[MoriartyL]

Originally Posted by Fop

What's the point of player interaction at all if the player doesn't know how his choices affect the character's behaviour (or the options don't perform consistently, which amount to the same thing)?

First of all, the fact that you're expecting to control what he's saying creates a connection between you and the character, even when you're not controlling him very much at all. I think that even with minimal interactivity, it does make the character feel like "your character". And with that investment comes a certain sense of responsibility for the outcome of the conversation, which I think is always good.

More importantly, interactivity adds a tremendous amount of depth. It makes it feel less like a script, and just a little bit more like a real conversation. At least, that was the intent. I think it worked out okay. When the tone of the dialogue is a bit flexible, it feels more real. Also, there is depth in the sense that the outcome is not necessarily certain. My blog post could have ended with the character running away so that he wouldn't have to hear my idea. Again, I think it adds a touch of realism that the ending is not taken for granted.

[MoriartyL]

Originally Posted by jjacob

Well, most adventure games don't have linear dialogue, which is why the dialogue tree was created. I'm not sure how that would translate to your idea. How were you planning to handle multiple topics in a conversation?

I don't see the problem- I could have the player character change topics in the middle, have the NPC change topics in the middle, or I could have one topic lead naturally into the next. I think dialogues in which the player controls when to jump to completely unrelated topics and the NPC goes along as if there's nothing odd about it (which is to say, most modern dialogue trees) feel very contrived. In a real discussion, you'll always try to bridge the gap between the two topics, and that should be certainly be scripted.

Well, as I said, a keyword accompanying an icon would be an improvement IMHO - it would avoid unnecessary confusion but still retain the core improvement of your idea (again, IMHO). Let me use it in an example:

Yes, I understood what you were saying. And I'm thinking about it.

[Fop]

Originally Posted by MoriartyL

More importantly, interactivity adds a tremendous amount of depth. It makes it feel less like a script, and just a little bit more like a real conversation.

Pressing a button and getting a seemingly random response is not useful interactivity. You don't need three buttons for that.

[Jackal]

There's a big difference between "control" and "awareness", but you seem to want to treat them as one. Either that or you're so concerned about the former that you're sacrificing the latter. Part of any good narrative is having the reader/audience/player identify with a character. You're pretty much eliminating that possibility with your approach. If your goal is to SURPRISE the player, then you're basically just jerking people around, not effectively presenting a character.

Unfortunately, you're really not even doing much to limit control. Players are never really in control in any game. There are always multiple options that are pre-scripted, and the player simply picks one. That's not control. All it does is create an illusion of control. So what you're doing is removing that illusion and leaving the player with pretty much nothing.

I find it strange that you're trying to refer to this as "interactive" storytelling when it has the opposite effect.

Originally Posted by MoriartyL

First of all, the fact that you're expecting to control what he's saying creates a connection between you and the character, even when you're not controlling him very much at all. I think that even with minimal interactivity, it does make the character feel like "your character". And with that investment comes a certain sense of responsibility for the outcome of the conversation, which I think is always good.

You can pretty much kiss that connection, investment, and sense of responsibility goodbye the second the player realizes that their input amounts to guesswork.

I'm not saying some form of this couldn't work. Like others here, I certainly SEE what you're trying to do, and find your claims that "you don't like it just because it's different" to be needlessly insular. And yeah, some forms of this have been used already, to varying effect. But none that I can think of regard the player's involvement as being so insignificant.

[Fop]

I liked the Gabriel Knight 3 system where the player picked a subject out of a bunch of icons and the actual talking was left to Gabriel.

[Dasilva]

The guy speaks truth, it can be a real drag when the dialogs are in trees, Its more spontanious this way. I like this idea.

[MoriartyL]

Just to clarify: The actions of the buttons are not random. Here is what they mean:

Blue ?: inquiry
Red ?: challenge
Blue !: agreeing statement
Red !: opposing statement
Blue ...: considering what has been said, in an effort to accept it
Red ...: considering what has been said, in an effort to reject it

The player is indeed in control of which of these courses the PC will take, but since the red and blue paths are always opposed to each other, which of the two colors is shown depends on the PC's personality and current mood, so that the player should not be allowed to negate the PC's identity.


Jackal, I don't understand most of your complaints. I call it interactive storytelling because that's what it is. It couldn't possibly work without interactivity. The player is controlling the character to a certain extent. He doesn't have complete control, because the identity of a character must have some control as well, but he is the one pushing the character around. And I don't understand what you mean by "guesswork"- what is the player guessing? It's always pretty straightforward.

[Jackal]

Well, of course it's still interactive in the sense that it requires the player to click buttons, but interactivity is all about giving the player more control (or at least creating the illusion of it). As soon as the player loses the sense of making a relevant contribution, then they're left feeling like all they're doing is clicking buttons. And if you're continually guessing at what a dialogue choice will actually elicit from the character, you're doing just that. And that's less interactive, not more.

Yes, I know you have a "meaning" for each button, but the fact that they create dialogue lines the player doesn't anticipate means they're left guessing. That's why you're getting comments about randomness. They're not random under a larger umbrella, but in terms of specific direction, they seem random.

You say you're trying to make dialogue more character driven, but if you're effectively communicating a character, then the responses should become intuitive. The complaints here aren't that the choices weren't what people WANTED them to be, but that there was no way to predict the direction they would take.

Don't get me wrong. I like the notion of having a dialogue evolve more naturally than simply picking one line after another. But you need to give players enough information to make them aware of what they're choosing. That's why you're already getting suggestions of more detailed options.

[Aj_]

Originally Posted by MoriartyL

The player is indeed in control of which of these courses the PC will take, but since the red and blue paths are always opposed to each other, which of the two colors is shown depends on the PC's personality and current mood, so that the player should not be allowed to negate the PC's identity.

This is not something I understood. Lots of games do this, they don't use colours to signify it, but the writer considers the response of the character through text instead.

I understood that the colour signified mood, but I don't really understand it in the context of the conversation. The icons seemed to go from red to blue, but the mood didn't change, or there was no reason for the mood to change.

My biggest problem is that three options out of a pool of six is very limiting. Agree/oppose, inquire, and consider. A sentence is a lot more complicated than that. A conversation is even more so. Take conversations from books or games, interactive and non-interactive. I think this system is going to fail to model a non-interactive one, where the path is already set. It's also not going to be able to depict conversations in a game like Fallout.

That and consider probably shouldn't be red or blue, but neutral, which also would be a nice option to have for the others. Agree and inquiry also being rather neutral, while challenge and oppose are definitly red. Of course, sometimes people want you to challenge or oppose them, not to agree with them, so oppose could be blue.

It's not much more than a simplication of what is already used in games, with a smaller pool of types of response, but instead of the actual dialogue, you're given an icon instead.

[MoriartyL]

Originally Posted by Jackal

Yes, I know you have a "meaning" for each button, but the fact that they create dialogue lines the player doesn't anticipate means they're left guessing. That's why you're getting comments about randomness. They're not random under a larger umbrella, but in terms of specific direction, they seem random.

That is, until you take a moment to think about why it happened, and see the character. If you mean that the player is guessing what the character will say next, I really don't see how this can be seen as a bad thing.

You say you're trying to make dialogue more character driven, but if you're effectively communicating a character, then the responses should become intuitive. The complaints here aren't that the choices weren't what people WANTED them to be, but that there was no way to predict the direction they would take.

Now that's not true at all. Take any branch, think for a minute about who the character is, and you'll know what he's going to say next.

Don't get me wrong. I like the notion of having a dialogue evolve more naturally than simply picking one line after another. But you need to give players enough information to make them aware of what they're choosing. That's why you're already getting suggestions of more detailed options.

I'm still thinking about whether any more detail could be beneficial. I'm really not sure. As it stands now, the player learns a lot about the character by seeing what he says. It's not unreasonable at all to think that over the course of a longer game, the player would eventually get to know the character well enough that he would almost always be able to anticipate what the character will do when he presses one of the buttons.

[jjacob]

Originally Posted by MoriartyL

I don't see the problem- I could have the player character change topics in the middle, have the NPC change topics in the middle, or I could have one topic lead naturally into the next. I think dialogues in which the player controls when to jump to completely unrelated topics and the NPC goes along as if there's nothing odd about it (which is to say, most modern dialogue trees) feel very contrived. In a real discussion, you'll always try to bridge the gap between the two topics, and that should be certainly be scripted.

But wouldn't that result in overly monotonous and linear dialogue? "Nice house you have here, so... Anyway, did you happen to see someone sneaking around your garden suspiciously?" I just don't see how your idea makes dialogue any less contrived than what we're used to in adventure games. I understand how you want to give a protagonist more character however, but wouldn't it really put even more distance between the player and the protagonist/PC?

Originally Posted by MoriartyL

Yes, I understood what you were saying. And I'm thinking about it.

I just think it would avoid unnecessary confusion Especially in the beginning of a game, where the player would still have to get acquainted with the protagonist's character.

Originally Posted by MoriartyL

That is, until you take a moment to think about why it happened, and see the character. If you mean that the player is guessing what the character will say next, I really don't see how this can be seen as a bad thing.

Why what happened? The player has to anticipate the dialogue intuitively, not evaluate it in hindsight. We may be talking about a significant amount of trial-and-error in places players would not expect nor desire it. And, as I said, it might take as long as half the game before the player gets acquainted with the protagonist's character. Besides, people (even fictual ones) are complicated, unless we're talking about extremely stereotypical two-dimensional "flat characters", or unless you'd want to use it as a game mechanic, where the player is in control of a schizophrenic in a mental hospital, and where the protagonist would constantly react very differently from what the player would expect, conveying his mental illness in the game (dialogue) mechanic.

Originally Posted by MoriartyL

Now that's not true at all. Take any branch, think for a minute about who the character is, and you'll know what he's going to say next.

I have to disagree with you there; how the hell would the player know what say, "inquiry" would result in? Besides, I don't want to think a minute before saying something, I don't want to have 20-minute conversations with an NPC and having to constantly relate to a PC's character, especially if that person is something I don't like or totally can't relate to (and I've had this with many protagonists already). I do want to have fluent conversations. Having played a fair amount of AGs, I usually have no trouble finding a dialogue choice that seems appropriate and fitting in its context, and usually within seconds. I also like how the dialogue choice in, for example, TLJ, could result in a totally different response from the PC; when you want to insult Zack for instance, April says something totally different than what you'd expect her to say. For some reason the dialogue in that game managed to convey her character pretty well; you instantly got the impression she was kind, well-mannered, helping, caring and insecure etc.

Originally Posted by MoriartyL

I'm still thinking about whether any more detail could be beneficial. I'm really not sure. As it stands now, the player learns a lot about the character by seeing what he says. It's not unreasonable at all to think that over the course of a longer game, the player would eventually get to know the character well enough that he would almost always be able to anticipate what the character will do when he presses one of the buttons.

As I said, is that the kind of trial and error a player would find desirable in a game? And as you say, "eventually" may not be quick enough for the many people with a relatively short attention span. Besides the quality of the dialogue would have to be of high quality to keep the player interested, perhaps even moreso than in adventure games with dialogue trees.

[MoriartyL]

Originally Posted by jjacob

(on keywords)
I just think it would avoid unnecessary confusion Especially in the beginning of a game, where the player would still have to get acquainted with the protagonist's character.

At the beginning... hm, it could work. What I'm concerned about is that it may seem like the PC is thinking too much about what to say. Often a story (especially if it's an RPG) starts with the PC in a comfortable routine, so that just wouldn't make any sense. Better to just have the early dialogues be fairly inconsequential, so that the player just gets the hang of the three buttons in a risk-free environment, and introduce the keywords only when trouble starts. But then taking them away would be awkward. Nah, it wouldn't work.

Why what happened?

Why the character acted differently than the player anticipated.

The player has to anticipate the dialogue intuitively, not evaluate it in hindsight. We may be talking about a significant amount of trial-and-error in places players would not expect nor desire it.

I don't understand what you're talking about. What trial and error? Didn't I say the idea was to have the player only go through each dialogue once? If it goes badly, you move on. Oh well. That's life. Or do you mean an error in anticipating the character's next move? There's no error unless the player makes a faulty assumption. That's the player's problem, not mine.

And why should the player know exactly what the character's going to talk about? Here, I'll give you an example. Early on in a game, the character should always know more than the player. I mean, he wasn't born at the moment the player started playing. Try to work around the player's inability to anticipate dialogue, and he can't get to know the backstory by being surprised by the PC's statements. The player should try to anticipate the dialogue, certainly, but I don't see why he should be able to right from the start. He should grow into the character with time.

And, as I said, it might take as long as half the game before the player gets acquainted with the protagonist's character.

Is this a bad thing? It's good for replay value, the connection between player and character will be more satisfying, the character will seem to have more depth. Where's the problem in keeping the player in the dark for a while? I think it's good that he should have to put something into it to get anything out. We're so used to culture spoon-feeding us, telling us everything we need to know right at the start, but it doesn't have to be that way.

I have to disagree with you there; how the hell would the player know what say, "inquiry" would result in? Besides, I don't want to think a minute before saying something, I don't want to have 20-minute conversations with an NPC and having to constantly relate to a PC's character, especially if that person is something I don't like or totally can't relate to (and I've had this with many protagonists already).

It's an acquired language. A little time with the three buttons, and you'd start to understand it intuitively. I have- Even though I charted out the whole thing, it's hard for me to think of any of that as I'm playing. All I see is that I'm giving a command and the character's following it in his own way. Then again, it may help that I've dealt with this character often before, so I know him well. At first, the player might have to struggle a bit, but I think it would quickly become second nature to control the character.

As I said, is that the kind of trial and error a player would find desirable in a game? And as you say, "eventually" may not be quick enough for the many people with a relatively short attention span.

I'm not interested if the player will find it "desirable", or if the player has ADD. If the player isn't sophisticated enough to handle a good character, screw 'im. I'm sick of all this lowest common denominator-targeting trash we see. Players can adapt to something more serious.

[Fop]

I wish you'd answer these critiques with a bit less hubris. It hardly seems likely that you've come up with a perfect system that doesn't even need tweaking.

[MoriartyL]

True, it does need tweaking. But on the content level, not the structural level. On the structural level, I think it's perfect.

Okay, so I'm in love with my creation. Sue me.

[Spiwak]

Originally Posted by MoriartyL

Okay, so I'm in love with my creation. Sue me.

I guess that's all that matters in the end.

Try actually implementing in this a game, if you can, where there'll be more than one rather simple conversation, and see how that works. Who knows, it might work quite well.

[Jackal]

Originally Posted by MoriartyL

That's the player's problem, not mine.

And here we have it in a nutshell. If you've got this kind of disdainful, antagonistic attitude towards the player, it's no wonder you've come up with such a self-indulgent system.

You're trying to pass this off as being about character, but character really has nothing to do with it. Yes, if the character is a bully, and the player anticipates him saying, "I want my wool mittens", then that's a character issue. But most people are more alike than they are different, so the vast majority of responses will fall somewhere along predictable lines. Except, as has been pointed out, conversations are complex and dynamic. It can go in one of hundreds of directions at a moment's notice, and if you leave the player out of THAT process, you're causing disconnection, not the connection you mysteriously seem to think you are.

I think you've created a nice little experiment in a kind of "discover the character" exercise. Or more accurately, "learn the system". It's a trial-and-error system with replay value, if one cares enough to replay it. But as an interactive storytelling device, it's counter-productive, as it diminishes the player's role. So, maybe if you used an interesting character study, it'd be more appealing on its own.

Is this a bad thing? It's good for replay value, the connection between player and character will be more satisfying, the character will seem to have more depth. Where's the problem in keeping the player in the dark for a while? I think it's good that he should have to put something into it to get anything out. We're so used to culture spoon-feeding us, telling us everything we need to know right at the start, but it doesn't have to be that way.

It's interesting you should call the other methods "spoon-feeding", when it's you who's making the player's role more passive. You say yourself it's about removing control from the player, then turn around and say they have to put more into it? Um, no. They put less into it, since they're now forced to follow blindly and make only vague attempts to direct progress.

Besides, you're making either/or arguments that don't apply. You're just obscuring the specifics to limit the player's involvement, that's all. Any dialogue system can reveal character at the pace it desires. The only question is how much the player is involved in the process.

I'm not interested if the player will find it "desirable", or if the player has ADD. If the player isn't sophisticated enough to handle a good character, screw 'im. I'm sick of all this lowest common denominator-targeting trash we see. Players can adapt to something more serious.

Huh? There's nothing more "serious" about your method, and really it's you who's dumbing down the process (unless you're only comparing it to a "click all the options until they're gone" method.) Good character? A good system doesn't make a good character, it only delivers it. The question is whether or not it delivers it effectively. It's clear you love your own system, but if you've made the player stop caring, it certainly hasn't done its job. And whether or not your character is good is dependent on the writing, as always.

[MoriartyL]

Okay, well, thank you for your opinion. Naturally, you know I'm not going to listen to anything you're saying, so I won't bother to make excuses.

By this point in the discussion, I've figured out exactly where progress needs to be made in the area of content, and also come to realize that I am madly in love with this form of dialogue, so there's really nothing more for me to get out of this thread. Bye.

[MoriartyL]

Oh, and thanks.

[Karmillo]

...

[Jeysie]

If I might ask, what were you planning on doing with this experiment, Mory?

If it is just a personal experiment, then I could understand ignoring feedback (although IMHO that makes posting a thread about it a complete waste of time).

But if you plan on ever using this for practical purposes - i.e. putting it in a game - then ignoring player feedback is the virtual equivalent of shooting yourself in the foot.

Personally I agree with Jackal... I think one of the problems facing the adventure game genre (at least in terms of "mainstream viability") is that it limits "real" interactivity compared to other genres, and this sort of system, while interesting in its own ways, would just exacerbate that problem.

I could maybe see it working in a "virtual dollhouse" sort of game, like a "The Sims" sort of deal, where the whole point of the game is to do nothing but observe characters you don't directly control anyway.

Peace & Luv, Liz