167 réponses à ce sujet

[In Exile]

Xewaka wrote...
How do you know the blinking button that appears for maybe a couple seconds means your character will grab and kiss his/her lover? That's the problem with action interrupts, they simply lack a way to inform the player of wether the action they propose is in-character with how the player wants its character to act. It's the same problem paraphrases have, but magnified.

With a reasonable scope, you can set up rational interrups.

Let's say you're playing a bounty hunter RPG, and you have to catch a criminal. Early on, you get a dialogue choice that determins whether you want to capture the target alive or dead. Some time later, that target takes a hostage.

You can then have choices. A [draw weapon] choice. A "Stop or I'll shoot!" dialogue option. And then, as the target responds, an interrupt button. Whether it's shoot to kill or not is determined by your previous choice. Alternatively, you could have two triggers (wound & kill, so labelled).

Alternatively, they could do it the ME way (the way it was originally advertised for ME1, not what they did for ME2) where if you chose a dialogue option before the NPC finished speaking, you'd cut them off.

[Xewaka]

In Exile wrote...
With a reasonable scope, you can set up rational interrups.
Let's say you're playing a bounty hunter RPG, and you have to catch a criminal. Early on, you get a dialogue choice that determins whether you want to capture the target alive or dead. Some time later, that target takes a hostage.
You can then have choices. A [draw weapon] choice. A "Stop or I'll shoot!" dialogue option. And then, as the target responds, an interrupt button. Whether it's shoot to kill or not is determined by your previous choice.

What if, since the moment you made the choice about killing or not killing, the player has received new information input that would cause the character to rethink its options? You're chained by a former choice, done before the new information caused the character to reevaluate its options.

In Exile wrote...
Alternatively, you could have two triggers (wound & kill, so labelled).

This would be more agreeable, but still, a single word interrupt would require the scene to be very contextualized to actually have a semblance of information.

In Exile wrote...
Alternatively, they could do it the ME way (the way it was originally advertised for ME1, not what they did for ME2) where if you chose a dialogue option before the NPC finished speaking, you'd cut them off.

I'd rather not get involved in a debate on how terrible the paraphrase wheel is as a mechanic, as it has the same misinformation problem. It could work, but the simple act of revising your options would probably always consume the time to interrupt the other party speech. Unless the dialogue wheel appears right after your last speaking choice, but then the player attention is divided between the speech and the wheel options, which probably means he's ignoring one of them.

[In Exile]

Xewaka wrote...
What if, since the moment you made the choice about killing or not killing, the player has received new information input that would cause the character to rethink its options? You're chained by a former choice, done before the new information caused the character to reevaluate its options.

That's a fair point, but you could easily add such trigger dialogues just prior to having an 'interrupt opportunity'.

Would you have any objection if there was no intermediate option that could influence the player's choice?

This would be more agreeable, but still, a single word interrupt would require the scene to be very contextualized to actually have a semblance of information.

I wouldn't use words. As with DA2, I think a standardized set of symbols are better and easier for players to keep in mind.

In Exile wrote...
I'd rather not get involved in a debate on how terrible the paraphrase wheel is as a mechanic, as it has the same misinformation problem.

For the sake of this example, let's ignore the paraphrase. Rather, think of it simply as allowing for dynamic dialogue input. If you make your dialogue selection before the NPC finishes speaking, it interrupts.

It could work, but the simple act of revising your options would probably always consume the time to interrupt the other party speech. Unless the dialogue wheel appears right after your last speaking choice, but then the player attention is divided between the speech and the wheel options, which probably means he's ignoring one of them.

I didn't consider that. I'd be a bigger issue with the wheel than without the well (not to mention that the spoken dialogue would change).

It might work with the dialogue wheel, but it won't without it, I can't see it. Anyway, I think dynamic dialogue (where the PC can be active in conversation) will be a major step foward for RPGs.

[Miashi]

I like the interrupts in Mass Effect 2...
I know that there were a handful of situations where you could involve your companions in the conversation, but why not make certain conversation interrupts happen if a certain companion is with you?

[Alex Kershaw]

Didn't really make any difference to the overall ME2 experience for me so if you want to do it, I'm not complaining but I'm not on the edge of my seat either. In fact, I'd rather you didn't just because I don't know how it could be implemented without a paragon/renegade meter - DA2 strayed too close to paragon/renegade already.

[Ohpus]

JohnEpler wrote...

I'm a big fan of the interrupt system, personally. I believe you can still implement it without the necessity of attaching it to a Paragon/Renegade sort of meter, though I would agree that you have to be careful about telegraphing them properly and ensuring that the person who uses them has at least some idea what's going to happen.

But wouldn't you ruin the whole suprise factor of thr Interrupt system by telegraphing the intent?

Now you could easily pull it of by the setting, such as Zevran putting moves on the Warden and using an Interrupt to break his nose.

[Sylvius the Mad]

[quote]In Exile wrote...

Haha.

I was using that in the colloquial sense of quick to act. Not to suggest an actual lack of thinking.

That being said, you can analyze a situation rapdily and throw yourself in the middle of it without being stupid, though I would wager you'd respond that the behaviour I suggest is in and of itself stupid. [/quote]
Maybe not stupid, but probably reckless.

You're far more tolerant of risk than I am.  I'm quite risk-averse, and I play risk-averse characters.
[quote]At any rate, I think it's just a reaction to a subjective feeling the feature is present moreso than it being well done or a standpoint compared to any other dialogue tone.[/quote]
It always bothers me when people form opinions based on a "subjective feeling" when they could just measure the differences to draw an informed conclusion.
[quote]Humour is meaningful (not for the player, but for the PC). I have to ask, what do you consider meaningful that humour does not fit?[/quote]
Humour doesn't convey meaningful information to my listeners, nor does it ask useful questions.

If my character is trying to acheive some objective withing a given conversation, humour doesn't help him do that.  or rather, the humourous options, as presented in DA2, don't help him do that, because I can't tell what information Hawke is going to convey or solicit within the joke.  The humour paraphrases tend to focus on the subject of the joke, and that's not what's important.

This could just be a paraphrase problem.

Also, it bothers me that I'm expected to choose an option simply based on its supposed humour, without knowing whether it is the type of humour in which my character would engage.  In DAO, my Warden could make jokes all over the place, and whether they were jokes was determined by my PC's perception.  I can't do that in DA2 because the wheel hides the content from me.
[quote]The diplomatic options in DA:O were more sycophant than actually silver-tongued diplomatic. [/quote]
They are, I think, the most effective sarcasm in the game, though.  Having Hawke be insincerely sycophantic really works, I think.

And there's the problem with the humour options.  If I want my character to be funny, I don't need the game to tell me when I'm doing it.  I know when what I've said is funny.
[quote]You know, replaying DA:O, sometimes I feel that speaking to you in person would be a little like speaking to Sten. [/quote]
Maybe that's why Sten is my favourite DAO companion.

Though, I didn't like how the Warden's conversations with Sten were written.  Sten I liked, but the Warden didn't ever adapt his expression to suit Sten's responses.  It's as if the Warden was unable to learn how to talk to Sten.
[quote]Really? I grew up in Eastern Europe, and they were my go-to treat. Found some recently in at Shoppers Drug Mart and was ecstatic. [/quote]
Apparently the FDA thinks they're a choking hazard because of the toy inside.  I can't imagine that ever being a problem unless you toss the entire egg in your mouth at once, but... their country, their rules.
[quote]I found the paragon interrupts in ME2 to be incoherent and unpredictable (though the renegade, being much more narrow to violent psychopath, were at least predictable in the violent psychopath-ness). What was it about the dialogue options themselves that you found unpredictable?[/quote]
Dealing with Cerberus.  Choosing Paragon options with Cerberus turned Shepard into a belligerent lunatic, but the whole of ME had taught us that Paragon = diplomacy.

I never understood it.
[quote]Let me clarify. When I mention the developers, it's not to say that their intentions have value in-character. Rather, what I meant was that from a design standpoint, they wanted that topic to be broached (so it was not, as with my lamentation over the lack of an attempted Cousland usurping of the crown at the landsmeet, simply something not implemented) but something the designers chose to allow an NPC to say but not the PC.[/quote]
As it happens, I think DAO should have allowed a Cousland to try to seize the crown.  I also think they should have allowed a Cousland to fail in that endeavour and die defeated at the end of the game.  But that's not relevant to your point.
[quote]To give you an example, the HN in DA:O meets with Cailan. Cailan asks why Papa Cousland is not present. The HN can mentions the father's death in some way. Cailin then addresses his question to Duncan. Now, that's fine. That's Cailan's business.

But the HN can't respond or interject. Duncan is speaking. My characters would interrupt Duncan to - among other things - accuse him of criminal behaviour for abandon the Teyrna and Teyrn to death to make away with a recruit and tell the story himself. [/quote]
While I'll grant that in that situation there should be some way to allow the PC to interject (this might actually be a good use for dialogue interrupts - have the interrupts trigger the dialogue wheel, so that players can choose how to intervene - thus allowing both active and passive dialogue participants to stay in-character), the writers can't imagine all of the things you might ever want to say.  That you might want to accuse Duncan of criminal behaviour (which would make no sense, given how the Wardens are governed) isn't something the game is always going to be able to allow.

You know as well as I do that the available options are finite.  But certainly I agree with you that if an answer is to be given the PC should at least sometimes be given a chance to offer it without first being deferred to by those around him.
[quote]To respond to your question, it's not that I view the character as the spokesperson (I actually love that in DA2, it made perfect sense for Hawke not to be a leader at all; Hawke only spoke most of the time because it was Hawke who badgered his friends to come along with him). Rather, my characters would simply talk anyway.

Like how other characters can chime in without being prompted when someone is addressing the Warden.[/quote]
I liked those interjections in DAO if they were directed at the Warden, but if my party members spoke to others without the party's consent I would want to be able to challenge them immediately for having disobeyed the will of the group.
[quote]To be honest (and this just came to me) if a game honestly had silent VO and allowed for frequent dialogue interrupts and otherwise let the PC be very active (KoTOR 2 actually does a good job of this at points) then I wouldn't object to silent VO anywhere near as much. In fact, the only thing that would bother me would be a subjective perception of difference, because unlike most, I cannot read with a tone. My reading voice is universal and monotone.[/quote]
My speaking voice is universal and monotone, so that's not really a problem for me.

This has been really helpful.  I think I've found a way that interrupts can work without being a barrier to roleplaying (have the interrupt trigger not a response, but a choice to respond), and you may have found a way for interrupts to mitigate the problems associated with a silent protagonist.
[quote]I want to volunteer the information most of the time, so I never run into the problem.[/quote]
You're so trusting.

The event in ME that bothered me most with regard to volunteering information was a bigger problem because the information being offered by Shepard wasn't even something I believed to be true.  I didn't think Shepard had enough evidence to draw the conclusion she was suddenly telling her superiors about, so not only was she not doing what I wanted by I thought she was making a fool of herself.
[quote]But ME2 did manage to break my Shepard and get me to the point where I had to debate restarting the game. And that was when Shepard started volunteering he was working for Cereberus.

I eventually came up with a tortured rationalization (Shepard was trying to leave a huge enough trail that the Alliance would catch the Illusive Man), but otherwise I never had the problem. [/quote]
If I ever have to rationalise my character's behaviour, the game is unplayably broken.  If I need to perform mental acrobatics to justify my character's behaviour, I need to be able to do that before my character performs the action.
[quote]See what I suggested above about silent VO interrupts. It may be that would be a system that works for us. I'd certainly compromise on the lack of a voice if the PC could be as active as I wanted to.[/quote]
And I would allow interrupts if I didn't need to make blind choices about whether to use them.

This is our best discussion ever.
[quote]I thought about this playing the Witcher 2, and contrasting how passive Hawke felt overall in DA2 by comparison.[/quote]
I'm sending you a PM about that.
[quote]Why do you feel the character I described holds an opinion (aside from valuing fidelity)? [/quote]
More importantly, I don't think the character you described necessarily holds no opinions.  He could well hold opinions about all sorts of things, but as long as those opinions are either over-ridden by his love of fidelity, or those opinions drive the initial promises and he never encounters information to change his mind - or perhaps he's stubborn and won't ever admit he made a mistake.

In DA2, I found valuing fidelity to be fairly difficult.  There was a quest involving recovering the son of a Magistrate, and I wanted to return him alive because I'd agreed to do so in return for some benefit.  But then when the time came to make the ultimate choice to return him alive or not, I could choose to return him only out of some sense of hope for his redemption, rather than just a dispassionate view of him as a tradeable good (which is how Hawke viewed him).

This is another place where I think the voice and intent icons limit roleplaying quite a bit.  DAO (and BioWare's earlier games) tended to offer a fairly neutral response at most dialogue option events, and I could safely choose those without worrying about whether the game thought I meant something different.  Even ME and ME2 offered those if you icked the middle option on the wheel.  But DA2 usually doesn't offer a simple "He's coming with me" that leaves the PC's intention unspoken.

And that's what the game needs.  DA2 should give us one neutral response where we can mean whatever we like by it and not have to bombard the NPCs with a bunch of pointless exposition.  If I'm taking the son away, or killing him, does Hawke really need to explain why he's doing it?  By writing the quest like that, the game limits us to playing one of the Hawke's they imagined.  But just letting us act without explaining ourselves doesn't have that problem.
[quote]Whereas there is a logically possible scenario that covers that, the best way to put it is... there isn't a scenario I can believably construct for my character that would have Alistair act in-character, my character act-in character, and be consistent with no aligment change and no character change for him.[/quote]
Perhaps you just don't know him well enough to know why he would behave that way.

This could be a difference between us.  I think the people I meet are so vastly and weirdly different from me that I don't really expect to understand why they do what they do, so I just view other people's behaviour as their behaviour, and I leave it at that.

It seems to bother you that Alistair's behaviour doesn't add up in your mind, while I would be surprised if it ever did.
[quote]Whereas there are an infinite number of possible states in between any two states, that does not mean any particular state imaginable fits. Infinite != all. [/quote]
Of course.  I agree.
[quote]Not at all. Behaviour isn't unpredictable; it's... reasonably predictable, to use that phrase. Which is to say that you have a higher than chance sucess rate at predicting it. It's quite honestly a science.

It comes down to what you consider an adequate success rate.[/quote]
This is why I think populations are predictable, but individuals are not.

I might be able to predict that 80% of people in some large group will behave a certain way, and I can be right about that over and over again.  I have tremendous confidence in that sort of claim.

But if you point to one individual in the group and ask me how he will behave, I have no idea.  There's an 80% chance that he's part of the majority, but 80% isn't nearly good enough.  Moreover, the group from which I'm selecting needs to be exceptionally homogenous for me to predict the behaviour of a single member with any confidence, an it's not.

I do some direct marketing.  I have mountains of data that tells me that between 1-2% of the people I contact with unsolicited mail will respond to it.  I can solicit a group of 40,000 people and get that 1-2%, and then I can solicit the same 40,000 people (minus the people who responded last time) and get another 1-2%.  It works every time.  I'm perfectly content to predict the behaviour of the group.

But will any given person respond to my mail?  Probably not - 98% chance says no - but if I need that level of homogeneity to have the confidence to act, then I can almost never act, because I almost never have the data available to tell me what the odds are.
[quote]Certainly, you could assume that people are entirely incomparable. But that's prima facie false. There are overlapping similarities between people.[/quote]
But not predictable overlapping similarities.
[quote]But you and I believe know means different things. You think know is binary. I don't.

I can have as reasonable a belief that a person will behave in X way as I can that the sun will rise tomorrow or that electical discharges may be fatal. [/quote]
Since I'm confident that you are aware that your predictions about the behaviour of individuals have a much higher error rate than your predictions about the earth's rotation, I honestly don't know what you mean by "as reasonable a belief".  If the error rate is higher, then it's less reasonable, isn't it?

And regardless of whether you think you can know how people will respond using your definition of "know", you're still going to be wrong a lot, so designing a character around the requirement that his oration is effective seems doomed to failure.

I also think that a game designed like that would feel really contrived to me, because my character would be able to make people do things just by telling them to do those things, and that's entirely unlike how I expect people to behave.

[Sylvius the Mad]

Xewaka wrote...

In Exile wrote...

With a reasonable scope, you can set up rational interrups.
Let's say you're playing a bounty hunter RPG, and you have to catch a criminal. Early on, you get a dialogue choice that determins whether you want to capture the target alive or dead. Some time later, that target takes a hostage.
You can then have choices. A [draw weapon] choice. A "Stop or I'll shoot!" dialogue option. And then, as the target responds, an interrupt button. Whether it's shoot to kill or not is determined by your previous choice.

What if, since the moment you made the choice about killing or not killing, the player has received new information input that would cause the character to rethink its options? You're chained by a former choice, done before the new information caused the character to reevaluate its options.

That would certainly be my concern.

I think interrupts should trigger a new dialogue wheel event wherein the player can choose to perform the interrupt action (or choose among a few options) or return to the scene without taing any action at all, as if he hadn't triggered the interrupt.

[In Exile]

[quote]Sylvius the Mad wrote...
It always bothers me when people form opinions based on a "subjective feeling" when they could just measure the differences to draw an informed conclusion. [/quote]

They do, as much as they can. You pressupose that it's possible to draw an informed conclusion over a feature that somehow sidesteps subjectivity.

[quote]Humour doesn't convey meaningful information to my listeners, nor does it ask useful questions.[/quote]

But that's not all communication needs. Humour can enrage a person your speaking with, take them out of their element. It can lighten the mood, making future discussion easier. It can help to ingratiate you. Humour does lots of things. Thinking about it in purely informationl terms is too narrow.

[quote]If my character is trying to acheive some objective withing a given conversation, humour doesn't help him do that.  or rather, the humourous options, as presented in DA2, don't help him do that, because I can't tell what information Hawke is going to convey or solicit within the joke.  The humour paraphrases tend to focus on the subject of the joke, and that's not what's important.[/quote]

Like I said: the information itself isn't anywhere near as valuable as controlling the social interaction. It's the person and their future actions that matters. Generall speaking, in DA2, when you the actual information options (e.g. interrogate and choices) were independent of tone.

[quote]Also, it bothers me that I'm expected to choose an option simply based on its supposed humour, without knowing whether it is the type of humour in which my character would engage.  In DAO, my Warden could make jokes all over the place, and whether they were jokes was determined by my PC's perception.  I can't do that in DA2 because the wheel hides the content from me.[/quote]

It's the only humourous option the game allows. But you and I take very different positions on what it means for a game to allow something.

[quote]They are, I think, the most effective sarcasm in the game, though.  Having Hawke be insincerely sycophantic really works, I think.

And there's the problem with the humour options.  If I want my character to be funny, I don't need the game to tell me when I'm doing it.  I know when what I've said is funny.[/quote]

I need the game to constantly tell my what I'm doing. That's how I know it recognizes that my action is correct. It's entirely unrealistic, but it's important feedback for the player. It's similar to a DM telling you whether a particular roll succeeded.

[quote]Though, I didn't like how the Warden's conversations with Sten were written.  Sten I liked, but the Warden didn't ever adapt his expression to suit Sten's responses.  It's as if the Warden was unable to learn how to talk to Sten.[/quote]

That also bothered me. It made me feel the Warden was an exposition tool for Sten.

[quote]Dealing with Cerberus.  Choosing Paragon options with Cerberus turned Shepard into a belligerent lunatic, but the whole of ME had taught us that Paragon = diplomacy.

I never understood it.[/quote]

I never chose Paragon options for Cerberus. I always pick the neutral one. It just didn't make sense to me. If I'm working with this organization, I'm not going to rock the boat.

[/quote]As it happens, I think DAO should have allowed a Cousland to try to seize the crown.  I also think they should have allowed a Cousland to fail in that endeavour and die defeated at the end of the game.  But that's not relevant to your point.[/quote]

I think a Cousland could win the Landsmeet. But like you said, neither here nor there.

[quote]While I'll grant that in that situation there should be some way to allow the PC to interject (this might actually be a good use for dialogue interrupts - have the interrupts trigger the dialogue wheel, so that players can choose how to intervene - thus allowing both active and passive dialogue participants to stay in-character), the writers can't imagine all of the things you might ever want to say.  That you might want to accuse Duncan of criminal behaviour (which would make no sense, given how the Wardens are governed) isn't something the game is always going to be able to allow.

You know as well as I do that the available options are finite.  But certainly I agree with you that if an answer is to be given the PC should at least sometimes be given a chance to offer it without first being deferred to by those around him.[/quote]

That was only an example of how powerless I felt in the scene, and with Duncan in general. The game was very clear on what attitude you had to have toward him (it apparently couldn't be hatred that he basically kidnapped you).

In general, my objection is only to being forced to be passive. Hawke did that a lot. I didn't like it.

[quote]I liked those interjections in DAO if they were directed at the Warden, but if my party members spoke to others without the party's consent I would want to be able to challenge them immediately for having disobeyed the will of the group.[/quote]

It comes down to politeness. But I don't think an RPG should restrict a PC on that spectrum.

[quote]My speaking voice is universal and monotone, so that's not really a problem for me.[/quote]

My problem is that my reading voice is universal. So if I was reading a label or reading dialogue, it amounts to the same. It can't sound like a person to me.

[quote]This has been really helpful.  I think I've found a way that interrupts can work without being a barrier to roleplaying (have the interrupt trigger not a response, but a choice to respond), and you may have found a way for interrupts to mitigate the problems associated with a silent protagonist.[/quote]

I've actually thought of that myself. The only issue would be how to implement the NPCs response (my idea is that the game pauses and then the NPC is cut off, maybe reacting to the interruption after you're done, or interrupting you). 

[quote]You're so trusting.[/quote]

It's a recurring problem. I just don't focus on controlling the content of what I say as closely as you do.

[quote]The event in ME that bothered me most with regard to volunteering information was a bigger problem because the information being offered by Shepard wasn't even something I believed to be true.  I didn't think Shepard had enough evidence to draw the conclusion she was suddenly telling her superiors about, so not only was she not doing what I wanted by I thought she was making a fool of herself.[/quote]

Was that with the Council? I had to restart the game at that point.

[quote]If I ever have to rationalise my character's behaviour, the game is unplayably broken.  If I need to perform mental acrobatics to justify my character's behaviour, I need to be able to do that before my character performs the action.[/quote]

It's no different, for me, with DA:O forcing me to out myself as a Warden and actively support the order as frequently as it did.

[quote]And I would allow interrupts if I didn't need to make blind choices about whether to use them.

This is our best discussion ever.[/quote]

Entirely. It's the fact I was so excited for DA2. The ways it let me down helped me refine what I really wanted in a game, and going back to replay DA:O and BG helped me to realize what I enjoyed about those games in the first place. 

[quote]This is another place where I think the voice and intent icons limit roleplaying quite a bit.  DAO (and BioWare's earlier games) tended to offer a fairly neutral response at most dialogue option events, and I could safely choose those without worrying about whether the game thought I meant something different.  Even ME and ME2 offered those if you icked the middle option on the wheel.  But DA2 usually doesn't offer a simple "He's coming with me" that leaves the PC's intention unspoken.

And that's what the game needs.  DA2 should give us one neutral response where we can mean whatever we like by it and not have to bombard the NPCs with a bunch of pointless exposition.  If I'm taking the son away, or killing him, does Hawke really need to explain why he's doing it?  By writing the quest like that, the game limits us to playing one of the Hawke's they imagined.  But just letting us act without explaining ourselves doesn't have that problem.[/quote]

I think that's precisely my problem with the old Bioware RPGs - the options were always neutral. Passionate NPCs (one way or the other) are my favourites. I can't enjoy a character if I can't have an outlet for that passion.

[quote]Perhaps you just don't know him well enough to know why he would behave that way.

This could be a difference between us.  I think the people I meet are so vastly and weirdly different from me that I don't really expect to understand why they do what they do, so I just view other people's behaviour as their behaviour, and I leave it at that.

It seems to bother you that Alistair's behaviour doesn't add up in your mind, while I would be surprised if it ever did.[/quote]

The world as you see it must be terribly chaotic. I think behaviour is predictable, and empirically speaking, I'm often right (if I had to ballpark it, 75% of the time, maybe?) 

[quote]This is why I think populations are predictable, but individuals are not.

I might be able to predict that 80% of people in some large group will behave a certain way, and I can be right about that over and over again.  I have tremendous confidence in that sort of claim.

But if you point to one individual in the group and ask me how he will behave, I have no idea.  There's an 80% chance that he's part of the majority, but 80% isn't nearly good enough.  Moreover, the group from which I'm selecting needs to be exceptionally homogenous for me to predict the behaviour of a single member with any confidence, an it's not.

I do some direct marketing.  I have mountains of data that tells me that between 1-2% of the people I contact with unsolicited mail will respond to it.  I can solicit a group of 40,000 people and get that 1-2%, and then I can solicit the same 40,000 people (minus the people who responded last time) and get another 1-2%.  It works every time.  I'm perfectly content to predict the behaviour of the group.

But will any given person respond to my mail?  Probably not - 98% chance says no - but if I need that level of homogeneity to have the confidence to act, then I can almost never act, because I almost never have the data available to tell me what the odds are.[/quote]

There's no reason to think about people as populations. So long as you take the notion of causality seriously, predicting the behaviour of individuals is nothing more than a question of the cause of any future behavour, and it's reflexive. If I know your behaviour now I can try and predict the causes of it, and if I know your circumstances (e.g. potential causes) I can try and predict your future behaviour. 

It's heuristic, but there's no real inconsistency. 

In your case, you have no knowledge at all about your population. But if you knew that all but one have a terrifying fear of paper, to the point of running away from any instance of it, whereas feels a moral obligation to respond to communications, you can come up with a reasonably (i.e. right most of the time) accurate prediction of who will and will not respond. 

Obviously the example is silly, but the idea is that we can predict behaviour so long as we can isolate the proximate causes of it. 

[quote]But not predictable overlapping similarities.[/quote]

Sure there are. Biologically. Not enough to know in the strong sense how much our experiences overlap, but certainly enough to know that they are relevantly similar, and so similar that any one person can use themselves as a template for any other person with reasonable success (unless there is some significant biological deviation). It's a continuum, which I know you don't like, but it's effective.

[quoteSince I'm confident that you are aware that your predictions about the behaviour of individuals have a much higher error rate than your predictions about the earth's rotation, I honestly don't know what you mean by "as reasonable a belief".  If the error rate is higher, then it's less reasonable, isn't it?[/quote]

Not if the error rate is as low as possible in comparison to the success rate. The goal is optimization. If I want to predict behaviour, and I can predit behaviour much better than chance, then even if my method is objectively worse when applied to people versus protons, it can still be a good method.

[quote]And regardless of whether you think you can know how people will respond using your definition of "know", you're still going to be wrong a lot, so designing a character around the requirement that his oration is effective seems doomed to failure.[/quote]

A game is different than reality in one crucial way: the game has to script every interaction. The only possible actions are those that the designers allow for. Thus, a line cannot be said other than how they create it to be said. One meaning to a line. For the purposes of the game world (which has a finite number of states, unlike our world) the meaning is actually not ambiguous.

[quote]I also think that a game designed like that would feel really contrived to me, because my character would be able to make people do things just by telling them to do those things, and that's entirely unlike how I expect people to behave.[/quote]

Your character can. That's like... a Bioware RPG in a nutshell.

[Sylvius the Mad]

[quote]In Exile wrote...

But that's not all communication needs. Humour can enrage a person your speaking with, take them out of their element. It can lighten the mood, making future discussion easier. It can help to ingratiate you. Humour does lots of things. Thinking about it in purely informationl terms is too narrow. [/quote]
But  those other objectives are achieved through the words chosen by the speaker.  If I want to enrage someone, let me choose the line that I think will enrage him.  Don't tell me which line will enrage him and let me choose the outcome.  That's an unrealistic approach to how language works.

You don't choose people's responses.  You choose words that you hope will elicit those responses.


[quote]Like I said: the information itself isn't anywhere near as valuable as controlling the social interaction. It's the person and their future actions that matters. Generall speaking, in DA2, when you the actual information options (e.g. interrogate and choices) were independent of tone. [/quote]
But every line is an information option.  The writers can't know when the PC is going to want to try to learn things, so they can't write the game assuming that.

[quote]It's the only humourous option the game allows.[/quote]
And that's fine, but let me know whether it's a humourous option my character would choose before making me decide.


[quote]I need the game to constantly tell my what I'm doing.[/quote]
You don't already know what you're doing?


[quote]That's how I know it recognizes that my action is correct.[/quote]
I don't understand what you mean here by "correct".  Your action is what you chose to do.  I'd say that it's correct if it's in-character.  What do you mean?


[quote]It's entirely unrealistic, but it's important feedback for the player. It's similar to a DM telling you whether a particular roll succeeded. [/quote]
I don't generally want the DM to do that.  I want the DM to tell me what's happening, and let me draw my own conclusions.

As a player, I know what calculations determine whether I will succeed, and I might have a good idea what the values of the relevant variables are in this specific circumstance, but I don't think the DM should tell me wheher I succeeded.

This is one way in which I think even BG failed.  Hiding-in-shadows became too easy because the player always knew with certainty whether the character was hiding successfully, and I don't think that's information he should have.


[quote]That also bothered me. It made me feel the Warden was an exposition tool for Sten.[/quote]
It created a bit of a disconnect for me, as I thought that I understood Sten, but for some reason my character didn't.


[quote]I never chose Paragon options for Cerberus. I always pick the neutral one. It just didn't make sense to me. If I'm working with this organization, I'm not going to rock the boat. [/quote]
I wanted them to think I was completely on-board and accepted all of their ideology.  But I couldn't seem to make Shepard say those things (actually, the Renegade options tended to say those things, which made even less sense - unless whoever wrote that section thought that Paragon-Renegade was actually a Good-Evil scale).
[/quote]My problem is that my reading voice is universal. So if I was reading a label or reading dialogue, it amounts to the same. It can't sound like a person to me. [/quote]
It's all just language.  I tend to convert spoken words to text in my head, so I'm just always working from text.  It all sounds the same, in that it doesn't sound like anything.


[quote]I've actually thought of that myself. The only issue would be how to implement the NPCs response (my idea is that the game pauses and then the NPC is cut off, maybe reacting to the interruption after you're done, or interrupting you). [/quote]
That was my thought at well.  Pause the line when the player pulls up the interrupt interface, and then either have the selected interrupt cut-off the spoken line or simply return to the spoken line if the player chooses not to act.


[quote]It's a recurring problem. I just don't focus on controlling the content of what I say as closely as you do. [/quote]
I think I benefit tremendously from viewing conversation as adversarial.
[quote]Was that with the Council? I had to restart the game at that point. [/quote]
It was talking to Anderson and Udina right after recusing Tali.  I don't recall the details of the exchange.


[quote]It's no different, for me, with DA:O forcing me to out myself as a Warden and actively support the order as frequently as it did. [/quote]
I don't recall DAO forcing you to declare yourself a Warden much.  There are certainly conversations in which you can do that, and they don't seem to go anywhere if you don't (the Templars in Lothering are like this), but usually you can not do it.

I'll be playing it again very soon - I'll check.


[quote]Entirely. It's the fact I was so excited for DA2. The ways it let me down helped me refine what I really wanted in a game, and going back to replay DA:O and BG helped me to realize what I enjoyed about those games in the first place. [/quote]
I think a lot of gamers - and developers - never go back to look at the older games to consider what they might have done better than the newer games.


[quote]I think that's precisely my problem with the old Bioware RPGs - the options were always neutral. Passionate NPCs (one way or the other) are my favourites. I can't enjoy a character if I can't have an outlet for that passion. [/quote]
As long as there's one neutral option - one way to say what needs to be said without going into greater detail - that would be fine.  I'm content to let youhave your passion.

Since there are dialogue options, the game should give us both.


[quote]The world as you see it must be terribly chaotic.[/quote]
The people in it certainly are.  That might be why I like rules so much - to constrain their behaviour in apredictable way.  Except that those rules get enforced and adjudicated by unpredictable people, so they often may as well not exist.


[quote]I think behaviour is predictable, and empirically speaking, I'm often right (if I had to ballpark it, 75% of the time, maybe?) [/quote]
So, when dealing with others, there's a 25% chance they'll do something you don't expect.  That's way too high a number for me to be comfortable.  Imagine if there was always a 25% chance that your employer would behave entirely unpredictably.  That would be an awful place to work.

When I work with people, for example, I get to know each of them well enough to know fairly well how they'll respond to specific types of behaviour, so I limit myself to that behaviour when dealing with them.  Outside of that behaviour, I have no idea how they'll respond, so I can't do those things.
[quote]There's no reason to think about people as populations. So long as you take the notion of causality seriously, predicting the behaviour of individuals is nothing more than a question of the cause of any future behavour, and it's reflexive. If I know your behaviour now I can try and predict the causes of it, and if I know your circumstances (e.g. potential causes) I can try and predict your future behaviour. 

It's heuristic, but there's no real inconsistency. 

In your case, you have no knowledge at all about your population. But if you knew that all but one have a terrifying fear of paper, to the point of running away from any instance of it, whereas feels a moral obligation to respond to communications, you can come up with a reasonably (i.e. right most of the time) accurate prediction of who will and will not respond. 

Obviously the example is silly, but the idea is that we can predict behaviour so long as we can isolate the proximate causes of it. [/quote]
Certainly.  I'm saying we can't isolate those causes with people we don't know extremely well.

I've worked in the same office for 11 years.  My direct superior is the same person it was 11 years ago.  My assistant has been my assistant for 3 years, and he worked here in another capacity for 5 years before that.  These are the people I work with on a daily basis, and I know them pretty well.  But people I met last month - I don't know them at all.

An example.  I write some things that need to be approved by the company's President, and he's only been here for about 18 months.  I hate writing those letters, because I don't know what he likes.  I have noticed that he tends to approve the letters I write, and reject letters that I have other people write for me (even if he doesn't know I didn't write them), but I haven't been able to identify the relevant differences between my writing and that of others in order to isolate the cause.

So I certainly can't do that within a 60 hour game where all of my interactions with these people were written by someone else.


[quote]Not if the error rate is as low as possible in comparison to the success rate. The goal is optimization. If I want to predict behaviour, and I can predit behaviour much better than chance, then even if my method is objectively worse when applied to people versus protons, it can still be a good method. [/quote]
There I would disagree.  The best method isn't necessarily an adequate method.

Imagine navigating at sea by using the stars.  If you don't know how to do that, but you've managed to create a system to predict stellar positions (badly), and that's the only means you have, that doesn't mean you should use that method to navigate at night.  What it means is that you shouldn't sail at night until you've found a better means of navigation.
[quote]A game is different than reality in one crucial way: the game has to script every interaction. The only possible actions are those that the designers allow for. Thus, a line cannot be said other than how they create it to be said. One meaning to a line. For the purposes of the game world (which has a finite number of states, unlike our world) the meaning is actually not ambiguous. [/quote]
But you're wrong.  The game doesn't need to script interactions at all.  It only needs to script individual lines.  The interaction doesn't really exist.

I don't understand why you think a line can't be uttered in a way other than how the writers intended (or expected).  The only requirement is that the game can only respond in a finite number of ways.  That restricts only the NPCs' behaviour, not everyone's behaviour.


[quote]Your character can. That's like... a Bioware RPG in a nutshell.[/quote]
That's not true at all.  With DA2's paraphrase and icon system, all I know is that I'm choosing the outcome.  I don't know how Hawke is going to acheive the outcome, but the game is telling me that it's going to work.

In BioWare's earlier games, I choose the line, and produces whatever outcome it produces.  I don't know in advance that it's going to work (and often it doesn't).  That's more like real-world interactions.

If the game is going to force my PC to know things and have (and use) abilities without my prior consent, then I can't be reasonably described as an active participant in the behaviour of that character.

Modifié par Sylvius the Mad, 25 mai 2011 - 08:21 .

[HTTP 404]

the most epic showdown: Sylvius vs In Exile....its not really an argument tho

Modifié par HTTP 404, 25 mai 2011 - 09:32 .

[In Exile]

[quote]Sylvius the Mad wrote...
But  those other objectives are achieved through the words chosen by the speaker.  If I want to enrage someone, let me choose the line that I think will enrage him.  Don't tell me which line will enrage him and let me choose the outcome.  That's an unrealistic approach to how language works. [/quote]

No. The delivery of a line can in itself enrage someone. Example:

That was a good idea! 
That was a good idea. <_<

THAT WAS A GOOD IDEA! 

The words chosen are identical. But the line varies in tone. It varies in body language. They are non equivalent, because the modifiers of speech, so to speak, vary. To say that communication is reducible to the words spoken is unrealistic. Even if you think, normatively, that words are all that should matter, people never speak as if only what they say matters.

[quote]You don't choose people's responses.  You choose words that you hope will elicit those responses.[/quote]

You choose much more than that: you need to choose body language, tone. And you have to be cognizant of what the other person is thinking of you. Because your behaviour in the past is used to evaluate your present behaviour.

[quote]But every line is an information option.  The writers can't know when the PC is going to want to try to learn things, so they can't write the game assuming that. [/quote]

They have to. Otherwise they cannot write dialogue. You used a very important concept above: elicit a response.

In RPGs, wherever there is player choice in dialogue, there follows a series of reactions after someone speaks. For the writers to even write dialogue, they must assume a way in which the line was said. As a result, the writers exhaust the world. It's finite versus infinite; there are only a limited number of possible reactions to a particular sentence. You are ''locked in''.

I think we've uncovered another interesting divide between us. You think that the writers must make every possible effort to accomodate the player, and that the ''buy-in'' (so to speak; I'm using this to refer to who has to accept certain premises to be true to play the game) is on the part of the designers on behalf of the player. I think it's on the part of the player.

This is why I strongly supported the Origins. It further disambiguates the possible future finite options, so you can plan a character that is not broken and the story is reactive to.

[quote]And that's fine, but let me know whether it's a humourous option my character would choose before making me decide. [/quote]

I agree with you there. But for any interaction to be real, the NPCs must meaningfully react to it in-game. I will gladly sacrifice vacuous control (for which there are no-in game consequences) for direction if it comes with greater reactivity.

 [quote]You don't already know what you're doing? [/quote]

I don't know what the game allows me to be doing. As I said: it is not a simulation; with finite options, I cannot be free.

[quote]I don't understand what you mean here by "correct".  Your action is what you chose to do.  I'd say that it's correct if it's in-character.  What do you mean? [/quote]

There are in-character options that are impossible in the game. Like debating with Morrigain or Alistair re: their beliefs. I had characters who would do this. They couldn't exist. Their existence was wrong in the sense that important properties to them could not be rendered in-game.

[quote]I don't generally want the DM to do that.  I want the DM to tell me what's happening, and let me draw my own conclusions. [/quote]

This is strange. You believe that the gameworld has fixed, natural rules. There are some features that are feedback for the player that are the equivalent to nature itself. For example, if you fall, you don't get to draw your own conclusion - you fall that the rate intrisinc to nature. The same happens when you get sick.

That being said, I don't think the influence metre is in-character knowledge. It's meta knowledge for the player.

[quote]As a player, I know what calculations determine whether I will succeed, and I might have a good idea what the values of the relevant variables are in this specific circumstance, but I don't think the DM should tell me wheher I succeeded. [/quote]

 [quote]This is one way in which I think even BG failed.  Hiding-in-shadows became too easy because the player always knew with certainty whether the character was hiding successfully, and I don't think that's information he should have. [/quote]

It's information the player should have. Why would you incoporate that decision into in-character?

 [quote]It created a bit of a disconnect for me, as I thought that I understood Sten, but for some reason my character didn't. [/quote]

Bioware did the same with Wrex. This feeling is the feeling I have with dialogue in Bioware games sometime (but particularly with silent VO), where the PC is used as a foil to develo an NPC.

The worst offender is Shepard being locked into ignorance with Wrex in ME, but there are multiple example in Bioware games were the entire design is build around the NPC getting a word off the PC.

 [quote]I wanted them to think I was completely on-board and accepted all of their ideology.  But I couldn't seem to make Shepard say those things (actually, the Renegade options tended to say those things, which made even less sense - unless whoever wrote that section thought that Paragon-Renegade was actually a Good-Evil scale). [/quote]

Different people wrote different sections of ME2 (I think only one person wrote the Normandy) and I don't think they ever had an hard outline of what the alignment means.

If you're going to force characters into an alignment, you need to have bullet points of what it means. A list that I, as a writer, could look at and decide whether the character response I am writing will be consistent with other responses.

Otherwise you have paragon Shepard threatening to break faces on Omega, which was another ME2 moment where the game just seemed to fly off rails for me.

[quote]It's all just language.  I tend to convert spoken words to text in my head, so I'm just always working from text.  It all sounds the same, in that it doesn't sound like anything.[/quote]

That's actually close to my experience of it. The more I think on it, the more my fundamental objection to silent VO comes not from the absence of tone itself. That doesn't have value for me other than for consistency; if NPCs speak and the PC doesn't the conversation is wrong because it forces me to switch between my reading voice and hearing and so becomes aversive; and I can't read the text myself (I tried) because then my character sounds like me, and that's useless.

DA2 made me realize that VO in ME added certain features of dynamic interaction that were simply missing from previous games and those features I liked. Tone (which is really what the voice is) is incidental to that.

 [quote]That was my thought at well.  Pause the line when the player pulls up the interrupt interface, and then either have the selected interrupt cut-off the spoken line or simply return to the spoken line if the player chooses not to act. [/quote]

It shouldn't make a difference from a standpoint of design. Like I said above: tone is just tone.

[quote]I think I benefit tremendously from viewing conversation as adversarial. [/quote]

In terms of naivety? Certainly. The loss, though, is the ability to steer a conversation. To use an analogy, it is the difference between a simplistic catch all 'attack' statistics and 'defence' statistic. Which I think makes sense, because I'm more active and risky than you are.

[quote]It was talking to Anderson and Udina right after recusing Tali.  I don't recall the details of the exchange. [/quote]

The Council Scene floors me. Shepard goes ballistic in a way that is incomprehensible to me. Why would anyone act that way?

 [quote]I don't recall DAO forcing you to declare yourself a Warden much.  There are certainly conversations in which you can do that, and they don't seem to go anywhere if you don't (the Templars in Lothering are like this), but usually you can not do it.

I'll be playing it again very soon - I'll check. [/quote]

There are converastions where you can avoid mentioned you are a Warden (Zathrian is a good example) and then, when you mention the Blight, he says you are a Warden. But you can never correct him. I don't like that.
 
[quote]I think a lot of gamers - and developers - never go back to look at the older games to consider what they might have done better than the newer games. [/quote]

It was your advice, indirectly. After all, you make the very good point I did like those games in the past, so there must have been some features I did like. Playing them made the focus on the precise structural differences between new-age Bioware games and older games.

 [quote]As long as there's one neutral option - one way to say what needs to be said without going into greater detail - that would be fine.  I'm content to let youhave your passion.

Since there are dialogue options, the game should give us both. [/quote]

Certainly. Even for me, passionate is sometimes inappopriate. Like with Shepard at the Council session. Shepard flying off the handle is just not something I want in my characters. Anger is generally useless unless you can use it to push someone, but then you have to have some power. When you're asking a superior for help, largely desperate and with little evidence, anger doesn't seem to do much.

 [quote]The people in it certainly are.  That might be why I like rules so much - to constrain their behaviour in apredictable way.  Except that those rules get enforced and adjudicated by unpredictable people, so they often may as well not exist. [/quote]

This is why frameworks are more important. Using a framework allows you to constrain the world.

I will respond to the rest shortly. I need more time to think on two points you raised. But I think this has been a very productive discussion for us.

[In Exile]

Sylvius the Mad wrote...
So, when dealing with others, there's a 25% chance they'll do something you don't expect.  That's way too high a number for me to be comfortable.  Imagine if there was always a 25% chance that your employer would behave entirely unpredictably.  That would be an awful place to work.

But you revise your framework on the fly. Yes, the framework as it stands (if it remains unrevised) is useless; but you constantly tweak it for the person until it predicts them well (close to 100%); it remains only generally predictive in the general case.

This is just like a scientific theory. In its infancy, it can predict some pheonema with some regularity and a small but significant margin of error. As you add more work, you can refine the framework to work well for a particular phenoma (say, understanding the photoelectric effect).

Think of it this way:

There is a domaing general module that can predict any behaviour 75% of the time. This is for any person you have never met.

Once you meet that person, you begin to build a person-specific module. It deviates from the general module by taking into account idiosyncracies. It gets close to 100% predictability as you are able to tweak assumptions further to account for idiosyncracy. There are dimishing returns, of course, but you get close to effectively always predictable.

Moreover, you are never stuck with a going from general to specific. If you meet a person that reminds you of another person, then you can use that person specific module to get a heard start.

All of this is dynamic. It is, to parallel a more mathematical concept, the difference between a connectionist network and an algorithm. It's just a structurally different way of approaching the same problem.

When I work with people, for example, I get to know each of them well enough to know fairly well how they'll respond to specific types of behaviour, so I limit myself to that behaviour when dealing with them.  Outside of that behaviour, I have no idea how they'll respond, so I can't do those things.

That's nothing more than saying that if the success rate isn't 100%, then it is equivalent to being 0%. And that's just a false dichotomy.

You need to appreciate that there is a continuum of possible knowledge.

It is true that you can be right all of the time. And it is true that you can be right none of the time. But can also be the case that you can be right some of the time. All of this from the same set of premises.

You discount the value in being right some of the time.

Certainly.  I'm saying we can't isolate those causes with people we don't know extremely well.

I've worked in the same office for 11 years.  My direct superior is the same person it was 11 years ago.  My assistant has been my assistant for 3 years, and he worked here in another capacity for 5 years before that.  These are the people I work with on a daily basis, and I know them pretty well.  But people I met last month - I don't know them at all.

I agree with you entirely. But this is just the difference between the person specific way of thinking and the person general way of thinking.

As I said above - it is not that you're not right about the substance of it. It's that you discount the ability to generalize or otherwise transfer knowledge across domains.

An example.  I write some things that need to be approved by the company's President, and he's only been here for about 18 months.  I hate writing those letters, because I don't know what he likes.  I have noticed that he tends to approve the letters I write, and reject letters that I have other people write for me (even if he doesn't know I didn't write them), but I haven't been able to identify the relevant differences between my writing and that of others in order to isolate the cause.

I once worked with someone who refused to give feedback. Ever. I couldn't understand it. I eventually confronted the person out of frustration, and she told me that her assumption was that if she wasn't complaining about my work, clearly I should assume it is good.

Which to an extent isn't unreasonable. But it was very difficult to appreciate my progress when I would show her data and she would only say ''Okay''.

There I would disagree.  The best method isn't necessarily an adequate method.

Imagine navigating at sea by using the stars.  If you don't know how to do that, but you've managed to create a system to predict stellar positions (badly), and that's the only means you have, that doesn't mean you should use that method to navigate at night.  What it means is that you shouldn't sail at night until you've found a better means of navigation.

That's a judgement call, though. It depends on what you set as the objective standard for 'good' navigation.

I agree with you that the best method isn't neccesarily an adequate method. But the fundamental question is what a good method is.

I would say that being right most of the time is good, whereas you would say it isn't.

But consider this: what if you had to navigate at night? No other option? This is what social interaction is like: we have to do it. It becomes a question of, given our isolated perspective, which is the best way to do it.

So to return to the issue of gaming, the parallel is like this: 

Your characters need to be able to have a great deal of control over their behaviour. The way you interat with the world is to always know perfectly what it is your doing; then you leave it up to the other person to make their move, similar to a game of chess. This is a passive strategy.

My characters need to be able to functionally interact with the
world. If I see that someone begins to misunderstand, or moves off my predicted path, I nudge them back. It is like trying to steer a boat in rapid current. Sometimes it is possible and sometimes it is not, but the fundamental approach does not change.

Do you see why, as a result, I value activity on the part of the PC in conversation?

If I choose a line and the NPC misunderstands, I would simply steer them back. It is an active process. I drive the conversation.

But you're wrong.  The game doesn't need to script interactions at all.  It only needs to script individual lines.  The interaction doesn't really exist.

The game scripts every possible interaction, in the sense that at the design level, there are a finite number of possible behaviours in conversations and the writers have exhausted them. There is no emergence.

Let me give you a counter-example. Suppose we have an open-world RPG. The designers give you tools: traps, spells and weapons. They design how each tool works. But what they don't design is their use - I could use my traps, spells and weapons in a way that no designer ever envisioned them being used; it would be an entirely unique combination that is made possible only by the combinatorial (and emergent) properties of these basic units.

To see a good example of emergence as a concept, google Cellular Automata.

I don't understand why you think a line can't be uttered in a way other than how the writers intended (or expected).  The only requirement is that the game can only respond in a finite number of ways.  That restricts only the NPCs' behaviour, not everyone's behaviour.

You agree with me that behaviour is causally determined, yes? That to act a particular way (even if it is unpredictable) there is a set state of physical affairs that led a person to acting that way?

A response is made up of (for the sake of simplicity): tone , wording , body language. Physical characterstics (like height, temperature, smell, etc.) also affect conversation, but these are impossible to ever include in a game so let's ignore them.

You think that for any tone or body language, a person could conceivably react in an identical fashion.

I am saying that this is impossible. You could imagine a player saying a line in an infinite number of ways: but the for the NPC to respond and remain the same person (even if you do not know what sort of person that is) the response must be consistent for that person.

That's not true at all.  With DA2's paraphrase and icon system, all I know is that I'm choosing the outcome.  I don't know how Hawke is going to acheive the outcome, but the game is telling me that it's going to work.

In BioWare's earlier games, I choose the line, and produces whatever outcome it produces.  I don't know in advance that it's going to work (and often it doesn't).  That's more like real-world interactions.

As I described above: that distinction only makes sense if you view conversation as passive.

If the game is going to force my PC to know things and have (and use) abilities without my prior consent, then I can't be reasonably described as an active participant in the behaviour of that character.

I agree with you on that; the problem is unpacking what this means. What you think it means to consent to an outcome in a video-game is not what I think it means to consent to an outcome. So we have an impass.

[Sylvius the Mad]

[quote]In Exile wrote...

[quote]Sylvius the Mad wrote...
But  those other objectives are achieved through the words chosen by the speaker.  If I want to enrage someone, let me choose the line that I think will enrage him.  Don't tell me which line will enrage him and let me choose the outcome.  That's an unrealistic approach to how language works. [/quote]

No. The delivery of a line can in itself enrage someone. Example:

That was a good idea! 
That was a good idea. <_<

THAT WAS A GOOD IDEA! 

The words chosen are identical. But the line varies in tone. It varies in body language. They are non equivalent, because the modifiers of speech, so to speak, vary. To say that communication is reducible to the words spoken is unrealistic. Even if you think, normatively, that words are all that should matter, people never speak as if only what they say matters. [/quote]
You're missing my point.  I'm saying that the writers can't know which of those lines my character might use to elicit a given response, so they need to let me know which one I'm choosing so I can make that decision.

I don't care which line actually elicits any given response.  I'm asking for the ability to choose among them without the game deciding for me which one I will use.

The way DA2 works now, the player tells the game what Hawke wants to accomplish, and DA2 has Hawke do the thing that accomplishes that.  That's backward.  The game should give the player options about what to do, which the player will select based on which one he thinks Hawke would use to elicit the response he wants.

[quote]You choose much more than that: you need to choose body language, tone.[/quote]
Absolutely.  The game should, as much as possible, let us choose that, too.

That's largely why I oppose cinematic dialogue and a voiced PC, but even with cinematic dialogue and a voiced PC the game could let the player choose among options.

[quote]And you have to be cognizant of what the other person is thinking of you. Because your behaviour in the past is used to evaluate your present behaviour.[/quote]
Umm, no.  Yes, the other person will use that in their evaluation, but the speaker doesn't have to use it.  They do if they want to be effective, but that's not mandatory.
[quote]They have to. Otherwise they cannot write dialogue. You used a very important concept above: elicit a response.

In RPGs, wherever there is player choice in dialogue, there follows a series of reactions after someone speaks. For the writers to even write dialogue, they must assume a way in which the line was said.[/quote]
Wrong.  For every line, the writers must assume a way in which it was understood.  That's the difference.

The writers need to know how the NPCs react to any given line.  That's all that matters.

[quote]As a result, the writers exhaust the world. It's finite versus infinite; there are only a limited number of possible reactions to a particular sentence.[/quote]
Yes.  There are a limited number of possible reactions.  No one is disputing that.  I'm saying that to do that, though, the writers don't need to force a specific delivery of each line.  That would only be necessary if there was only one possible interpretation of each line, and that's demonstrably false.

And, we're drifting away from the issue.  I'm talking here about the need of the game to let the player choose the line that he thinks his character would choose, rather than choosing the line that produces the outcome the character wants.

Those two things are only equivalent if his character has a perfect understanding of his audience under all possible circumstances, and that's not necessarily true.
[quote]I agree with you there. But for any interaction to be real, the NPCs must meaningfully react to it in-game. I will gladly sacrifice vacuous control (for which there are no-in game consequences) for direction if it comes with greater reactivity.[/quote]
This wouldn't reduce reactivity at all.  It would just make that reactivity less predictable for some character designs.

Furthermore, in-character there's no way to know whether the NPCs have reacted to a line, as you have no external frame of reference from which to make a comparison.  You can't know if the NPCs would have reacted differently to a different line, because the different line didn't happen.  That's why reactivity usually isn't a concern of mine, because for the player to notice it he has to be metagaming.

[quote]That being said, I don't think the influence metre is in-character knowledge. It's meta knowledge for the player.[/quote]
Then it can't affect gameplay unless you're metagaming.  The influence meter then ceases to be a problem at all.

[quote]It's information the player should have.[/quote]
No, it's information the player should not have.  The player needs to know how likely it is that he could succeed at hiding here.  He does not need to know whether he did succeed at hiding.

[quote]Why would you incoporate that decision into in-character?[/quote]
Why is it available at all?  It encourages metagaming.  It encourages the game's design to allow for and even require metagaming.  It's not information the characters would have, so the game shouldn't show it to the players.

Yes, the player could well ignore it, but that it's there produces some strange design outcomes.  This is much like cutscenes that show events taking place of which the PC isn't aware (NWN2 was terrible for this).  The player can ignore these scenes when making in-character decisions, but the game's design might assume that the player will not do that.

Intentionally handing the player metagame information as part of gameplay serves only to reinforce that the game is a game, rather than a roleplaying environment.

[quote]Bioware did the same with Wrex. This feeling is the feeling I have with dialogue in Bioware games sometime (but particularly with silent VO), where the PC is used as a foil to develo an NPC. [/quote]
I never really spoke to Wrex.  I found his armour on my own, and that ended all of his dialogue before it ever happened.  That was weird.

I don't mind the design you describe, most of the time, because I see the dialogue as an opportunity to express my character's personality, rather than develop the NPCs.  Sten was a problem because there weren't generally in-character responses available.  In fact, my first time through DAO I hardly spoke to Sten at all because I wasn't really allowed to, and thus I never learned anything about him or about his quest.  I didn't even kow Sten had a companion quest until playthrough number 3, where I was playing a far more earnest and personable character than is typical for me.  She got to know Sten well, and ever since that Sten has been my favourite DAO companion.

[quote]DA2 made me realize that VO in ME added certain features of dynamic interaction that were simply missing from previous games and those features I liked. Tone (which is really what the voice is) is incidental to that. [/quote]
This is partly why I distrust BioWare's market research data.  I don't think their respondents necessarily know what it is they like about a given feature.  That even you, one of the most thoughtful gamers I know, didn't realise it initially only makes that more likely.

I agree that the dynamic interactivity offered by ME is a positive.  But the paraphrasing basically made the system impossible to use, and the fixed tone dramatically reduced roleplaying options.

[quote]In terms of naivety? Certainly. The loss, though, is the ability to steer a conversation.[/quote]
An ability I don't think actually exists, so I'm okay with that.

[quote]There are converastions where you can avoid mentioned you are a Warden (Zathrian is a good example) and then, when you mention the Blight, he says you are a Warden. But you can never correct him. I don't like that. [/quote]
That's just sloppy writing.  DA2 does this throughout Act I: half of Kirkwall seems to "know" that I'm going into the Deep Roads with Bartrand, even though it's not true.  Hawke hasn't told Bartrand he's doing that; Hawke hasn't told Varric he's doing that; Hawke hasn't even decided himself that he's going to the Deep Roads with Bartrand.  And if he did, he'd want to keep it a secret and not tell anyone anyway.  And yet, somehow the whole city "knows" where Hawke is going, and Hawke is never allowed to correct them.

[quote]It was your advice, indirectly. After all, you make the very good point I did like those games in the past, so there must have been some features I did like. Playing them made the focus on the precise structural differences between new-age Bioware games and older games.[/quote]
Exactly.  People can't discuss these features in detail without first examining these features in detail.

Modifié par Sylvius the Mad, 26 mai 2011 - 06:43 .

[Sylvius the Mad]

[quote]Xewaka wrote...

How do you know the blinking button that appears for maybe a couple seconds means your character will grab and kiss his/her lover? That's the problem with action interrupts, they simply lack a way to inform the player of wether the action they propose is in-character with how the player wants its character to act. It's the same problem paraphrases have, but magnified.[/quote]
I remember when i first learned of the interrupts for ME2.  Somewhere on the ME boards, I'd just seen an ME2 developer admit that the ME paraphrases were poorly done, and needed to be improved for ME2.  And then, in the same thread, he told us about the interrupts.

And he didn't seem to understand why I completely lost my mind.  The interrupts undo all the good work at fixing the paraphrases (which didn't really get fixed - let's face it).  How could they think that both of those were good design ideas at the same time, when they accomplish diametrically opposed things?

I see now from talking to In Exile that the interrupts did actually offer a benefit that the regular dialogue didn't - interactivity - but they just did it so badly that I couldn't see it.
[quote]In Exile wrote...

But you revise your framework on the fly. Yes, the framework as it stands (if it remains unrevised) is useless; but you constantly tweak it for the person until it predicts them well (close to 100%); it remains only generally predictive in the general case.

This is just like a scientific theory. In its infancy, it can predict some pheonema with some regularity and a small but significant margin of error. As you add more work, you can refine the framework to work well for a particular phenoma (say, understanding the photoelectric effect). [/quote]
Okay, that makes sense.  I just don't want to use the model until it's vastly more accurate than 75%.  I'm more likely to use the parts of the model that are more accurate, and avoid interactions beyond those domains until I have more data.  For example, I'm confident that I can predict the behaviour of the cashier at Starbucks.  I don't know this person at all, but as long as I limit our interaction to me ordering a drink and her taking my money, the interaction works.

But beyond that, I'm just not going to talk to her unless I've had more opportunities to observe her behaviour.
[quote]All of this is dynamic. It is, to parallel a more mathematical concept, the difference between a connectionist network and an algorithm. It's just a structurally different way of approaching the same problem. [/quote]
That's an interesting way to look at it.  I generally think connectionism is just a way to construct the eventual algorithm, which you ultimately do need to draw a useful conclusion.
[quote]That's nothing more than saying that if the success rate isn't 100%, then it is equivalent to being 0%. And that's just a false dichotomy.

You need to appreciate that there is a continuum of possible knowledge.

It is true that you can be right all of the time. And it is true that you can be right none of the time. But can also be the case that you can be right some of the time. All of this from the same set of premises.

You discount the value in being right some of the time.[/quote]
I don't think this is about being right.  I think this is about being confident.

I don't think it's reasonable of the character to always have a high level of confidence when choosing what to say.  If he wants to elicit a specific response, how likely he is to know what will achieve that result is dependent upon his personality.

If I'm choosing dialogue options, the question is about how confident I am that this will produce the result I would like.  At each dialogue decision hub, my character wants to achieve a specific result.  So, among the various options, which of those would my character choose to produce that result?  Whether the option actually does lead to that result doesn't matter.  In fact, the way DA2 is written almost none of the dialogue options lead a desired result, because so few of the dialogue hubs are action hubs - most of them are personality hubs, and they don't elicit different responses.

If my character wants to intimidate someone, then I should choose the dialogue option that my character thinks will result in intimidation.  That allows a varying level of confidence in the outcome.  I should not just choose the option with the intimidate icon next to it, because that short-circuits the whole in-character aspect of the dialogue.
[quote]I once worked with someone who refused to give feedback. Ever. I couldn't understand it. I eventually confronted the person out of frustration, and she told me that her assumption was that if she wasn't complaining about my work, clearly I should assume it is good.

Which to an extent isn't unreasonable. But it was very difficult to appreciate my progress when I would show her data and she would only say ''Okay''.[/quote]
My problem is different from that.  He'll object to specific phrases, insisting on swapping them out for a different phrase, but next time he'll accept that same phrase.

I think the difference has to do with sentence structure, but I haven't been able to pin it down.
[quote]That's a judgement call, though. It depends on what you set as the objective standard for 'good' navigation.

I agree with you that the best method isn't neccesarily an adequate method. But the fundamental question is what a good method is.[/quote]
I think that's an in-character decision.  As such, the writers need to leave it to us to decide that.
[quote]I would say that being right most of the time is good, whereas you would say it isn't.[/quote]
A straight 51% majority?  And again, what if you can't tell when you'll be right?  Or what if those times when you're confident you'll be right are in fact the times when you're most often wrong?

I don't think it's as simple as you're describing it.
[quote]But consider this: what if you had to navigate at night? No other option?[/quote]
I'd employ the most risk-averse application of the best system I had.
[quote]This is what social interaction is like: we have to do it.[/quote]
Not as much as people seem to think.
[quote]So to return to the issue of gaming, the parallel is like this: 

Your characters need to be able to have a great deal of control over their behaviour.[/quote]
I need to have a great deal of control over my character's behaviour, yes.
[quote]The way you interat with the world is to always know perfectly what it is your doing; then you leave it up to the other person to make their move, similar to a game of chess. This is a passive strategy.[/quote]
It mimics the real world.  This is how conversation actually works.
[quote]My characters need to be able to functionally interact with the world. If I see that someone begins to misunderstand, or moves off my predicted path, I nudge them back. It is like trying to steer a boat in rapid current. Sometimes it is possible and sometimes it is not, but the fundamental approach does not change.

Do you see why, as a result, I value activity on the part of the PC in conversation?[/quote]
Certainly.  And as I think we've established above, we can both have what we want.  BioWare's just missing the mark with their experiments, right now.
[quote]The game scripts every possible interaction, in the sense that at the design level, there are a finite number of possible behaviours in conversations and the writers have exhausted them. There is no emergence.[/quote]
Granted.  But from an in-character perspective, the interactions still don't exist.  As such, every consequence is an emergent property.
[quote]To see a good example of emergence as a concept, google Cellular Automata.[/quote]
As an aside, have you seen Stephen Wolfram's speech on how he plans to formulate a unified field theory using cellular automata?
[quote]You agree with me that behaviour is causally determined, yes? That to act a particular way (even if it is unpredictable) there is a set state of physical affairs that led a person to acting that way?[/quote]
Yes.
[quote]A response is made up of (for the sake of simplicity): tone , wording , body language. Physical characterstics (like height, temperature, smell, etc.) also affect conversation, but these are impossible to ever include in a game so let's ignore them.

You think that for any tone or body language, a person could conceivably react in an identical fashion.[/quote]
Since the causal relationships that produce behaviour in others are not wholly known to us, we cannot say for certain that any reaction to any stimulus is necessarily impossible.  Especially since we're also not aware of what the sum total of the stimulus is.  The people aren't reacting just to us; they're reacting to us in combination wtih everything else they know about the world right now.  Sometimes that leads to people breaking down emotionally for reasons those around them don't understand at all, and certainly couldn't have predicted.  We're not privy to every aspect of these people's lives, so we can't know what they're going to do right now, or even if what they do right now has anything at all to do with what we just said them.
[quote]I am saying that this is impossible. You could imagine a player saying a line in an infinite number of ways: but the for the NPC to respond and remain the same person (even if you do not know what sort of person that is) the response must be consistent for that person. [/quote]
No.  The PC delivers a line in one specific way, and the NPC reacts to that.  The next time through the game, the PC might deliver that same line in a very different way, and the NPC reacts to that.  Within each instance of the game's reality, the behaviour makes sense.  There's no reason to think that the behaviour makes sense in both realities simultaneously, however, and the two realities don't exist simultaneously.

Whatever drives Isabela's reaction in one game might be different from what drives her reaction in another game.
 
You don't see it because you assume that from playthrough to playthrough the NPCs are always exactly the same people.  Not only is that an unnecessary assumption, but it's a metagame concern.  Within each playthrough, your character has no knowledge of or reason to suspect alternate realities.

In-character, Hawke says something and Isabela responds.  How Isabela responds tells Hawke something about Isabela.  If you change the details of the scenario, then what Hawke learns from the scenario also changes.