Over at The Alexandrian, Jason Alexander is looking at some similar things to Vector Theory. I walked along a similar path for a while. My early game module designs were flowcharts with decision points indicating "combat here [if win, go to A; if lose, go to B]", and similar story elements. It was an interesting way to design a game because I thought I'd be able to come up with an answer to everything, how naive I was...
I'm prompted to write this because someone just made an insightful comment.
Interesting thought to keep in mind though... some players/player groups don't want a web, they actually want a line.
I've played with two groups before who, given choices (more than, say, two) of what type of adventure to pursue, even when it was connected to certain other elements of character or past adventure, etc. would actually feel like they didn't know what to do.
Eventually, they would follow a path just to follow it, but hated the idea that other paths existed.
Full comment and context here
I'm seeing the two ideas as different perspectives on the same picture.
Vector theory isn't saying that one method of design is "more right" than the other, I'm merely using it as an instrument to observe why games have been designed in that way...and what sorts of outcomes can be expected from such design decisions.
Let's look at the typical structure of a store bought module. This is also the traditional structure of most early games. Forget the mechanisms, die rolls and other details for the moment.
The GM defines a basic plot, then leads the players through the plot. Their decisions don't really matter, maybe a character dies along the way, but there will be time to add in a new one to ensure the GMs plot is fulfilled (or maybe the player simply sits out until this story reaches its conclusion, then joins in on the next story).
Each node along the path is not so much a narrative decision point, because the players can't alter the direction the story is going in. But each node could be an additive or subtractive filter (Additive: During the impetus, does the party collect the magic widget? Subtractive: During the complications, does the party suffer injuries?). There must be decision points in the lead-up to the climax, otherwise, this would not be a game, it wouldn't be interactive entertainment at all, it would just be a story leading to a conclusion.
It's a simple story, but it's the kind of story that a lot of people are comfortable with. Among others it's called railroading, because you can't deviate from a pre-ordained path.
Let's look at another option.
The branched path is traditionally considered the opposite of the simple path. But in effect it really isn't all that different. The Branched path offers meaningful story decisions for the players to engage in. They can choose to deviate their path to follow a new storyline, and their next set of choices will be based on the choices they have made so far. The example illustrated above may seem overly complicated, but it could be a lot worse. With three directional decisions from each node, there are potentially 3x3x3x3=81 possible endings.
...but the GM still writes them all up. The players still follow a story defined by someone else, even if theye are actually following one from a potential range of story outcomes.
This method of scenario design can be found in many store bought modules, but in these cases they often limit the choices to two possible options (to reduce a wasted page count filled with storyline that probably won't be followed). I've known GMs who've spent weeks writing up games in this way, only to be incredibly disappointed when their favourite path wasn't taken.
As "Morrisonmp" wrote in the comment quoted above, there are a lot of players who will be just as frustrated knowing that they haven't encountered certain experiences along the way.
Keep in mind that I'm not talking about dungeons or maps here. The story path is something very different. A dungeon or map allows a group of players to go back to a previous point and then explore new avenues if they didn't like the first one...a story path only moves in a single direction, from introduction through to climax and denouement.
So while the branched path may look like a good design strategy for offering player choice, I would certainly caution away from it.
What else is there?
This was my favourite method of scenario design for many years. You'll note that every node encountered still has the three possible outgoing paths, but now they link across one another. Instead of 81 possible endings, there are three different endings....far less work for the designer/GM, and far less missed opportunities for the players.
The catch with designing a scenario in this manner is making the choices meaningful, and making the choices accurately link up to the new nodes. By the time the complication stage of the story has been reached, each node has three possible entry points and a well designed game will ensure that each entry point makes sense given the choices that will be faced within the node.
I'd like to think this method of design combines the best elements of both the straight path and the branched path. The GM has a good idea of where the characters will be heading (specific character agendas might be pulling them to the left or right, specific die rolls might do the same thing), but players still have decisions to make.
The other advantage to designing games in this way is that specific nodes can be highlighted for their ability to change a narraton's wavelength structure.
For example: The leftmost node at the impetus level has a chance of providing a magic sword, the rightmost node in the complication level has a chance of providing a healing potion. As long as the characters follow the correct path through the story they'll pick up all the tools they need for their climax.
Another example: The leftmost node at the complication level provides a proton disruptor, while the rightmost node at the complication level provides a phase shield. Both of these will be useful at the climax, but the players will only ever get one OR the other.
Some would still call this a form of "railroading", because it forces the characters to a specific ending designed within the constraints of the scenario. But since there are valid choices to be made, this is where the "railroading" definition starts to become contested.
To be continued...