Up to this point I’ve been fairly vague about exactly what I mean by a “node”. This is largely because there isn’t really a hard-and-fast definition of the term.
In generic terms, you can think of each node as a “point of interest”. It’s the place (either literally or metaphorically) where something interesting can happen and (in most cases) information about other interesting things can be found.
In my experience, nodes are most useful when they’re modular and self-contained. I think of each node as a tool that I can pick up and use to solve a problem. Sometimes the appropriate node is self-evident. (“The PCs are canvassing for information on recent gang activity. And I have a Gather Information table about recent gang activity. Done.”) Sometimes a choice of tool needs to be made. (“The PCs have pissed off Mr. Tyrell. Does he send a goon squad or an assassin?”) But when I look at an adventure, I tend to break it down into discrete, useful chunks.
Chunks that become too large or complex are generally more useful if broken into several smaller nodes. Chunks that are too small or fiddly are generally more useful if grouped together into larger nodes. The “sweet spot” is about identifying the most utilitarian middle-ground.
(To take an extreme example: “All the forestland in the Kingdom of Numbia” is probably too large for a single node. On the other hand, 86,213 separate nodes each labeled “a tree in the Forest of Arden” are almost certainly too fiddly. Is the appropriate node the “Forest of Arden”? Or is it twelve different nodes each depicting a different location in the Forest of Arden? I don’t know. It depends on how you’re using the Forest of Arden.)
Let’s get more specific. Here are the sorts of things I think of as “nodes”:
LOCATIONS: A place that the PCs can physically go. If you think of a clue as being anything that “tells you where to go next”, telling the PCs about a specific place that they’re supposed to go is the most literal interpretation of the concept. Once PCs arrive at the location, they’ll generally find more clues by searching the place.
PEOPLE: A specific individual that the PCs should pay attention to. It may be someone they’ve already met or it may be someone they’ll have to track down. PCs will generally get clues from people by either observing them or interrogating them.
ORGANIZATIONS: Organizations can often be thought of as a collection of locations and people (see Nodes Within Nodes, below), but it’s not unusual for a particular organization to come collectively within the PCs’ sights. Organizations can be both formal and informal; acknowledged and unacknowledged.
EVENTS: Something that happens at a specific time and (generally) a specific place. Although PCs will often be tasked with preventing a particular event from happening, when events are used as nodes (i.e., something from which clues can be gathered), it’s actually more typical for the PCs to actually attend the event. (On the other hand, learning about the plans for an event may lead the PCs to the location it’s supposed to be held; the organization responsible for holding it; or the people attending it.)
ACTIVITY: Something that the PCs are supposed to do. If the PCs are supposed to learn about a cult’s plan to perform a binding ritual, that’s an event. But if the PCs are supposed to perform a magical binding ritual, then that’s an activity. The clues pointing to an activity may tell the PCs exactly what they’re supposed to be doing; or they may tell the PCs that they need to do something; or both.
NODES WITHIN NODES
In other words, at its most basic level a node is a person, a place, or a thing.
As suggested above, however, nodes can actually be fairly complex in their own right. For example, the entire Temple of Elemental Evil (with hundreds of keyed locations) could be thought of as a single node: Clues from the village of Hommlet and the surrounding countryside lead the PCs there, and then they’re free to explore that node/dungeon in any way that they wish.
Similarly, once the PCs start looking at the Tyrell Corporation they might become aware of CEO James Tyrell, the corporate headquarters, their shipping facility, the server farm they rent, and the annual Christmas party being thrown at Tyrell’s house — all of which can be thought of as “sub-nodes”. Whether all of these “sub-nodes” are immediately apparent to anyone looking at the Tyrell Corporation or if they have to be discovered through their own sub-network of clues is largely a question of design.
In short, you can have nodes within nodes. You can plan your campaign at a macro-level (Tyrell Corporation, Project MK-ALTER, the Chicago Sub-City, and the Kronos Detective Agency), look at how those macro-nodes relate to each other, and then develop each node as a separate node-based structure in its own right. Spread a few clues leading to other macro-nodes within each network of sub-nodes and you can achieve highly complex intrigues from simple, easy-to-use building blocks.
A friend recently turned me on to your site, and I’ve been reading through these. As a GM, I’m finding them useful, and they really make clear a lot of realizations I’ve had, especially recently, about scenario design. Thanks a lot for that!
I’m also noticing now, that where I often find scenario writing to be a real chore, and a bit stressful, I’m really looking forward to writing some now. I hope that feeling lasts. 🙂
I am just starting out as a DM, and I found this article a real inspiration for writing my first adventure.
With this approach to structure it seems you’ll be able to avoid a lot of potential frustration both for the DM and the players. Thanks a lot for taking the time to explain your reasoning in such detail, it was really helpful.
I have no idea if you are reading questions on your 2010 posts, but here is one:
Is it wise to have a node that is “blocked” by a moral choice? E.g. one of the clues from A points to “break into the sheriffs office to obtain information”.
The PCs might not want to do that for moral reasons.
So, would this be an “extra” clue/node then? E.g. Node A has 3 clues pointing to “morally okay” nodes B, C,D and a fourth one pointing towards E, which is the illegal/immoral option? Or would it still need only 3 clues and the player´s choice to not pursue this lead is simply a limiting of options?
Very good question! And one I hadn’t explicitly considered before.
There might also be other, non-moral reasons to avoid certain avenues of investigation: You find clues indicating that there’s cult activity around a warehouse on N Parrish St in Baltimore, but when you go to check it out you discover it’s basically an armed compound. Sine you’d rather not tackle security guards with SMGs strapped to their sides, you decide to continue your investigation somewhere else. (I had a group who literally went to a different continent rather than deal with a particularly well-armed compound.)
I think the more specific answer to your question is how much pressure you want to put on that moral choice.
If it’s pretty easy to route your investigation around the immoral choice, then that choice will probably be fairly easy to make. It’s like saying, “Do you take the free lollipop from the sample table or steal the lollipop from the small child?”
The fewer paths there are around the immoral choice, the more likely it is that the PCs will be put in a position where their only choice for continuing the investigation is the immoral choice. (Because they’ve missed or misinterpreted the other clues that were available to them.)
And even with node-based design, you can force the moral choice: Use a funnel-shaped design with all of your clues ultimately pointing to the moral choice node, and they’ll need to choose between continuing their investigation or doing the immoral thing.
With that right group, that can be a really meaningful dilemma. And, if that’s the case, you need to be prepared for them to make the choice to NOT pursue the investigation. (And by prepared I don’t mean “find another way for the investigation to continue”; I mean, be prepared for the investigation to fail. If you’re going to present the players with really powerful, meaningful choices, you don’t want to immediately strip the consequences away and turn them into meaningless choices.)
You can also create dilemmas that aren’t all-or-nothing: The moral choice node, for example, might just be the logistically easier path. Breaking into the sheriff’s office after hours is easy, but ethically questionable; tackling the armed cultist compound with SMGs is going to be really tough, but less ethically fraught (at least for some groups). Now it’s not a choice between continuing the investigation or doing the immoral thing; it’s a choice between doing the easy, immoral thing or the really difficult, potentially lethal thing.
And that can be an even more interesting choice.
I’m really glad you addressed the “be prepared for the investigation to fail” aspect.
Thanks for the really quick answer!
I really don´t want to put too much pressure on the (tougher) moral choices, it´s more on the side of “I want the players to at least think about it” and “I want them to have different viable options” side.
And also, I do not want to railroad them too much, that´s where my question originally came from. Thinking about this overnight, I think I identified where my problem comes from:
I often think about these things with a preference in mind, a bad habit. So if I had clue A (the dangerous option), clue B (the immoral option) and clue C (the GM-preferred option) I feel like I am railroading in a passive-aggressive way.
The point with the heavily guarded warehouse makes sense of course, I should have thought of that. “It´s too dangerous!” came about occasionally in my previous campaign.
Perhaps it´s clear from my disjointed questions that this is my first try at going “node-based”. I´m still in the scenario-prep phase, so I have no concrete examples yet, but the help is greatly appreciated!
Is it sometimes useful to conceptualize important objects as nodes in some scenarios, or is it adequately covered by all of the other categories?
I can see how objects could be subsumed into other categories either as clues, as ways for NPCs and organizations to conduct events, as the independent causes of events, as window dressing at locations, or as props for PCs to perform activities. Would a MacGuffin best be conceived of as a transferable quality of some node?
Less likely, but are qualities too simple and abstract to be useful as nodes most of the time?
It seems like most of the ground covered by qualities as nodes will be taken by the descriptions, clues, revelations, and so on of the already typified nodes and that qualities would merit no independent existence in the preparation. Plus they can be added or subtracted easily in play owing to their simplicity, so it doesn’t seem like there would be much use in spending valuable prep-time on them rather than determining them on the fly at run-time.
@Questwriter, if you look at all the examples Justin has published in the last decade of his actual prep, usually his nodes are based around “when the PCs get here, a scene will take place.” Sometimes it’s a chain of scenes, sometimes the scenes are conditional, but the scenes are key — ultimately your game is built out of scenes. So nodes are usually locations because scenes take place in locations, and it’s common for characters to have set locations where they’re found, and so on. But it’s also possible for a node to be designed around a character (when you find them, or they find you, this scene takes place), or an event (at 3am, wherever they are, the PCs hear the bomb go off). The point is to have a framework for gathering the notes you, the GM, are likely to need at one time, on one page, to make it easier to run a scene.
So can an object be a node? If finding the macguffin is going to trigger a scene, then maybe. (Though in most cases an object is either hidden in a set location or carried by a set NPC, so it still might make more sense to use one of those as your node. The idea is to do what makes your prep easy to manage.)
Can a “quality” be a node? I’m not sure what that even means.