Dec 17

2017

A Taxonomy of Choices: Establishing Character

Posted by: Jason Stevan Hill | Comments (1)

As part of our support for the Choice of Games Contest for Interactive Novels, we will be posting an irregular series of blog posts discussing important design and writing criteria for games. We hope that these can both provide guidance for people participating in the Contest and also help people understand how we think about questions of game design and some best practices. These don’t modify the evaluation criteria for the Contest, and (except as noted) participants are not required to conform to our recommendations–but it’s probably a good idea to listen when judges tell you what they’re looking for.

If these topics interest you, be sure to sign up for our contest mailing list below! We’ll post more of our thoughts on game design leading up to the contest deadline on January 31, 2018.


There are a variety of different types of choices in a typical Choice of Games title. This is the first of a three-part series. Today, I’m going to discuss some basic types of choices: Fake Choices, Establishing Choices, and Forking Choices.

As a template, I’ll be working with a theoretical police procedural game. In the game, you’re a detective with a broken marriage and a corrupt partner, trying to take down the local mob. In this post, I’ll describe four types of choices that establish character, then conclude by talking about Forking Choices; in the next post I’ll discuss Testing Choices. (NB: if I refer to a variable, and it’s not immediately preceded by a + or -, I’ve prefixed $ to clarify that it is, in fact, a variable.)

The simplest type of choice in ChoiceScript is a Fake Choice. A Fake Choice is one that has no mechanical effect upon the game. (NB: A Fake Choice and a *fake_choice aren’t necessarily the same thing, though they can be.) Like an Establishing Choice (below), a well-done Fake Choice establishes character, but does so in the mind of the reader, not in the internal logic of the game (eg no primary or secondary variables are affected). A well-done Fake Choice can be just as suspenseful as a “real” choice.

A classic example would be where there are several dialogue options, and those dialogue options help to immerse the player, but don’t have effects on the game. For example, if the story asks the player what type of ice cream they want to order, to which the narrative always says, “Yum!”

Fake Choices are also used for comedic effect, like in the opening of Hollywood Visionary:

#I quit!
#I quit!
#I QUIT!

That said, if a Fake Choice fails to conceal itself as being a Fake Choice, something of the magic of our games is lost. Therefore, you want to use Fake Choices judiciously, so as to not run the risk of exposing them as being Fake; when players know in their bones that their decisions don’t have any effect on the game, they get frustrated and feel cheated. (Conversely, if the Fake Choice is used for comedic effect, then the magic isn’t lost; the choice from Hollywood Visionary isn’t trying to disguise the fact that it’s a Fake Choice. However, you still shouldn’t overuse them; the joke gets old quickly.)

A variant on the Fake Choice is the Flavor Choice. A Flavor Choice has no effect on Primary or Secondary Variables—on “winning” or “losing” the game—but instead customizes the narrative to the reader’s desires. Classic examples of this are choices like “what’s your name?” “what’s your gender?” or “what’s your sexual orientation?”

In Choice of the Dragon, whether you have leathery, scaly, or feathery wings has no effect upon the game. However, the game does actually record the value of the choice, so it can be used to reflect the choice back at the player at the end of the prologue.

In the case of our police procedural:

Every morning, the PC meets her partner at a diner near the police station. There's a new waitress today. She approaches your table, unaware that you always order:

*fake_choice
  #Coffee, black.
    *set favorite_drink “coffee”
  #Coffee, with cream and sugar.
    *set favorite_drink “coffee with cream”
  #Green tea, actually.
    *set favorite_drink “green tea”

Assuming you set $favorite_drink, and then recall the value of $favorite_drink later, this would be a Flavor Choice. Thus, Flavor Choices are remembered by the game—because they affect variables—but the variables affected either can’t or shouldn’t have any effects on the outcome of the game’s various plots (the End States).

(If you asked for a favorite drink and recorded its value, but then never recalled the variable later, it would functionally be a Fake Choice.)

Fake and False Choices aside, we move to describe an Establishing Choice that affects Primary Variables. In an opening scene of our police procedural, the PC manages, through no fault of her own, to apprehend the nephew of the local mob boss in the middle of a crime.

How did you take him down? 
*fake_choice
  #I talked him into putting the gun down.
    *set diplomacy %+20
  #When he ran for it, I chased him down.
    *set athletics %+20
  #I explained, in detail, the consequences of shooting a cop.
    *set intimidation %+20

Thus, regardless of the reader’s decision, the nephew is arrested; but the decision tells us what the PC is good at. It establishes the character and the qualities of the PC in the reader’s mind and in the mechanics of the game.

Another Establishing Choice might be something like:

Once you had the cuffs on him:
#I read him his rights. (+principles)
#I took the opportunity to vent some of my frustrations. (+bloodthirst)
#I searched his pockets before I sent him to central booking. (+greed)

The nephew goes to jail, and the reader has defined something about the contours of their character’s personality.

Establishing choices will happen throughout the game. However, they play a particularly important role in the first chapter, as the reader develops their character (commonly referred to as “character generation”) and learns about the rules of the world and the stakes of the game.

The next type of choice is an Objective Choice. This may seem like an Establishing Choice—in that there are no tests here—but the crucial difference is that Objective Choices affect Secondary Variables—variables that describe the world or the consequences of the player’s actions—instead of Primary ones. Because SVs are used to measure the PC’s success towards their Narrative Goals, Objective Choices give away the proverbial candy store without risk of cavities, and therefore should be used sparingly in Choice of Games titles.

In our detective’s story, she finds herself with a weekend to do with as she pleases.

#show up at my kid's ballgame. (+family_peace)
#pick up some overtime to help cover a sick coworker. (+cash, +police_rep, +career)
#shake down some corner dealers. (+cash, +stash, +street_rep)
#spend time documenting my partner's crimes. (+evidence_against_partner)

Here, we’re not affecting the PC’s skills (she’s not practicing her marksmanship, or working out at the gym, or learning how to be a better hacker). Instead, her decisions are directly affecting the Secondary Variables that determine the End States of the game: her work-life-family balance, her career prospects, creating a criminal case against her partner, and her street rep. However, one or two instances of an Objective Choice in the early midgame can help establish the different axes of success in the game, as an Objective Choice asks the player to choose between the game’s Narrative Goals.

If you want to use more than a couple of OCs in your game, the way to balance them out is to have negative SV-effects paired with the positive ones: so, “spend time at your kid’s ballgame” (+family_peace) might be paired with a -street_rep. However, you want to be judicious with such choices, as you don’t want players to specialize their way out of narrative tension.

A Forking Choice is a choice that forks the game in a substantive way. This is the historical basis of branching narrative: do you go left or do you go right? But in the context of variables, forking choices take on multiple dimensions and uses.

Because of their origin, they often have spatial overtones: do you go to the bar down by the docks, where the mob you’re hunting is known to hang out, or do you go to the cop bar next to the police station? But they don’t have to: Do you turn in the drugs you found at the stash house, or do you keep them? A Forking Choice is one where the reader chooses between mutually exclusive actions that significantly branch the narrative, either immediately or in a delayed fashion. Often, the results are recorded by the game in either a boolean or enumerated variable.

I say “substantively” and “significantly” because all choices that have unique subordinate text technically branch the narrative. But subordinate text isn’t substantive or significant; the game itself doesn’t remember subordinate text, for example, so it can’t affect the endgame. Similarly, it may be possible to go to two different “places,” but because of the clever use of variables, the encounter is functionally the same, just skinned to look different. For example, if the detective is questioning witnesses, and she and her partner each speak to different ones, but the two witnesses offer the same information and stat-effects to the player—they’re just “skinned” differently—I wouldn’t term that a Forking Choice.

But if the choice of which bar to visit results in different types of encounters—in one you can either intimidate criminals or establish a rapport with them, and the other you can either improve your relations with the police department as a whole or investigate your partner’s past misdeeds—that’s a Forking Choice; the two branches offer substantively different narrative moments and help the reader move towards different Narrative Goals. Similarly, if the PC decides to keep the drugs from a bust or dispose of them, two radically different futures are possible.

The important distinction above is that some Forking Choices have immediate consequences, while others have their consequences deferred. Deciding which bar to go to is an immediate consequences; keeping or turning in the drugs may have immediate consequences, but in a “Chekhov’s gun” way, it’s implied that the real consequences unfold later in the narrative.

The best time for Forking Choices with immediate effects is the endgame. There, the pursuit of wildly different storylines/outcomes can be pursued without concern for merging consequences back into the main story.

That said, Forking Choices in the early and midgame—usually with deferred consequences—are very powerful things, but it is because of their very power that we caution their use. For these early Forking Choices to be effective, the story has to reference/recall both branches of the fork later in the game. To be do this well, both branches have to be equally satisfying. Forking Choices in the early game with immediate consequences, such as Choice of Rebels’ “helot or aristocrat,” requires the author to maintain the consequences of that choice throughout the rest of the game, which is a daunting endeavor.

Designing a game where multiple choices and their multiple outcomes all successfully have satisfying recalls later in the game is very difficult, giving rise to combinatorial explosion. Choice of Robots is powerful for the way that early and midgame Forking Choices are recorded and then recalled throughout the game.

For example, in Robots, one of the important questions asked is “What kind of hard-drive will your robot have? Media-enhanced, multiblade, or encrypted?” Then, later in the game, there are specific moments where each of those hard-drives have a chance to shine. Similarly, you get to choose what kind of hands the robot has: clawlike grippers, T. Rex arms, Inspector Gadget arms, humanoid hands, gun arm, or multitool hands. Again, each one of those has particular moments where there is a narrative payoff. Those callbacks are very different from Rebels’ ongoing, omnipresent distinction between helot and aristocrat.

Notably, it’s important for these early and midgame Forking Choices to be perceptible to the player; Intentionality needs to be communicated. If the text offers the detective—after a long day on the job—the opportunity to take a shower at the precinct or to wait she gets home, and those two choices spin the game off in wildly different directions, the text has failed to communicate the presence of a Forking Choice. Something exciting and different might happen as a result of that choice, but if the reader doesn’t have some inkling of what else might have been, the value of the difference is lost. Because they’re not valued, we strongly recommend against them.

Finally, a Motivated Forking Choice is one that turns up when a choice is largely binary but the Choice of Games Style Guide demand at least three options for a choice. “Do you turn in the drugs to evidence or keep them for yourself?” That’s an example of a Forking Choice. But:

Do you...
#turn in the drugs to evidence. (+principles, $turned_in_drugs = true)
#keep them for myself; I might need to fabricate evidence later. (-principles, $turned_in_drugs = false)
#keep them; who knows when I’ll need a little pick-me-up. (+hedonism, $turned_in_drugs = false)

Whether or not the PC keeps the drugs is the fork, but the motivations behind keeping the drugs reflect and therefore affect personality Primary Variables, which will hopefully have deferred consequences in other contexts.

That’s it for today. Next time, we’ll outline the ChoiceScript Machine and the different types of Testing Choices.

Subscribe by E-mail