Story and Plot Design

- by demagogue

Dark Mod FMs, like the long tradition of Thief FMs before them (which we should all take notes from), allow for a much bigger role for story and plot than other first person genres, and most players expect a good story to come as part of their experience playing FMs. It's not really sufficient anymore to build a mission first and try to throw together some plot or story to fit it without putting a little thought into what makes for good storytelling in an FM first, and ideally designing the mission's architecture, objectives, gameplay, AI pathing, plot, and story all as a coherent whole from the beginning. One story that inspires me is that, as is well known, Terry Pratchett (the famous fantasy writer) was a long-time player of our FMs and would often post questions and comments (like all of us), and *he* said that *our* FMs were among the best story-telling going on in all video gamedom! High praise from a master, and something we should all strive to live up to!

The purpose of wiki entry is to give a mapper some good methods and things to think about when working out the story and plot-progression of their FM. It's about the mechanical side of how to construct a plot that works in terms of gameplay to tell a story, as opposed to giving story ideas per se. It will walk through a number of ideas mappers can think about when plotting their FMs, although of course since storytelling is a creative thought process, there is no universally "right" way to design a good story and plot. But hopefully some of these ideas can inspire your own thoughts on the issue.

NB, For the next few days or weeks, this is going to be work in progress and probably subject to serious revision and complete re-writing. So it's not really a trustworthy tutorial until it's done.

Introduction: Plot, Story, and "Making Progress" in FMs

"Plot", in the context of an FM, really means two things that are interconnected and work together:

(1) The gameplay plot, the progress and logic of the gameplay from the start of the mission to fulfilling all the objectives and winning the FM. E.g., sneaking past guards-A & B allows you to get into room-A, that gets you key-B, that opens door-B, which lets you into the room that has a readable telling you how to get access to the attic, etc ... until you finally reach the room to swipe the holy golden trinket of summoning, fulfilling objective-1; etc. The purpose of each step just has meaning insofar as it makes progress towards your ultimate goal of meeting your objectives; and

(2) The narrative plot, the progress and logic of the FMs storytelling from the briefing, to story details the player gathers as he explores the FM, to finally the "narrative climax" of the FM. E.g., if you had a FM with 10 rooms, various rooms will contain storytelling elements (readables, overheard conversations, visual-storytelling elements, etc), which the player will see or hear in the order he or she explores those rooms. The "narrative plot" would be the progress and logic of how those story-elements reach the player, e.g., the player reads readable-A before readable-B because the room with readable-B requires a key you found under readable-A, whereas readable-C is in an open room and could be read at any time, and altogether A, B, & C dole out the story to the player in an interesting way that carries the player in a "narrative flow" through the story, start to end, etc (this time A->B->C, another time A->C->B). The purpose of each storytelling element has meaning insofar as it contributes towards making progress in the (or a) narrative flow towards the ultimate goal of the narrative climax (which is the engine of the flow; the player can feel that "all this is going somewhere", and they feel a sense of gratification when they get there).

The two are interconnected, as one could guess from those descriptions, insofar as (1) the progress of the storytelling will flow primarily from the progress of the gameplay as the player explores the FM and opens up new areas [in contrast to normal fiction], and (2) in turn the "storytelling" directs gameplay by giving directions and cues to players what to do next or how to get past some gameplay barrier they face [affordance theory]. Both sides of the story-gameplay coin are driven ultimately by whatever gameplay events or narrative information gets the player closer to fulfilling the objectives/narrative-climax(es) of the FM. The two are connected at the hip and often there isn't a clear distinction between gameplay and narrative plot, and we capture both ideas in the shared concept of "making progress" in the mission closer to the objectives/story-fulfillment. [in contrast to story & gameplay that does not make progress. story/gameplay != narrative-plot/game-plot. The former are mechanics; the latter is the structure of the mechanics to create a flow for the player from start to objectives/climax.]

Spreading Story out in Time and Space & the Concept of "Narrative Space"

Now let's get very concrete. Both gameplay and narrative plot for an FM are ultimately about structuring the flow of plot-carrying information and physical "progress-ability" (like keys or hints) to the player in the space of the FM as he moves through it, understanding that sometimes the author may want very controlled releases of these according to a carefully planned script, and other times she may allow uncontrolled and random releases at the player's own pace, or sometimes some semi-controlled variation in between.

In this respect, it is very useful to think of your FM in terms of "rooms" (not necessarily always a literal room), with various rooms containing storytelling and progress-granting elements, or interaction opportunities that reward (correct) interaction with those elements. Then we can think about structuring the flow of information and progressability as ultimately about structuring the player's path through space and "time" (advancement to new rooms, or old rooms in new circumstances with plot developments) to access those elements. For the gameplay plot, this is very intuitive stuff we are familiar with from the FMs we have played. We understand the basic idea of scattering keys, open windows, climbable walls, hints in readables, and object-interaction opportunities to reward player exploration and engagement with the gameplay-intersecting plot with further physical progress in the FM space & time, so I don't think I need to spend much time talking about that.

The concept I want to talk more about here is how story-telling and making progress in the narrative-plot follows a similar idea. And that idea is, carrying a narrative-plot in an FM is largely about literally spreading the story, not in time like a movie or novel, but into the space and "time" of the FM (in its "rooms", or following plot-developing events) so that it is doled out to the player over time as they explore the FM space. Thus, to make progress in the physical space is to make progress into the narrative-space of the (narrative) plot.

Controlling the flow of Information and gameplay progress.

Unlike a movie where the filmmaker can control all information to the viewer, narrative plotting has graded levels of control of information-releases, which has pros and cons; some things the mapper can enforce (like you MUST read this book), some aspects the mapper can try to direct (putting things in earlier rooms makes it more likely they'll be read early on), and some aspects the mapper has little control over whether or how or when the player gets the info.

A Method for Story Writing and Plot Design

- start from the poles, the set-up (through the briefing and information in the first few rooms) and the climax. Then draw a line between the two and have nodes which connect the set-up to the climax. This is the plot-line. (Later we will talk about the fact that in games the line is not always a single linear line, but can be more like a spider web where you can travel across the nodes in different orders from start to finish.) You can think about each node is a plot-progressing event.

The number of nodes you draw depends on how long your FM is. For a small contest-sized FM, you'll want maybe 5-10 nodes. For a very large campaign-sized FM, you might want 20 or more. A basic rule of thumb I use is think about one node per major room or scene (like a building in a city) in the FM. So if your FM has 10 "rooms", you can think about 10 nodes, and then add or subtract a few. ...

Connecting Gameplay Flow and Story Flow through Plotting

Thinking about controlling plot flow is pretty intuitive, so the main task of this section is just to make it explicit. At its core, it involves the mapper literally looking down at their map and imagining the different routes players might arbitrarily take, and then parceling out information and plot-progressing elements in ways that make for an interesting plot flow as they go about their path, perhaps even structuring the map some to work with the plot development.

The idea of "plotting" in this situation means putting some structure to how these plot elements come to the player, while still (ideally) keeping the map rather non-linear and allowing the player some free reign to engage with the plot at their own pace and even in their own order and not being too heavy handed, but not too open either. There is a sweet spot here. Some players feel deflated if the map is way too closed at the start (like every door is locked and unpickable), but at the same time they can feel deflated if the map is way too open, like every door is openable and the player can just fly through the mission, not read anything, and skip straight to the end and beat the FM. The feel of "flow" in this sense is charting a middle course, keeping the mission just closed and challenging enough so the player has to really engage with the game to make progress, but not too closed or hard so they get frustrated, but not too open and easy they are never forced to really engage with the plot.

By "connecting gameplay flow and story flow", I am referring to the idea I mentioned before, that making progress in the gameplay plot also doles out progress in the narrative plot, so the mapper can, e.g., control the story plot by making sure readable-B is read after readable-A because the player has to get a key nearby readable-A to open up the room to read readable-B. This is a way to enforce that readable-B is not read before readable-A, allowing the mapper to "set up" B with A, so that e.g. B is a revelation or development on to A, and it develops a flow from A to B in the player's experience. Thinking about this in terms of the development of the whole FMs story from start to finish is what I mean by plotting, and using this mechanic as the main tool for plotting. Of course the development in B can't just be 'any' new information over A to be interesting or create a flow, but really has to be something interesting that really carries the story to a new place. (Something I'll talk more about below.) The term I'm going to use for how this mechanic works is "dependence relations".

Plot Dependence Relations

A "dependence relation" is a kind of technical term (from logic) that in this context refers to some pieces of story information requiring that previous pieces of information are accessed first before they are accessible. If your FM is set up so that readable-A should (or must) be read before readable-B can be read, then we can say readable-B is in a dependence relation with (or depends on) readable-A.

The typical tool of dependence relations in games is the "key" mechanic (not always a literal key), which just means some gizmo you find in one room allows you to get access to new rooms, thus the later rooms depend on the earlier ones. This is one way for the author to police or enforce the release of story information in a set order. But note that there are other ways to enforce or nudge the order of information as well, such as putting information closer to the starting point ensures that it's more likely the player will see it before information placed much farther away in the map, although that order is not literally "enforced" by a mechanic, so they're not strictly in a necessary dependence relation, just more likely than not.

Why it's useful to think in terms of dependence relations is that the author should really get in the habit of thinking about their bits of information getting released in any arbitrary order as the player arbitrarily wanders the map. If you watch FM playthroughs, you should know that every player is different. A dependence relation is just a way to say, whatever order the player goes in to get information, at least I know (with this key mechanic) that A will always precede B. That way B can refer to information given in A and the player will be able to follow it.

[Gradations of dependence and enforcement (or policing, guiding, cuing, nudging, etc.)] A necessary dependence relation would be "A must be read before B", like the room B is in doesn't open up unless A is read (either reading the readable triggers an objective & script that opens up the new area, or it literally gives the player a key, etc.) A strong but not necessary dependence relation would be "A should be read before B", which is like putting the key next to readable-A, so that the player could technically grab the key without reading A and run off to read B, but it's very unlikely. You can see how the strength of the dependence could vary according to the situation. If the key is on the other side of the room, it's more likely the player might miss the readable.

In any event, I think most (if not all) players are in the habit of reading any readable at the time that they find them (so in the order they find them). So that makes the author's job easier.

But dependence can still be relevant. Good FMs are not long linear hallways, but should be open spaces that allow the player to freely explore the space in any order they want. ...

I will get to what this means for plot flow in the next section.

Linear, Semi-Linear, and Non-Linear Plot Flows

As for the variety of structures of plot flows, I think the terms in the title alone are pretty intuitive. A linear plot flow is when the story is doled out to the player in a direct linear sequence, A, B, C, D... A very non linear plot flow would be something like a completely open urban space -- streets and buildings, with each building as equally enter-able as any other, and each building containing some plot information. Then the player can visit the buildings in any arbitrary order they please. A semi-linear plot flow would be technically open, but there are some built in biases so the player is still likely to go through certain blocs of the space/story in a certain order, but with some fuzziness on the edges too so a few things along the way they may see in a different order. E.g., a long hallway with rooms containing progressive story bits is semi-linear since the player is free to run to the last room first, but chances are they'll explore the rooms sequentially down the hallway, allowing the mapper to develop a little mini-story across the hallway as they go.

This last example also introduces the idea that plot flow can happen at multiple levels of scale, at a more global level in terms of large scale advancement in the map (entire regions) and much more local levels such as the path down a single hallway, or even a single room as a player looks around it (probably starting near the door and exploring progressively inward).

That's the image in terms of physical layout. In terms of the mapper drawing out the storyline, the image is more like the image I gave in the first section about writing the story in the first place, as a line stretching from a starting scenario to a finale with plot-developing nodes along the way. In terms of that image, a linear plot flow is a straight line with all the nodes lined up. A very non-linear plot flow is like a spider web where many of the nodes have lines going to all the other nodes, and the player starts on the outside and can choose any line they want to go down to go towards the center, but they have to "touch" all the nodes before the center opens up (or there can be layered nests like an onion, as progressively inner-layers in the web are opened up after the player "touches" all the nodes in the outer layer.) Then a semi-linear plot flow is like the spider web image, but some areas of the web locally have lines either cut or privileged, so the player is either directed or more likely to go down a certain path locally through that plot area, but other areas are more non-linear, or the player is free (on a more global level) to choose which local linear path they'll go down (more locally).

Thinking about these lines and plot-advancing nodes is also useful as the image for "gameplay" space generally, and the idea that the player "advances" in the plot as they advance through the gameplay space. In terms of linear plots (or locally linear plot threads), this is in terms of literally making progress along the line. In terms of the non-linear web image, this is in terms of the player making their way towards the web center as they meander their way around the web touching nodes.

I don't think a mapper always has to literally draw a web like that, but it's an image they can have in their mind looking at their map and gameplay design.

Modus Operandi

Modus Operandi, or MO, is an important concept that's been implicit in everything I've said above, but it's time to make it explicit. The short definition of an MO is the core motivation of a character that drives their behavior in a story, particularly in plot-advancing ways. An MO, or a clash of MOs or of an MO against nature, is usually the main engine of a plot. Probably in reality human motivations are fickle diffuse things that come and go, but in good storytelling MOs are more of a mechanic or device that pushes a plot forward, or a character through a plot, in a natural and compelling way that players can intuitively understand. So it's sometimes better to think of an MO in this mechanical way, as a plot device, than as *merely* human motivation per se, much less some random motivation like wanting to eat which may be very compelling in reality but not very good for carrying a plot or drama like an MO of survival or love or revenge can.

While an MO is important in any kind of storytelling, and the MO of NPCs in a game will work like the MOs of characters in books and movies (since game NPCs are scripted, except that NPC scripting can allow for a little more freedom than a book character), the MO of the player character in a game is very special because it's not just about the motivation of the player-character itself, it's also about the motivation of the literal player playing him. So special care has to be taken with constructing and engaging the MO of the PC.

Another thing special about a PC MO is that it's an "active" MO. That is, the MO of the player is usually connected to the goal or objective of the gameplay, so a PC MO in a game has a formal role in the mechanics of the game itself. That is, the mapper can create or modulate the player's MO through the objectives, and also the MO will push the player to behave in certain ways in the game that the mapper can (and should) expect and account for in their mission design. [...]

The most common and compelling MO for a Dark Mod FM is of course "the job". Of course, some jobs are more motivating than others. Stealing a random widget to make a little money is one thing, stealing [X to save the world] is more compelling. [That said, it can be off-putting to have an MO that's so fantastic it breaks immersion.]

Some other MOs include survival gameplay (probably the most compelling MO), helping NPCs, ...

It should be emphasized that the flow of a plot is (in my experience) proportional to the strength of the player's MO. The more compelling the player's MO in a game, the more the player will really be caught up in the compelling flow of events engaging with that MO.

[To give one example I read recently, the MO of the player in Halflife 1 was very compelling, you are trapped in a compound when all hell breaks loose and you need to survive and escape, and every game scene reinforces the desperation of the player in that MO. In contrast, the basic MO of Halflife 2 is to do some chores for some NPCs that until the very end do not seem to make much difference to what's happening in the world anyway, and definitely not at the local level of individual game scenes. So it's often been said that the MO for HL2 is a puddle compared to the flooding compulsion of the player's MO in HL1.]

Objectives are the main game mechanic that directs and makes an MO. [...]

A very important contribution to the MO is how the briefing and setup is designed. This is an important point because sometimes mappers drop the ball on this and unnaturally thrust the player into an improbable MO, which is a good way to alienate the player from the story altogether. The example I recall is the briefing for Equilibrium, otherwise an awesome FM in its gameplay, but storywise the mapper made the mistake of starting the player off with an entirely arbitrary MO to visit the Keeper compound for no good reason, and then as soon as the player entered it, a gate came down trapping the player inside for no good reason, giving him the completely arbitrary MO to escape a compound he never really had any good reason to enter in the first place. Great gameplay; terrible MO and storytelling.

Plot Stages: Beginnings, Middles, and Ends

Beginnings, Middles, and Ends, and the "Developments" That Move the Story Through Them

More Detail on Story-Telling Elements

Before getting to the actual plotting part, I think it's worth a brief look at the typical elements that go into FM storytelling.

- Readables

Journals, letters, books, etc, readables do storytelling obliquely. They don't usually narrate directly "X"; they are the writings of an NPC in the game-world thinking about "X" (unless that NPC himself is narrating something he saw). I think all of us have played enough FMs to know how readables work to get out the story and plot elements.

One thing I want to mention about them, though: A common criticism with FM readables are that they are "too long" or have irrelevant information. What I think this criticism really boils down to is exactly what I was talking about above. I don't think it's the length per se that's the issue. I think the feeling of "making progress" in the mission (or its lack) also applies when a player reads a readable. ...

- Conversations

Conversations happen when NPCs speak and act and the player is in proximity to listen and watch, usually either 2+ NPCs talking to each other, or an NPC and the player talking. The nice thing about conversations over readables is you can add some action, gestures, have the AI walk around a little or perform some action, and also it's a part of action live right in front of the player, so it's imminently relevant to the world & gameplay. The downside is the conversation can get interrupted or the player can wander off, so they're harder for the mapper to enforce.

- Visual Storytelling

Visual storytelling is any method of conveying narrative elements without using words, in effect, showing the player rather than telling. A classic example is finding an NPC body in a room. The player knows something dramatic happened that's probably relevant to the story he's wrapped up in.

- Mise-en-scène.

A mise-en-scène is a more formal term referring to how scenes are visually presented to help with the storytelling. The concept was developed in cinema -- where it largely refers to the layout and blocking of items in a scene, and how they are framed with certain angles and zooms of the camera -- but it has a natural application to game design as well, although the idea has to be adjusted to context. It is a little different from the way I was talking about visual-storytelling above, which was about certain visual elements that contribute to the story itself. The mise-en-scene is more about constructing visual presentation itself.

Since the mapper cannot literally control the camera in a game -- the player is in direct control of the camera -- in the context of gaming mise-en-scene is more about directing the eye of the player, and then framing the scene the player sees as they move and look around. A good example from an FM is the way Saturnine framed the mansion facade in Rose Cottage as the player entered the mansion grounds, making an awesome visual impact as the player entered, which he was able to do with a conscious and thoughtful design of the scene's geometry. Mise-en-scene fits with visual storytelling insofar as the mapper often wants to put some visual accent on important storytelling elements that not only directs the player's view to it, but also sends a visual cue to the player that this is something important to the story, not just an arbitrary set piece. Placing key items in the center of a room or scene, or lighting it in a certain way, or placing other objects around in ways that highlight it are all ways to put a visual accent on it. This is all part of thinking about a good mise-en-scene design for your FM.

- esp take advantage of sneaking gameplay (eavesdropping, events acted out around the player)

- Narrative Economy

Narrative Economy is about the ration between your map taxing the player's attention and communicating story information. To put it plainly, when you have something special that grabs the player's attention, if it carries some story information along with it, then your story will have a bigger impact, or signal-to-noise ratio. Conversely if you have a lot of random bits, then the story can get diffused and lost in the noise.

One classic rule of thumb is the Chekov's shotgun approach, which is if the author draws the player's attention to a shotgun on a mantle in an early scene, by the end of the story somebody should use the shotgun, or in someway it should be relevant. Another classic approach is to go scene by scene and cut the fat to carve out a harder-hitting storytelling punch left. Look at each readable or room and ask how hard it hits the player as they're going through. A really strong and compelling storytelling FM is going to pack hard punch after punch on the player room after room.

- Anticipation

Anticipation is about building expectations in the player's mind that "something" is coming, which is good both for building suspense for its own sake, and so that when that something finally does come it has a bigger impact. This is another good reason to have a Chekov's shotgun approach, to cue the player's mind about certain directions the story may go, which then allows the mapper to manipulate those expectations... The player can get an idea that "somebody is going to use that shotgun", but you can develop multiple potential-narrative threads that open up the possibility that this NPC might use it for this reason, and this other NPC for some other reason, and the uncertainty and puzzle of it gets the player engaged in the drama of it.

- Revelations

At its core, storytelling is mostly about the controlled (or uncontrolled, as the case may be) release of information in a structured way, where later pieces of information contribute to a narrative flow on top of the previous pieces in a directed way towards the goal or climax of the story, which lets the player feel like they're getting caught up in a dramatic flow and they can have a sense of making progress, and a sense of accomplishment when they arrive at the conclusion after a well-delivered build-up. (As we all recall from playing FMs with great stories, even if it's somewhat smoke & mirrors, it's a real feeling of fun when we're in a world we don't understand at first and we feel pieces start coming together in really interesting ways, and we want to push onwards to see what they're leading to; even if we can guess, we still feel that drive to discover more. Good plotting sparks that feeling.)

Essential to this picture are two things, (1) all the information pieces add up to an interesting and dramatic story when put together and (2) the player doesn't have access to all the information at the start, but they are revealed over time, and the new pieces must be important revelations or discoveries about the plot to really count as progressing the plot.

Now people sometimes complain that "plot twists" are cliche, or "I could see the end coming a mile away", and I think as a reaction that sometimes makes some mappers shrink from really writing in dramatic revelations into their plot and you get the opposite problem of a watery story where not much really happens at all.

- World or Character State Changes (or Apparent Changes)

The nice thing about storytelling in a game is that player interaction can actually change the state of the world they are exploring. Scripted events can be cued, but also more functional state changes can occur.

- Symbolic Interaction, McGuffins (and active McGuffins), and Rituals

Symbolic interaction is a storytelling technique where a plot concept of theme is reified into an object ("reify" means "to turn into an object"), then the object can stand in as a metaphor for that concept, and interaction on that object can represent some kind of engagement or confrontation of the player character with that concept. This is of course a kind of classic symbolism, but a game can take the symbolism the next step by adding meaning in the interaction with the symbol as a symbolic act itself. [...]

A McGuffin is one kind of use of symbolism. The classic definition of a McGuffin is the object that serves as the engine of a story or plot, while the object itself is inert, like the Maltese Falcon or the Ark in Raiders of the Lost Ark (except at the end where the Ark actually does affect the plot). It has a natural place in Dark Mod FMs because, of course, the engine most TDM FM plots is the task to steal some prized object, which object drives the plot but usually isn't much of an active part in it (because finding and taking it is usually the end of the story for most FMs.)

A ritual is another approach towards symbolism in FMs that is I think unique to gaming (in contrast to novels and movies). The idea is that often a mapper will have the player engage in some kind of sequence of actions (usually interactive with some objects or in space) that advances the story or serves as some kind of symbolic engagement with the story.

- NPCs

Like a lot of games, it's sometimes useful to distinguish functionary AI (like guards) who are mostly there for gameplay reasons, and NPC AI, who play a role in the FM's story, although of course an AI can play both roles or switch between them.

Specific Techniques and Story-Types FMs have Used

- Jobs & Events

- Reconstructing Past Events and Leading them to the Present

- NPC-Relationship Webs

- Private Spaces in Proximity: Storytelling in City & Mansion FMs

Common Problems and Things to Avoid in Plotting

- Does the plot make sense? Does it pass the sanity or straight-face test?

- Is your plot breakable? (And is that so bad?). Can specific game mechanics screw up the plot flow?

- Is there a good balance between free gameplay and directed plotting?

- One plotting problem I encountered in my FM. In order to get the player to see an important event in one room, the mission design led him away from the path that continued the mission, requiring them to backtrack to get back on the "main path" of the plot flow. In retrospect this was a design mistake. ...

- Be wary of game-y puzzles and gameplay (which isn't to say you can't have fun). Game-y is when you leave the goal of gameplay moving the plot forward and move into the territory of just having the player do crap because it sounds cool on paper.

- Avoid "premonition" puzzles, avoid situations that bust or kill the player without fair warning (especially for ghosters), or require lots of reloads

- Never leave your player stuck without a way to win.

- Cuing. Try to let the player always have an idea something to try next. Use cues. Consider graduated levels of hints (so a puzzle is found early on and open throughout the FM, early things give loose hints, as you open up more areas conventionally other rooms can give increasingly detailed hints). If the puzzle has multiple states, use the failure states to give hints about the success state. Something somewhere should be obvious the way through. It should be possible to complete an FM without having to post a question. Avoid cultural or language puzzles (or math puzzles). Remember that a large number of our players are not native English speakers. Be sensitive.

-