Hobo,
Can you tell us how your game is linear? That might be more helpful when trying to find out how your game differs from The Shaman's.
Fair enough. I'd say most of the linearity in my games tends to be front-loaded. Instead of having players create their characters, show up at the first session, and hope that they find enough common ground to adventure together, I start out with a premise. "You're all passengers on such and such a ship, bound for the port of Razina, hoping to find work." "You all work for the magistrate of the city of Terrassa, and while he gives you a free hand to pursue your mission of keeping the peace as you please, from time to time he hands you specific cases and tells you that they must be solved." "You're all guards on a caravan bound for Sinharat. You can come up with your own reasons why you want to go that direction". Something like that.
I also like to use the background method from Spirit of the Century that creates the barest hint of a backstory for each character, and more importantly, creates ties in the backstory for characters. They're not just complete strangers who happen to be standing around in the tavern at the same time when adventure calls; they're people who know each other and to some extent may have worked together in some capacity before.
For a session or two, I actually bring a game. I envision a specific problem that needs to be solved, and quickly. Pirates or bandits attack. You find out that the crown prince is on board incognito. Once the pirates or bandits are fought off, he invites them to the palace to be rewarded. Once in the palace, several court functionaries approach the PCs with problems that they believe the PCs might be their best agents to resolve. There are more problems for the PCs to solve than can really realistically be solved. Some of them might even be mutually exclusive.
This is where the linear function starts to dissolve and the players start to control the game. They pick which ones they think sound most interesting. They start figuring out how to tackle that problem. As they dive into it, I start actually making up the ramifications of the problem, usually based on what they decide to do rather than something I've got preconcieved. The example right there was one I ran not long ago. Another recent campaign, the prisoners woke up in prison, with no memory of the last three months. They find out they were accused of murder of a major crime lord. A few minutes later, a corrupt guard who's been bribed to look the other way lets them go. They have not idea what happened to them, or for that matter, what to do... so they start asking questions and digging around for answers.
Actually, come to think of it, that scenario isn't really all that linear, except in the set-up itself.
Notably, I like to imagine what might happen with the stuff that they don't pursue. So, the Queen asks them to go find out what's going on in her homeland where rumors of her father's kindgom being besieged by fey are filtering in by ship. The Crown Prince asks the PCs to check up on rumors of an army amassing in Cryx. The ambassador of Jekkara makes a lot of noise about invading Praetorians from Kurushat, the militaristic hobgoblin empire. The players decide that they can do the Crown Prince's work for him on the way to the Queen's homeland, so they attempt to kill two birds with one stone. However, the Praetorians of Kurushat don't stop just because the PCs aren't interested in them, so later when they come back, they found out that the King and his entire household have been slaughtered, the city is under martial law, and a puppet king, the Crown Prince, is on the throne. And that the army that was amassing in Cryx was actually at his behest in a hostile takeover bid.
Anyway, a few actual examples. When it's obvious to me that a campaign has run its course, or if I'm tired of it myself, I start pulling stuff together so that on a given session, all of the stuff that's been going on can get wrapped up, cut off, and satisfyingly concluded, rather than left dangling.
I'm not sure if that's actually an example of linearity so much as it is of me manipulating the pacing a bit, though.