I think the problem with "RPGs on Rails" is pretty much when the PCs should reasonably have an opportunity to do something, but cannot -- not because of bad luck, bad planning, bad information, or any other reasonable in-universe reason, but because THAT WOULD GO AGAINST THE SCRIPT.
Now, the trouble here is that there might be a difference of opinion between GM and player. Players might try something, and it fails, and they think, "It's because GM is being heavy-handed and not letting us do our awesome thing!" while the GM thinks, "Sorry, that was actually kind of stupid, and the game universe shouldn't WORK that way." There, a bit of meta-communication might be required to make sure the GM and PCs have an idea what expectations are for how things work in this universe.
However, there are indeed games that are pretty scripted, and a GM may rather ham-handedly attempt to plug a "plot hole" to the point where the seams start showing in the story universe. Or, in some cases, the module will even explicitly instruct the GM to under NO CIRCUMSTANCES ever let the PCs come up with a clever way out of this predicament. (There was a classic Deadlands/Last Colony adventure that ticked me off by doing exactly this. And then there was Zombie Run.)
Now, running an open-ended sandbox game can be a real chore. I might well propose to the players as part of the price of entry that this adventure is going to have certain bounds, since I've got a day-job to worry about, and limited free time. E.g., this is the story of a group of heroes taking on The Grand Mega-Dungeon. The objective is to reach the bottom level and defeat the Uber-Boss. If the first thing the PCs do upon starting at the door is to turn 180 degrees and head off THATAWAY in search of random encounters, so all the GM's dungeon-planning is for naught, then they're being jerks. I'd just better get this established from the get-go, if it's a potential issue. But even then, there's still the potential that the PCs might overcome various obstacles in unique (and, alas, perhaps "anticlimactic," from the GM's point of view) ways, and he might have to make up a few things along the way to account for player actions nobody predicted -- but it's all still within the scope of the domain that the GM was (mostly!) prepared for.
If that doesn't work, then we have to negotiate until we find a scope the players are actually interested in. And if, as the story demands, the players have a perfectly good reason to leave this area to go somewhere else to do a side-quest ... well, if y'all go off the map, we might have to end the session here tonight, and give the GM some time to write up something for what's going to happen on the Side-Quest, rather than winging it the whole way.
no subject
Now, the trouble here is that there might be a difference of opinion between GM and player. Players might try something, and it fails, and they think, "It's because GM is being heavy-handed and not letting us do our awesome thing!" while the GM thinks, "Sorry, that was actually kind of stupid, and the game universe shouldn't WORK that way." There, a bit of meta-communication might be required to make sure the GM and PCs have an idea what expectations are for how things work in this universe.
However, there are indeed games that are pretty scripted, and a GM may rather ham-handedly attempt to plug a "plot hole" to the point where the seams start showing in the story universe. Or, in some cases, the module will even explicitly instruct the GM to under NO CIRCUMSTANCES ever let the PCs come up with a clever way out of this predicament. (There was a classic Deadlands/Last Colony adventure that ticked me off by doing exactly this. And then there was Zombie Run.)
Now, running an open-ended sandbox game can be a real chore. I might well propose to the players as part of the price of entry that this adventure is going to have certain bounds, since I've got a day-job to worry about, and limited free time. E.g., this is the story of a group of heroes taking on The Grand Mega-Dungeon. The objective is to reach the bottom level and defeat the Uber-Boss. If the first thing the PCs do upon starting at the door is to turn 180 degrees and head off THATAWAY in search of random encounters, so all the GM's dungeon-planning is for naught, then they're being jerks. I'd just better get this established from the get-go, if it's a potential issue. But even then, there's still the potential that the PCs might overcome various obstacles in unique (and, alas, perhaps "anticlimactic," from the GM's point of view) ways, and he might have to make up a few things along the way to account for player actions nobody predicted -- but it's all still within the scope of the domain that the GM was (mostly!) prepared for.
If that doesn't work, then we have to negotiate until we find a scope the players are actually interested in. And if, as the story demands, the players have a perfectly good reason to leave this area to go somewhere else to do a side-quest ... well, if y'all go off the map, we might have to end the session here tonight, and give the GM some time to write up something for what's going to happen on the Side-Quest, rather than winging it the whole way.