Consider the case of "The Sword" example of play from BW. "I'm not into this, let's just skip this scene", resolves absolutely nothing.
OK, celebrim has already said he sees no real motivation to your BW Sword scene. Cancel it - what's the next scene, and let's see if we find it any more engaging.
I don't fully understand these posts.
"The Sword" is an introductory scenario for BW. The players look at the pregen PCs and hear the GM's description of the situation, which is pretty straightforward ("Having made your way through the dangerous ruins, you finally see it sitting on an altar through the doorway before you - the Sword!"). If the players aren't interested in any of this, then the scenario's not going to be run, is it - I guess we'll be doing something else this afternoon! Most of the time, though, if the group has agreed to playtest BW, then I would assume they'd have a go at it to see how it plays.
A lot less scenes will engage the player if the player dismisses the scene at the outset rather than making some effort to interact with it and see what it has to offer.
This really isn't my experience. It's not that hard, at least in my experience, to set up situations so that the players buy in from the get-go.
[MENTION=22779]Hussar[/MENTION] has made it clear in his case how this could have been done - Get the action to City B, which is where we're all super-keen to get to! Let us hire our spearcarriers so we can get back to the grell already! When the player goals are as clear is that, I don't find it hard to frame scenes around them.
Can a GM make mistakes? Sure. No one's perfect. But when mistakes are made - when a scene is framed that turns out not to enage the players (whether some, or all, of them), why keep pushing it? For some playstyles there may be answers to that question - for instance, maybe we're all committed to serious setting exploration, and even if the setting turns out to be boring in parts, we're committed to sticking to it. But for some, perhaps many playstyles, there's no reason to keep pushing a scene that's losing the players. Let it go.
All this talk about 'scene framing' and 'narrative versus simulationist play' is IMO really irrevelant and its moving goal posts.
I think it's higly relevant in at least one respect - it's relevant to showing that there are viable methods of play which (i) don't rely on GM force over plot and don't rely on exclusive GM force over content, and (ii) treat the framing of a scene that fails to engage as a mark of GM failure (to frame a decent scene) rather than player failure (to pick up on the hook that the GM has offered).
It's also not at all clear that saying players should have the power to frame scenes or that GM's should frame scenes that are interesting to the players in any fashion readily addresses the issues that this particular surprise raised - or any issues from any other case of "surprising the DM" brought up in this thread. The problem with using that as a blanket panacea for all table conflicts is that it doesn't explain how to deal with conflicts over player priorities and goals, conflicts between GM and player priorities and goals, or really anything else.
It doesn't, no. But it does make the point that there are some posts on this thread that make assumptions about GM force which are, from other points of view, quite untenable.
For instance, the idea that a GM might allows a particular combo once, but then veto it - exercising some sort of ad hoc fiat over the action resolution rules - is something I personally can't fit into my conception of how to run an RPG. If the action resolution rules, or some particular rules elements, are broken then fix them (or if they're too marginal to bother fixing, then quarantine them via gentlemen's agreement). If you want variety in combos, then look for a system that will produce that variety.
An article by Roger Musson in an early White Dwarf, discussing dungeon design, emphasises the need as GM to anticipate player creativity in placing treasure: don't put it in the dungeon if you don't want them to have it, no matter how well trapped and/or hidden. My own view is that this same principle applies to mechanics - eg if you don't want certain mosnters SoDed, then don't play with a game with SoD - don't just rely on a really good save bonus and take the chance that you won't roll a 1 (or fudge it if you do).
In which case, we must <snippage> ask the players to explain what the do want. The GM must then take this player input and begin narrating or at least judging a wholly new scanario. But even this misses the point, because if we do ask the players what they want there ends up being a space of negotiation or often lack of negotiation between the players over what the game should be. This can be every bit as fierce and problimatic as negotiation between a GM and player over the scene, because ultimately everyone is just playing the game together and everyone can't have exactly their way. In my experience, one of the main reasons that most RPGs have authoritative GM's is that this exists as a way to resolve conflicts about narrative scope.
<snip>
The DM has that authority because most of the players at the table don't really care to hear about negotiations over what the story is or should be about. They'd rather spend as little time in that sort of metagame as possible because it wastes time at the table that could be otherwise used doing things they want to do. In other words, it's not at all clear to me that 'cutting to the scenes the players are interested in', means in practice anything different than what is going on at a table by default or that discussing what scenes we want to have above and beyond IC choices, "do you turn left and seek out the goblins, or go right and search for the druid" really leads to more content per session.
BW expressly assumes that at the start of a campaign the players and GM will discuss what the game is to be about - with the GM as the leader of that discussion, but not the sole voice. (This is set out in the OP of the current "Is BW useful for D&D" thread.)
Whether or not many players want to approach RPGing this way, it seems fairly clear that some do.
But that sort of discussion is not a constant recurrence in scene-framed play, at least the sort that I run. The way that you get scenes that the players are invested in is to follow their cues. In BW this is particularly straightforward because they are core to the PC build mechanics. In my 4e game it is a bit more informal, but still - in my experience - pretty straightforward.
I also think that framing things as in-character or out-of-character choices does not always shed light. For instance, in my 4e game one of the PCs is a member of a player-invented elvish secret society. When the PCs were hanging out with some elves during (what MHRP would call) a transition scene, the player mentioned to me (as GM) that his PC was making the secret sign of the "Order of the Bat" to the elvish captain, hoping to learn that the captain was also a member of the secret society. Now in one sense this is an in-character action. But it's also a clear request by the player to bring part of his backstory to the foreground. So responding as GM, for me at least, isn't just a matter of narrating the setting. It's deciding how to respond to a clearly signalled player desire.
Up to that point I'd given no thought to these elves' relationship with the Order of the Bat, but I knew that as well as the elvish captain there was one other interesting NPC with the elves, namely, their crafter, whom this PC had already asked to fashion a tooth taken from a slain dragon into a Wyrmtooth dagger. Now I didn't want the captain to be a member of the society, because for various reasons (pacing, plus my own sense of what sorts of scenarios I can and can't run well) I wanted to downplay rather than intensify links between the PCs and these elven NPCs. But I did want to give the player something for his efforts. And I also knew that I wanted the crafter to come back into play down the track, because I wanted the dagger, when finished, to be delivered to the PC in question. So I told the player "The captain doesn't respond to your secret signal. But the crafter notices it, and signals you back."
That's a pretty basic example, but I think it shows how in-character choices can be pretty clear vehicles for expressing player preferences, and also shows how a GM can use certain techniques to make decision that are responsive to those player preferences whilst still preserving desired control (however much that might be) over broader aspects of the scene. One of those techniques is "Yes, but . . ." (in this case, "Yes, but it's not who you were hoping for - it's this other person who will fit better into the GM's agenda). Another of those techiques, and I think perhaps the most important one, is No Myth, or at least a degree of No Myth - treating the world and the backtory as provisional and fluid until actually brought out into play.