November 13th, 2006
|11:44 am - Player Knowledge, Planning, and Intent|
So, I'd like to discuss the use of player intentions in resolution systems. This follows on a couple of threads: notably Fred Hicks' (drivingblind's) LJ post featuring his diagram of "Clarity & Relevance of Player Intent". This relates to a couple of threads: TonyLB's RPGnet thread, "[Craft] Why I won't take task resolution seriously", along with a couple of threads on Story Games: first, "Intentions in Task-Based Resolution", and then two follow-up threads which I attempted: "Intentions in *ANY* Resolution" and "Player Knowledge, Planning, and Intent".
First, let me clarify what I mean. I'm mostly fine with Fred's diagram, though I disagree on some of the conclusions about the nature of play. A slight nitpick is that true player intent can't be taken into account by mechanics -- only declared intent or perceived intent. Games only work based on declared game actions -- though note that I say "game actions" which is different than "character actions". The game actions can always be done with different intents. For example, I can take a game action in Polaris by saying "But only if his hand is chopped off". There are many reasons why I might declare that: I might be trying to tempt the other player into accept that, or trying to get him to drop, or even just trying to get a particular reaction regardless of whether the hand gets chopped or not.
For any success check system, there is a scope of what will happen from a single application of the mechanics (sometimes called the "Stakes"). If the full situation is known, then it may be that a number of steps are needed to apply the mechanics. For example, for breaking into a safe, first you must find the alarm, then disarm it, then circumvent the lock. However, there is a potential problem if the player doesn't know what steps need to be done to get what they as a player are after.
Here's an example. Suppose the players know that someone was shot in a Dogs in the Vineyard game. They want to catch the killer. However, they can't simply resolve finding the killer in a single conflict -- and in that case they may not have a definite number of steps to get what they want. As a parallel case, the PCs might be trying to find a secret entrance to the castle somewhere in the sewers -- but they don't know how to find that out.
The intent solution to this dilemma is that the player says what they're trying to eventually accomplish, and the GM guides them towards what steps they need to get there. As Brand Robins put in a late post on the thread:
The big thing here is that because the Player isn't letting the GM know what his end goal is the GM can't help steer him at it until the end -- assuming the GM would want to steer the player. The player never knows which roll will end up with him at the sewers, or near the sewers, or if he can even get to the sewers. So pretty much he keeps trying different things until he finds one that will work. Task by task, no clear intent.
While this works, it's not ideal to me. The problem is that before the example even starts, we have player ignorance. The GM knows what the PC would need to do to accomplish his goal, but the player doesn't. Both as GM and as player, I would prefer to give the player enough information to accomplish things before resolution begins. Then I don't have to steer the players, they can steer themselves.
As a small-scale example of this, if I'm going to play out a combat in detail, I will generally lay out a diagram of the area and describe it. Now, I could instead have my own vision and tell players what they would need to do to, say, surround an enemy. But I find it is more interesting if the players can come up with plans that I wouldn't have thought of... as opposed to me steering them to a plan based on what they tell me they want.
In general, I would say that if the player can't make informed decisions about what game actions she needs to take to accomplish what she wants, then resolution is at the wrong level. Either the player needs to be given more information such that she can make informed decisions, or the resolution should be done at a higher scale. For example, if the players don't have enough information to determine what they should do to find the sewer entrance, then I might simply have them make a single roll (say, Gather Information in D20) to find it. If I don't want it to be a single roll, then I have to provide them with enough information to make informed decisions.
All other things being equal, if there is only one player decision point, then there should be only one player roll. If the players have something they want, but don't know how to get it, then they should ask questions and I'll give them more information. However, I don't want to steer them with what they should do.
A particular pet peeve of mine is times when hiding intent is effective. In my experience, when informed of the player's intent, the GM often makes it hard for them to get what they want, in order to be challenging. This leads to the intent paradox that the players can accomplish some things more easily if they lie (or at least are deceptive) about what they're trying to do. In traditional games, this would often involve being deceptive about the PCs' plan. In some recent resolution systems, this moves onto a meta-level where it can be easier to effect something if it isn't the stakes.
"Here's an example. Suppose the players know that someone was shot in a Dogs in the Vineyard game. They want to catch the killer. However, they can't simply resolve finding the killer in a single conflict"
|Date:||November 14th, 2006 05:09 am (UTC)|| |
Since setting stakes is left wide open in Dogs, that's part of the premise rather than an absolute. i.e. We're assuming that the group didn't want to subsume all investigations of the town in a single conflict roll.
Technically, yes, it's possible for the Dogs to begin a conflict before they describe even riding to the town. They heard that the church burned down, and the first thing they say is: "I want to find out who burned it down." and start a conflict there, with riding into town to ask the Steward being the first raise. However, there are many groups who don't do this, and I'm assuming here that this is one of them.
I suppose that in general, the problem of intentions beyond a single roll is solvable by always rolling for everything that you are reaching for -- and thus the players not having goals beyond the next success check. I don't think I've seen this done in practice, though.
"We're assuming that the group didn't want to subsume all investigations of the town in a single conflict roll."
Oh, okay. Cool.
I thought you meant "can't" as in "it's not allowed".
I gotta admit that the dialogue following my posting of the chart has started to get over my head. :>
I'm gonna bookmark this one and try to reread it when the words don't get all swimmy on me (long day).
I think you're onto something w/r/t the whole declared/perceived stuff. I never meant that chart to be an end-state; I'm glad to see it evolve.