Conflict v. Task Resolution: An Outmoded Distinction?

[Sean]

I tend to think that it is lately. I'm happy to be smacked down though.

I think that they're just examples of different kinds of stakes-input into the resolution mechanics.

Case 1: The stakes are: "Does he get the safe open?"

Case 2: The stakes are: "Does he get the safe open and find the blackmail material?" (Or just: "does he find the blackmail material?")

Case 3: The stakes are: "Does something good or bad ultimately result from his attempt to open the safe?"

Case 3 is important, as some examples from discussions of this topic Sorcerer and Sword show quite clearly.

Case A: "Does my sword blow hit?"

Case B: "Do we win the fight with these clowns or not?"

Case C: "Does something good result from our having picked a fight with these clowns, or something bad, irrespective of whether we win or lose?"

I guess I'm mostly posting this to (a) ask the question: once we understand the concept of stakes in its full generality, in relation to the Lumpley Principle, is there any need to keep thinking that task v. conflict is important? Sometimes they overlap, we all know that. But what's left that's important about the non-overlap?

And (b) to suggest that even though cases 2 and B and 3 and C are often lumped together as both examples of conflict resolution, there's something very different going on in them in terms of credibility apportionment and the idea of what your resolution system is resolving both.

[GB Steve]

Are you asking the questions in a general sense or are you refering to a particular game or mode of play? I can see that I'd give different answers depending on this kind of information.

For example, gamists (and to a lesser extent sim players) tend to have a problem with large scale resolution, which conflict resolution tends to be,  because they can see it (wrongly in my view) as not taking full and proper account of tactical considerations.

In DitV, a conflict occurs when a GM says 'no'. That's the definition. Some games would represent that as a single skill roll (healing in GURPS) or a simple contest (healing in HQ if the GM thought the contest not so interesting) or multiple skill rolls (persuasion in Dying Earth) or an extended contest (HQ).

I think what's important is that games have consistent definitions about what the system resolves and the means to resolve it. And probably about what the system should not be used to resolve.

So in DitV the system resolves everything where the GM says 'no', everything else he says 'yes' to. In MLwM, everything is either violence, villany or a connection and must be resolved in one of these three ways (although there is a little indeterminancy as to what happens with "more thans"). As to whether these are tasks or conflicts, it doesn't matter. For players to enjoy these games it is a necessary condition that they sign up to these resolution methods. Is that what you mean in relation to the Lumpley principle?

[Andrew Morris]

Sean, I don't see any significant difference in your first three cases, and I'm not sure what you're trying to point out with the second three. As to task and conflict overlapping, I personally don't see that as happening, though I'm open to being convinced, so I don't think we can say "we all know that."

Steve, I just wanted to point out that scale has nothing to do with the definition of task/conflict resolution. You can have large-scale task resolution and small-scale conflict resolution. I agree that we usually see conflict resolution applied at a larger scale than task resolution, but it's not definitional, so I'd stay away from it for this sort of discussion.

[Valamir]

Outmoded?

That would have to be one of the quickest obsolescence lifecycles I've ever seen.  Games have only just started to regularly call on conflict resolution as a featured part of the design...we're just at the tip of the ice berg of seeing its applications...I don't see how it could be considered outmoded at all.

If you're suggesting that "in the end Conflict and Task resolution are pretty much the same thing so why label them seperately"...well, they're the same thing in the sense that something gets resolved, sure.  But what gets resolved is pretty fundamentally different.

The distinction may seem very subtle and hard to pin down in discussion...but in actual play...man...the actual play experience is night and day different.  For anyone whose gone from true old school task resolution to total conflict resolution, there's just no comparison.  Its a COMPLETELY different way of thinking about how to approach obstacles in the game.

Just as Director Stance is a completely different way of thinking about how a player relates to his character and the game world.

Just as shared narration is a completely different way of thinking about the distribution of credibility.

Its a pretty core fundamental difference...its not a "fad technique" that can become outmoded.


Now thinking in terms of MECHANICS that are purely Task Resolution or purely Conflict Resolution is a different matter.  Task vs. Conflict Resolution is a Technique of play that can be supported or hindered by, certain mechanics (and rules presentation thereof) but is not dependent or tied to them.

Can thinking in terms of this mechanic is a Task Resolution mechanic ant that mechanic is a Conflict Resolution mechanic become an outmoded way of parsing the issue.  Yeah...that I agree with.  It probably already is.  Certain mechanics will be more or less suited for one or the other but no mechanic is exclusively one or the other.

If that's the point you're trying to get to, then yeah, I agree.  But the difference between the two techniques of resolution.  No...definitely not.

[xenopulse]

I think that they're just examples of different kinds of stakes-input into the resolution mechanics.

You think that's a just? :)

Of course framing the stakes is part of the difference: It's whether the stakes reliably escalate or resolve a conflict.

Here's how my current AD&D 2e game goes:

GM: You enter a room. There's a safe, a desk, and some shelves.
I: I search the desk.
GM: Make a search roll.
a) [Failure] You don't find anything. (Don't know if anything's there)
b) [Success] You don't find anything. (There wasn't anything there)
-- Rinse and repeat for the shelves and the safe --
GM: What do you do now?
I: Uh, I was hoping to find dirt on this guy.
GM: Well, what's your character's next action?
I: ... is there any dirt to be found at all, or am I just wasting my time?
GM: Well, your character doesn't know that. So what does he do?

That's not necessarily how all task resolution play goes, but if we were using a conflict resolution system, it wouldn't even be an option to play like this. And you know what? The GM loves playing like this, and so does the other main player. Why? It fits their idea on how play should go. Characters act on insufficient information. They might waste time.

That's fine and appropriate for how those two want to play. Me, personally, it bores to death. I want my actions to matter. I want them to reliably address the issues at hand. I don't care whether it's "realistic" that my characters only do things that have that potential.

So, the difference between TR and CR might be revolving around stakes if you look at it that way, i.e., whether the stakes are tasks or conflicts, but that does in no way diminish the difference between the two.

[Sean]

Xenopulse, yeah, sure, I agree. Psychologically, there's a big difference. (This goes to Ralph's points too.) But either way you're just negotiating what kinds of 'moves' people get to make in the game and then resolving them.

One kind of move goes "My guy takes this action. What happens?"

Another kind of move goes "My guy and this guy are going to throw down and only one of us is going to leave the room standing. Let's figure out who it is."

Another kind of move goes "My guy and this guy are going to throw down and he's going to lose (say) but long-term I don't know if that's a good thing or a bad thing for my guy. Let's find out." (Maybe you win the gunfight automatically but there's a question whether the girl you're trying to get is going to see you as a cold-blooded killer or a brave hero. Another example: your long-term goal is to face down the wizard in his throne room. If the best way to face the wizard is to get taken prisoner by his guards then maybe winning the roll for the conflict with the guards means getting your clock cleaned in a tough fight followed by being thrown in a jail cell. This is what I mean with the third kind of example, when resolving the situation isn't about resolving that situation itself at all but about progress or regress relative to some larger-scope stakes. Does that help explain the difference between the three types I'm thinking of, Andrew?)

And of course there are all the moves that don't start with 'my guy' too.

There are a whole lot of RPGs in which only the first kind of move is allowed, or only explicitly acknowledged to be allowed, and there are some other RPGs in which only the second kind of move (say) is allowed.

But I don't see why this is a real, genuine decision, why it has to be either one or the other kind of move.

I suppose what I mean by that is that all any resolution system of the relevant type is adjudicating is what statements enter the game and how: what order statements enter the game or become candidates to, whether or not a statement is allowed to enter the game as-is or modified, what subsidiary statements enter the game automatically because of other ones that entered or failed to enter.  

Setting up stakes for conflicts is an explicit way of determining what statements will or will not enter the game depending on who wins.

So if you restrict stakes to 'do I open the safe' without any player input into what that means or its consequences for what the player's really interested in, leaving the latter up to GM fiat, well, I guess you can call that 'task resolution', and it's true that it's a default assumption of many RPGs that you're going to play that way, or at least that when you start to argue about what rolls mean they're going to be interpreted this way.

What I'm really resisting is this idea that there are Two Different Mutually Exclusive Ways To Resolve Things That Are Totally Disjoint and Completely Exhaustive of All Possibilities, which often seems to be at least implied by 'conflict v. task' arguments. Because I don't think that's right. It seems to me that there are a whole bunch of different ways for players to set stakes, and heck, if you wanted a game where you could roll to open the safe and leave it up to GM fiat what you found, OR roll to see whether you get the dirt with the safe-opening happening either way, OR agree that you're going to open the safe and get the dirt because you both want it that way, OR have the GM say "OK, fine, you open the safe, and you find..", and you say "no, wait, I want some adversity here, I don't think I should just 'get this for free' - maybe I don't want to know what's in that safe after all" OR agree to open the safe but then roll for whether your buddies will reject you as a criminal when they find out the next day well, it might be somewhat unfocused in its resolution mechanics, but there's no principled reason to think that you couldn't do both.

I mean, maybe there's an argument that for any specific game a more specific choice about how stakes are set for resolution is a good thing to have. I'm open to that.

So OK, fine. Maybe there's a family of games, call them 'traditional RPGs', that have a fairly narrow conception of what kinds of stakes it's OK for a player to set and when, and then another family of games, call them 'cutting edge indie RPGs', that have realized that there's no reason at all to restrict yourself to that assumption, and use wonderfully different stakes-setting principles for players to create different play-experiences. I'm all good with that.

I'm just really wary of saying that there are two different kinds of resolution on view here. The first family is broader than it looks in the caricature, the second family is way broader than that, and it's not clear to me that there aren't ways of cutting up how stakes and statements enter into the game that wind up making this distinction seem rather artificial from the theoretical standpoint.

[Bankuei]

Hi Sean,

I think a more worthwhile difference between the two styles of resolution would be to look at whether everyone at the table knows what the stakes are before the dice get rolled or not.  You can still have conflict resolution without player input into the stakes, as long as the players know up front what the stakes are before committing to action.

Consider, "I want to drive a little reckless and try to get there quickly" could have failure stakes ranging from:
-"You get a ticket"
-"You get in a tiny fenderbender"
-"You explode in a ball of flames"

Obviously, most players would not undertake an action if the last option was the stakes, and I've seen plenty games where the costs were higher than what the player anticipated.

Once you know what the stakes are, and when mechanically it's going to be resolved (a single roll, best of 5, when someone runs out of points/tokens, etc.), then you have Conflict resolution.  That's why D20's combat system is an example of conflict resolution, while the skill system is an example of task resolution.  It's also the reason you'll find that things that a player would find very important to play (such as a character 's being alive or not) get conflict resolution across the boards.  

I think the difference between the two is very important, and that most gamers have an subconscious understanding of that, hence the reason most combat systems and magic systems aren't based on wishy-washy Task Resolution.

Chris

[Andrew Morris]

Does that help explain the difference between the three types I'm thinking of, Andrew?

Not really, no. Or rather, yes, but the restatement seems entirely different than your original one, leaving me a bit confused. Going from "The stakes are: Does he get the safe open?" to "My guy takes this action. What happens?" seems like a huge change to me. In the first, you're just dealing with standard conflict resolution: I define stakes as this, then the resolution system determines whether or not that occurs. The second is task resolution: I use my (whatever) skill, do I suceed or fail? So it really seems like those are two different animals, rather than a clarification of the first one.

One of the key differences between task and conflict resolution is that task resolution determines the success or failure of an ability/skill/action/whatever, while conflict resolution determines the outcome directly, with variable stakes. It may or may not require the successful use af a particular skill. The concept of setting stakes is more important for conlict resolution than task, because in task, the stakes will always be "does my skill (or whatever) work or not?"

So I'm pretty much on the other side of the fence here, I think. I do see task and conflict as mutually exclusive, but if you can provide some good examples of where they overlap, I'd be interested.

[xenopulse]

Sean,

I must confess I am confused by what you're trying to say. On the one hand, you outline three distinct ways of approaching a situation. Then you say that it doesn't have to be one or the other. But they sure look different to me. Are you saying that you don't see a reason why any specific game should not support all of these options, i.e., that the only difference is the type of stake, but that all stakes should be open for negotiation with the GM in any game? You might think that is a trivial thing, but in this gaming culture, that's pretty radical.

I do think there is a pretty fundamental difference between choosing actions and then determining the consequences, or chosing conflicts and then figuring out which tasks got us to the result.

Think about trying to play a BattleStar Galactica game with D20 rules v. Primetime Adventures rules. It's Starbuck's spotlight episode.  There are 8 cylons out there, and it's just her and Hot Dog. Now, in the D20 rules, by virtue of the combat system, several of the stakes and how to resolve them are already determined. Are they going to get killed (thereby ending their story possibilities)? Will they kill all the Cylons? How much damage does each one cause? Unless you truly tweak the system against the way it was designed (i.e., drift), your choice of stakes is small. Now, using PtA, you get this stake as a possibility: Sure, they'll wipe out all the Cylons, but is Starbuck going to be shot down and crash land on a planet (thereby opening up more story possibilities)? For Hot Dog, you could ask: will he overcome his self-doubt (independent of how many Cylons he manages to take out)?

Notice that with PtA, if Hot Dog wins, the narration could have him shoot down many, few, or no Cylons at all, but he'll still have overcome his self-doubt. That's the reverse approach you could take with D20 where you say, if he shoots down 3 or more Cylons, I'll say he also overcame his self-doubt.

So it's not only about the stakes. It's about the direction from which you approach the story development.

[Andrew Morris]

Christian, to your whole post -- yup.

[Sean]

Xenopulse asked:

"Are you saying that you don't see a reason why any specific game should not support all of these options, i.e., that the only difference is the type of stake, but that all stakes should be open for negotiation with the GM in any game?"

No, I wouldn't say that.

I think there are lots of very good reasons for specific games to choose some ways of arbitrating stakes and ruling on the inclusion or non-inclusion of statements over others. One main one being, you can have a huge effect on the kind of play-experience people are likely to take away from your game by the way you handle this.

I just don't think the move from something called 'task resolution' to something called 'conflict resolution' really characterizes properly what's been accomplished by the new designs. Rather it's a move from an overly narrow sense of how stakes are set and statements introduced to a game to a more general one, which includes the previous, narrower one in its purview.

In other words I think task resolution is a special case of resolution-in-general, where you're only allowed to introduce very particular kinds of things: "my guy does x" type statements. I suppose if you want to say conflict resolution is this other special case where you're only allowed to conflict over "who gets the girl?" or "who wins the fight", maybe making it grainy like in Dogs so it's not all one roll like in some iterations of The Pool, OK, we can call that conflict resolution and say it's different.

But it seems to me that the REALLY IMPORTANT THING here is that we have this much broader sense of what kinds of stakes can be handled with rpg resolution mechanics and who gets credibility over what kinds of statements, etc. Which is why I tried to distinguish between 'local' and 'nonlocal' conflict resolution mechanics, and point out (not advocate!) the possibility of a resolution mechanic which you could specify either "does my guy open the safe?" or "does my guy get the dirt on him" as the stakes with equanimity. Maybe what's at stake in a different kind of game is setting elements.

[Emily Care]

So if you restrict stakes to 'do I open the safe' without any player input into what that means or its consequences for what the player's really interested in, leaving the latter up to GM fiat, well, I guess you can call that 'task resolution', and it's true that it's a default assumption of many RPGs that you're going to play that way, or at least that when you start to argue about what rolls mean they're going to be interpreted this way...

What I'm really resisting is this idea that there are Two Different Mutually Exclusive Ways To Resolve Things That Are Totally Disjoint and Completely Exhaustive of All Possibilities, which often seems to be at least implied by 'conflict v. task' arguments. Because I don't think that's right. It seems to me that there are a whole bunch of different ways for players to set stakes, and heck, if you wanted a game where you could roll to open the safe and leave it up to GM fiat what you found, OR roll to see whether you get the dirt with the safe-opening happening either way, OR agree that you're going to open the safe and get the dirt because you both want it that way, OR have the GM say "OK, fine, you open the safe, and you find..", and you say "no, wait, I want some adversity here, I don't think I should just 'get this for free' - maybe I don't want to know what's in that safe after all" OR agree to open the safe but then roll for whether your buddies will reject you as a criminal when they find out the next day well, it might be somewhat unfocused in its resolution mechanics, but there's no principled reason to think that you couldn't do both.

(bold emphases mine)

Hey Sean,

I hear very interesting things in what you are saying.  Let me paraphrase:

[*]You are saying that task and conflict resolution are distinct types of resolution. However, they are different but...
[*]They are two different ways of resolving outcomes that exist along a huge continuum of different ways to do so because...
[*]What they really are is two different sets of distributions of credibility (say), or two different types of rule sets that determine which participant gets to say what about the Shared Imaginary Text.  [/list:u]
Yes?  If so, then yes indeed, I would agree. Does that make the terms outmoded? Perhaps. But only if better, more precise terminology can be formulated that describes the phenomonen you're talking about, and that reduces the confusion that arises around task & conflict res.  

best,
Emily

[Andrew Morris]

Sean, Emily: How would defining resolution as a spectrum be more useful than the current split?

[Sean]

Hi Emily,

Yep. That's it, exactly.

The terms aren't necessarily outmoded in (what I'm arguing to be) their proper place. That is, there are resolution systems where the stakes are always "do I bring about the general result I want out of this conflict?" and systems where the stakes are always "does my guy successfully perform this action?", and I've got no beef whatsoever if we use the terms specifically to apply to those types of resolution system. But I also think that there are some systems that maybe blur the lines a little bit, and a lot of variety within 'conflict resolution' that could be better appreciated in terms of how the varying systems apportion credibility.

The best way to make progress with the question and the terminology both would be to design systems that cut across the lines of those assumptions, I think, at this point, but some serious compare-and-contrast of a wide swathe of existing designs might be useful as well.

[Sean]

I had a minor breakthrough on the kind of thing I'm talking about here, I think, with the discussion of 'reverse illusionism' in John Kim's thread:

http://www.indie-rpgs.com/viewtopic.php?p=167308#167308