Facade (http://www.interactivestory.net/). Probably the most innovative "interactive story" to date.

[jefferytitan]

Interesting. Added to my wishlist, along with Ralph Koster's Theory of Fun. Until I have more money or more generous friends, what really struck you theory-wise from his work?


You might want to look into Chris Crawford's approaches to interactive storytelling. He's devoted much of his life to tackling issues surrounding "true" interactive storytelling. This is not the simple branching approach like in "Choose your own Adventure" novels.

Here's his book - Chris Crawford on Interactive Storytelling (http://www.amazon.com/gp/product/0321278909/002-0921482-6535208?v=glance&n=283155)
and here's his current project surrounding his "research" - StoryTron (http://www.storytron.com/)

Also you might want to check out the game Facade (http://www.interactivestory.net/). Probably the most innovative "interactive story" to date.

[Christian]

Hi, i am helping a little in the project of Chris Crawford, im evaulating the effectiveness of the tutorial of the authoring tool, and i have to tell you that storytron is pretty revolutionary, it has infinite possibilities, it also has a unique business models where authors of interactive stories get paid for their work, you should really check the "overview" part of the storytronics site because there you will find an excellent explanation of what it is about, much better that i could give you here.

[MikeRozak]

I have written a number of random ideas about stories in games on http://www.mxac.com.au/drt.

After thinking about this for awhile, I've concluded that you don't wan't to include stories (exactly), but elements of stories. If you try to create a story (as opposed to merely using story elements) you end up with a "game" where the user can't make any real choices, which is bad. See http://www.mxac.com.au/drt/GUT.htm for more thoughts.

[Greg Squire]

Interesting. Added to my wishlist, along with Ralph Koster's Theory of Fun. Until I have more money or more generous friends, what really struck you theory-wise from his work?

I accually haven't read his book on interactive storytelling (Yet), but I did listen to him at his session at the last IGC. He had a very unique approach to storytelling. The way I understood it is that the StoryTron, essentially creates/generates all the possible story branches at that moment based upon a set of rules and the current "story state". The rules are also to help remove "uninteresting" branches as well. But it only creates one branch level and not branches under that, at least not until the user takes one of those branches (based on user input). ;)

Don't know if that made any sense or not. Hopefully some.

I've read Chris Crawford's other book on Game Design and liked it. I also enjoyed "A Theory of Fun" as well.

[jefferytitan]

I guess it depends how big your lego pieces are. You could have some big fancy lego pieces that only come into play where appropriate, and a whole bunch of little ones which aren't as cool, but you can use for just about anything. For example, you could have a subplot about assasinating the evil king, but it only gets presented to you if you prove yourself to actually be against the "evil king".

I guess I just worry that with too much granularity your "story" could just turn out as random crap. I'm sure rules systems etc help a lot... but I'd need examples to be convinced.


I have written a number of random ideas about stories in games on http://www.mxac.com.au/drt.

After thinking about this for awhile, I've concluded that you don't wan't to include stories (exactly), but elements of stories. If you try to create a story (as opposed to merely using story elements) you end up with a "game" where the user can't make any real choices, which is bad. See http://www.mxac.com.au/drt/GUT.htm for more thoughts.

[mahlzeit]

From what I've read of Chris Crawford's ideas, he is simulating not the rules of everyday life (as in The Sims) but the rules of drama. If you study "old fashioned" storytelling, you run into concepts such as the character arc, conscious vs. unconcious desire, progressing complications, etc. But for an interactive story you only provide the basic building blocks at the thematic level. And then the non-player characters react to the player actions depending on how they are supposed to serve the theme.

So instead of programming a fixed storyline of Romeo and Juliet falling in love, fighting with their parents and families, and ultimately killing themselves, you would only set up the characters (the Capulet and the Montague family), the props (poison), the locations (Verona), and the theme (for example "love vs social obligations"). Basically, you design your interactive stories one level up from traditional storytelling. They are meta stories.

[jefferytitan]

Okay, I get that. I think that understanding the intent of the player in an arbitrary style of game (i.e. some have conversation, some don't, etc) could be tricky, and the intent is presumably what you'd be reacting to dramatically. I'm sure it works great for StoryTron.

I must admit that the StoryTron thing seems a little beyond what I intended the thread to be about. I was aiming at "what simple steps can you do to make arbitrary game x more interactive"... unless StoryTron is something that you could just license and apply to your next Half Life based game, or MMORPG or survival horror.


From what I've read of Chris Crawford's ideas, he is simulating not the rules of everyday life (as in The Sims) but the rules of drama. If you study "old fashioned" storytelling, you run into concepts such as the character arc, conscious vs. unconcious desire, progressing complications, etc. But for an interactive story you only provide the basic building blocks at the thematic level. And then the non-player characters react to the player actions depending on how they are supposed to serve the theme.

So instead of programming a fixed storyline of Romeo and Juliet falling in love, fighting with their parents and families, and ultimately killing themselves, you would only set up the characters (the Capulet and the Montague family), the props (poison), the locations (Verona), and the theme (for example "love vs social obligations"). Basically, you design your interactive stories one level up from traditional storytelling. They are meta stories.

[Davaris]

So has anyone seen a working demo of this StoryTron? Do you think it could be used in an RPG?

[jefferytitan]

That's one of the questions on my mind. I saw a (one) screenshot, but wasn't sure if that was playing or designing.


So has anyone seen a working demo of this StoryTron? Do you think it could be used in an RPG?

[mahlzeit]

I played a very old demo of his Erasmatron (the old name for the technology) once. The "story" was about a company meeting and you had to resolve a number of conflicts between the employees. It was only slightly interesting to play. That was about five years ago, though.

[Davaris]

I think if they made it so you can insert conversations into your games with NPCs it could be useful. But from the impression I get, its not going to work that way.

I'll bet women will love games made with it though. :)

[Christian]

Not only women i assure you, any kind of story can be made with it.

And no, it cant be used on an rpg, but you could create a story with dragons and magic and monsters and fighting and deaths and wars and epic stuff in storytron, it would be really cool, basically you could make something BETTER than an rpg.

And the screeonshot you saw was tof storytron, the playing part, the designing part is made with a tool called Swat (storywolrds authoring tool).

[Davaris]

I don't see how you can make something that is better than an RPG if you can't fight using real stats, level up and explore maps. If all you can do is talk, whats the point? They really need to make an API so programmers can use it in thier games. If they did this people would be lining up to use it.

[jefferytitan]

I wouldn't say "what's the point", but certainly it has much more potential to impact the game market if it's general-purpose. Personally I thought their method of talking to NPCs looked weird, so I'd like to be able to do my own UI. And have other styles of games, of course.


I don't see how you can make something that is better than an RPG if you can't fight using real stats, level up and explore maps. If all you can do is talk, whats the point? They really need to make an API so programmers can use it in thier games. If they did this people would be lining up to use it.

[MikeRozak]

Davaris wrote:


I don't see how you can make something that is better than an RPG if you can't fight using real stats, level up and explore maps. If all you can do is talk, whats the point? They really need to make an API so programmers can use it in thier games. If they did this people would be lining up to use it.

As I noted in http://www.mxac.com.au/drt/GUT.htm, CRPGs, adventures games, most MMORPGs, and even FPSs follow the following pattern:

1) The player controls one character (or a small group).

2) The character is given a goal, such as "Stop people from arguing in the meeting." At the same time, a story (or other incentive) is used so that the player internalizes the character's goal, and actually wants to end the arguing. Fable does a good job on this front: http://www.mxac.com.au/drt/FableAnalysis.htm.

3) The player must do some problem solving to determine how to accomplish the goal. FPSs require the player to figure out what obstacles to take cover behind. Adventure games require the solution of a puzzle. CRPGs require the player to determine what objects to buy, where to go, and how to overcome the baddies or traps. A well-designed problem-solving element allows for several different approaches: either sneak past the guards or attack them outright.

4) The player must be given means of acting on their solution to the problem. Platformers let players overcome goals by jumping from place to place. CRPGs (typically) rely on combat. Adventure game actions are usually trivial once the puzzle's solution is know. FPSs allow jumping and shooting. Again, choices are important.

5) Once the player completes the goal, the player is rewarded with a happy ending, their character is rewarded with loot/xp (which also rewards the player), and/or the world changes. Again, if possible, provide the player a choice over the outcome.

6) Repeat the process.


Just because combat is the traditional solution to the problem, doesn't mean it has to be. Stealth (aka: Thief) is another possible solution. Talking to NPC (not just clicking on a dialogue tree) is a third. You could even use "combat" tactics with speech, by spreading rumors, convincing co-workers to do stupid stuff and embarass themselves, etc.

[Christian]

Well, if you like rpgs just because you level up and explore maps then i dont think that storytron is the thing for you, but if you think on the story part of rpg, then you really could make something better than a common rpg. Also, you can do more than talk, thare are many other verbs that you can make, think on any other than "talk", and you can do it in storytron. Also, you have stats in storytron too, and you could level up if you wanted to, and you could kill monsters, and people, and fight, and o magic, and explore places, but not in the way that rpgs do.

[Davaris]

Also, you have stats in storytron too, and you could level up if you wanted to, and you could kill monsters, and people, and fight, and o magic, and explore places, but not in the way that rpgs do.

Dunno... :) I'll have to see it in action before I'm convinced of this.

One of the hardest parts for me in RPG creation is the dialogues. The work burns me out and I know its not very good. If I could create characters that react according to given characteristics, then it would make RPGs so much easier to make.


Just because combat is the traditional solution to the problem, doesn't mean it has to be. Stealth (aka: Thief) is another possible solution. Talking to NPC (not just clicking on a dialogue tree) is a third. You could even use "combat" tactics with speech, by spreading rumors, convincing co-workers to do stupid stuff and embarass themselves, etc.

Yes you're talking to a rabid Fallout fan here. :) So I expect this sort of thing in RPGs as well.

[shaft]

It will be interesting to see if "storytron" does become a reusable "interaction engine" for use in games.

As an application, storytron definately looks interesting, but based on what I've seen (which is very little, and likely early builds) it looks like storytron is missing a very important aspect to storytelling... Total immersion.

The GUI looks functional, but clunky.

Integrated into a specific game, the game could overcome that barrier, and the combined result would be very good.

Now granted storytron will probably evolve into something with a great front end, and allow for very immersive story telling. But it would be truly GREAT if games could use it and add game elements like rules, physics, other forms of AI, etc. to create something more than a story (aka A game).

If I were the business person there, I'd be using storytron as a profitable way to beta test the story element engine, and then eventually license it out for use in other applications. And I highly suspect that is there plan ;)

[Christian]

You say that "to create something more than a story", well, storytron are not stories!, they are interactive storyworlds, that is, you dont really right a story, but you create all the verbs that could happen in a world, and you just throw the player in to do whatever s/he wants, so its trully interactive and inmersive because all what the player do will affect the outcome in totally different ways each time s/he plays the storyworld, ending with hundreds of posibilities for each storyworld.

You can download and play with Swat right now, its the pre-alpha but its usable, storytron beta will be for testing in summer, and there is a lot to learn, so if someone is interested in it, i suggest starting now because this thing is complicated, and its going to grow into a big thing.

[jefferytitan]

I think the previous commenter means that although the story is an important component of the game experience, the UI is as well. Certain UI features may be required to get the game designer's vision across, as well as the fact that certain groups of gamers feel more comfortable with a specific UI.

For example, if Half Life 3 was released as Storytron (with the Storytron UI I've seen examples of), a lot of long-term customers would be confused and put off. Advanced storytelling would be very useful in the Half Life universe, but 3D FPS is a necessary component of its game model.

BTW, where would I find Swat?


You say that "to create something more than a story", well, storytron are not stories!, they are interactive storyworlds, that is, you dont really right a story, but you create all the verbs that could happen in a world, and you just throw the player in to do whatever s/he wants, so its trully interactive and inmersive because all what the player do will affect the outcome in totally different ways each time s/he plays the storyworld, ending with hundreds of posibilities for each storyworld.

You can download and play with Swat right now, its the pre-alpha but its usable, storytron beta will be for testing in summer, and there is a lot to learn, so if someone is interested in it, i suggest starting now because this thing is complicated, and its going to grow into a big thing.

[Christian]

www.storytron.com and then go to the part that says "Swat".

[Savant]

Then download the Java run time and then unzip the swat download and then burrow in 2 or 3 directory levels to find the JAR file to run...

Couldn't they have packaged this thing any better?

[Christian]

Its a pre-alpha! even storytron is not yet released, so, its just to start learning the Swat and practice a little., bugs, lack of functionality and problems like you have mentioned are understandable.