A difference I see, with your sites method, is that, as a PC or a GM, I seldom see the in-game freedom to just check stuff out. I'm not talking about railroading, I'm talking about how once life starts rolling, you got things to get done, and there's no time to smell the roses (or explore extra dungeons).
We tend to find ourselves in the bind of "we got 3 days ride to ThereVille to stop the villain, and we just found a dungeon...hmmm...dungeon, villain...let's ride!"
How were the players introduced to the villain? How did they come to be pursuing the villain? How did they learn of the villain's plot?
Janx said:
It's very easy for a plot to guide the player's path where if they are vested in it, other choices don't matter.
How they become vested in the events of the game is one of the important characteristics of sandbox play, in my experience.
Janx said:
For example, the PCs have decided to ride to ThereVille to the village's aid. Within any encounter, there's tons of choice on how they handle it. Odds are good, whatever they choose, is something that works towards their goal.
However, in that same framework, they are NOT likely to choose to do anything that deviates from that goal. Sure, they'll stop for directions, help a beggar, buy a sword. But they won't do a u-turn and do something else, just for the heck of it.
From one angle, following a plot (an objective) cuts down on certain kinds of choices. What is really happening, is the players already made a major choice and barring a change, automatically discard any option that doesn't move toward the goal. This looks like a lack of choice, but it isn't.
Characters have goals and pursue them, putting off or aside other options, in sandbox play, too.
Choosing
how the adventurers will pursue their goals, from among multiple options, is not the same kind of player choice as
whether or not to engage with the situation in the first place.
Janx said:
Winding this back to RC and his sites method, I tend to see in game play (through the groups I started), that the party seldom has free time to just explore dungeons. Every dungeon is delved for a plot reason, rather than a "let's go find stuff" reason.
In a sandbox game, dungeons are explored for the characters' reasons, not the referee's.
Janx said:
Thus, it seems like the sites method doesn't come into usefulness for me.
The other day I was fiddling around with what some gamers would probably call a 'relationship map' of some of the significant NPCs in the
Flashing Blades campaign I'm working on. 'Relationship map,' in my estimation, is just a fancy term for a flowchart, so I was struck by the comment upthread that a dungeon is also an elaborate flowchart, because when I look at the boxes containing the names of the different NPCs and the lines showing the relationships between them, well, darn if it doesn't look like a lot like a dungeon.
From this perspective, characters are 'rooms,' some with one or two connecting 'passages' (relationships), others are major intersections with a half-dozen or more connections. There are 'secret passages' as well, relationships that are out of the public eye. Each 'room' may offer hazards and rewards, in the form of the goals and resources of the character.
The adventurers can explore the 'rooms' and 'passages' in the course of the game. The thing to remember is this: I'm not deciding for them what rooms and passages they will follow. They can visit as many or as few rooms as they want, engage as many or as few of the non-player characters as they choose. They may find that in the course of their explorations that challenges flow from some rooms across the 'dungeon' to find them: make an ally in one room, and you may find that gets you an enemy in two or three other rooms elsewhere.
So sandbox games don't require 'sites' to be physical localities, in my opinion.