"Say 'yes' or roll the dice' is not about whether or not the GM "agrees" with something. It's about stakes, pacing, dramatic momentum.if the GM didn't agree with it there would be a roll to determine whether it did or did not take place.
If the GM "said 'yes'" every time, then the game would not really be a game at all. It would be something like a version of what Gygax derided as Monty Haul play - ie some sort of strange fantasy wish-fulfillment thing.
I'm pretty sure I quoted from p 72 of BW Gold upthread:
Unless there is something at stake in the story you have created, don’t bother with the dice. Keep moving, keep describing, keep roleplaying. But as soon as a character wants something that he doesn’t have, needs to know something he doesn’t know, covets something that someone else has, roll the dice.
Flip that around and it reveals a fundamental rule in Burning Wheel game play: When there is conflict, roll the dice. There is no social agreement for the resolution of conflict in this game. Roll the dice and let the obstacle system guide the outcome. Success or failure doesn’t really matter. So long as the intent of the task is clearly stated, the story is going somewhere.
Flip that around and it reveals a fundamental rule in Burning Wheel game play: When there is conflict, roll the dice. There is no social agreement for the resolution of conflict in this game. Roll the dice and let the obstacle system guide the outcome. Success or failure doesn’t really matter. So long as the intent of the task is clearly stated, the story is going somewhere.
Roll the dice when something is at stake; "say 'yes'" when nothing is at stake - the free roleplaying is colour, framing, etc; don't declare failure by way of fiat.
I think it's reasonably clear as a basic framework. The practical issue is having a good sense of what counts as something being at stake, or not - as I posted in post 164 upthread, to which you replied (and which I therefore assume you read):
I regard distinguishing between what is mere framing, and what is an outcome, as a very important domain of GM judgement. If you get it wrong, in either direction, then play will suffer.
For instance, suppose that - following the initial set up of the Cortex viking game - I ask the players, "So, what do you do?" rather than frame them into their trek to the north where they crest a ridge and see a steading, what is going to happen? The players will be confused - what was the point of all that set-up if we're not now going to cut to the action? I send mixed signals - I suggest that there is potentially something else of significance in the neighbourhood of their PCs that has no connection to the stuff we just spent 10 or 15 minutes working through. Why would I want to do that?
Conversely, if I treat not only the trek and the cresting of the valley as framing, but go further and tell them "So you enter the steading, and the action opens with you discussing matters at a feast with the giant chieftain", then there is the danger that I have mistaken an outcome for framing. For instance, one of the PCs in the game is a sneaky type who can influence animals and change into a wolf. By framing that PC into open negotiations with the giant, and prevent the player from expressing those aspects of his PC in the way that he actually did - namely, by sneaking into the Steading, finding a giant ox in the barn, and then trying to trade that ox for a favour from the giants (relying on the fact that giants are notoriously stupid and so won't recognise their own ox).
That's not to say that the boundary between what is framing and what is outcome is always clear-cut. There may be a zone of reasonable choices by the GM, and those judgement calls - in conjunction with the players' own concerns, motivations etc which are both elicited and responded to by the GM's framing - will influence what events unfold in the game.