And both that morning equally lay
In leaves no step had trodden black.
Oh, I kept the first for another day!
Yet knowing how way leads on to way,
I doubted if I should ever come back;
–Robert Frost

In an interactive novel, the reader influences the story through choices. Upon replay, that reader can make different choices and explore a whole new path. Eat your heart out, Robert Frost! Let’s talk about choices in Harrowing Adventures as they relate to traditional gamebooks, digital interactive fiction, and what new challenges arise with multiple players.

Interactive Fiction

First, let’s talk about the philosophy of 4th Wall Games with regard to regular old interactive fiction and how our design choices have influenced the way we are making Harrowing Adventures. As Mark mentioned before, we’ve decided to make this story all about the characters and how those characters relate to the world. It’s written in third person, and you follow one protagonist and make decisions for that character.

This has the effect of clearly separating the player from the protagonist when it comes to making choices. It gives you, the player, a little additional psychological freedom to role-play a character other than yourself. You don’t have to feel tempted to keep making the same-old decisions replay after replay simply because that’s what YOU would do. Instead, you can re-imagine this character in any number of ways and make choices accordingly. The guys from Inkle once said in a podcast that they noticed a peculiarity when players of Sorcery! start a new game or press on after undoing a significant portion of the story to back out of a fail state: the players often seem to want to recreate a version of a previous save state. It’s like they’re trying to get back to the same places in the story they’d seen before rather than explore elsewhere. Maybe that’s what they’re doing, and maybe they’re just trying not to miss out on any bonuses they had before. On the other hand, we suspect that they might also be instinctually making decisions that are consistent with what they would personally want to do, and therefore often end up in similar circumstances when replaying the game. Can it be that making decisions for someone else allows you more flexibility in making choices?

There are some story structure benefits when using third-person as well. When you are reading a third-person interactive novel, the choices can easily become an extension of the story itself. Mark talked about this in his excellent and comprehensive article, but I’m going to just reiterate a few of the ideas here. In third person, there is a natural flow of the choices within the context of the story. For instance, the sentence of the choice can become the first sentence of the next section of the game, and therefore the story continues fluidly. You could even just scan the choices and ask yourself, “Which of these stories would I rather read?” 80 Days is a nice example of how this can work, even in first-person.

Mark and I have spent a good long time thinking about the criteria for what makes a good choice. First of all, we don’t want a player to feel like they’re making an arbitrary decision. Any choice should feel like it has advantages and disadvantages, and that you might have reasons to feel that one is a better option than the other given your particular goals and skills. How many gamebooks have paused the story to ask you if you wanted to go left or right with little additional context? Even more senseless are the books that require you to make a foolish or insane decision in order to access some part of the story that otherwise you’d never see. Mark and I want to make sure that there is never a choice that someone wouldn’t plausibly want to take. Hopefully, this helps remove the need for the meta-gaming where you say to yourself, “If they’re going to put such an obviously terrible choice in the game, there must be an ulterior reason for it. Maybe I should take it even though I really don’t want to.” If you have to make an obviously terrible decision in our game, it’s probably because your character has found himself in a dire situation with little hope for survival, and all the choices are desperate.

Digital Interactive Fiction

By using digital media, Harrowing Adventures is joining others in adapting and expanding the boundaries of what an interactive novel can be.

Moving to digital media opens up the possibility of having interesting new ways of making choices, such as moving around a map as in the Sorcery! series. We can also make randomized decisions based on chance or skill rolls without having to tip the player off that this is happening. We’re free to break from the strict narrative structure and throw in mini-games too. We can even introduce a puzzle that doesn’t need to have a numerical solution (the numerical solution then informs you of which entry to turn to, a commonly-used tactic in pencil-and-paper gamebooks). For our first installment of Harrowing Adventures, we’re going to keep the game a fairly pure narrative. You will have a character with stats that are tracked, and we’re going to have some RPG mechanics working behind the scenes as well, but we’re probably not going to include any sort of real mini-game in our first book. We have enough technological hurdles to overcome initially, without having to take time to make a game within the game. When we make a second book, we’ll see if a mini-game or two makes sense and improves gameplay.

Even though we’re sticking with more traditional interactive fiction gameplay, the digital nature of Harrowing Adventures opens up a lot when it comes to the choices. Consider the following: The main character is trying to convince another character that they are trustworthy enough to buy some black market item. The main character is going to have to delicately balance when they apply pressure, when they back off, and what they should say to try to gain confidence. For interactions like this, we can make the game follow probabilistic pathways based on our mechanics and your character’s skill at persuasion. They can build rapport, and finally a roll will determine if a choice has yielded success. One nice effect of this kind of behind-the-scenes skill roll is that it also adds to replayability. There is no longer a correct series of choices to make to achieve the desired result. When the character is walking down hallways and making turns, the deterministic approach works best, but if she’s having a conversation, things should be a little squishier. That was something hard to achieve with paper gamebooks.

We can also selectively offer choices to or hide them from the player based on the stats of their character or a history of decisions. For example, only a particularly observant character or one with experience riding horses would notice that the girth on their saddle has been loosened and be able to correct that before falling prey to sabotage on a ride. Being able to vary the choices on the fly like that is something we’re making extensive use of in Harrowing Adventures. You’ll experience that in the prologue.

Digital Multiplayer Interactive Fiction

The addition of multiplayer sends 4th Wall Games trekking into unknown territory. You may have played the old Multi-User Dungeons (MUDs). They are “multiplayer interactive fiction” in the sense that there are words on the screen and you can read them, but generally it’s the story of how you walked east 20 times in a row, south four times and scrolled past 264 lines of procedurally-generated text describing the strength of the blows exchanged with some monster (or mob, if you want to stick to the lingo). MUDs don’t really fit a narrative structure, and so you, as a player, are responsible for stitching together a narrative for yourself. We haven’t found anyone doing something like Harrowing Adventures. There are some other interesting multiplayer projects out there, but we have yet to find one that’s doing quite what we’re doing. Eon Altar was the last game that sounded like it might come close, but it turns out that it didn’t.

Harrowing Adventures consists of three proper digital interactive novels that are interwoven in a meaningful way. The decisions you make actually influence the story of the other characters, and therefore alter their choices. We’ve got a whole bag full of tricks and an in-house software tool helping us keep this from exploding into an unwritable mess of story. Interactive novels have always relied on tricks to keep the whole thing from spiraling out-of-control. If you find a good gamebook, you’ll notice after a few replays that there are story choke points and funnels that keep sending you back to a handful story paths (or killing you outright). Consider a single-player interactive novel with the following constraints: All entries are unique, all entries have the same number of choices C, and all story paths have the same length L in number of entries. If you were to employ no clever tricks, that would require a number of unique entries N of

For a game length of only ten passages, with three decisions in each, if you used no clever tricks, you’d have to write 29,524 unique passages, and you’d only get to make nine decisions in the whole game. You can see why nobody writes single player interactive fiction without some tricks. What happens if you throw in multiple players P and assume a choice by one player always affects the subsequent passages of the other players? The equation becomes

Extending the example above to multiplayer, in which three players each have three choices and there are ten levels of story, how many unique entries would have to be written? If you guessed 24 trillion entries, you’d be correct! It’s 23,756,669,087,844 to be exact. I guess Mark and I had better get writing! At least we know why nobody else ever thought this was a good idea before. But don’t worry. We’ve got some tricks up our sleeves, which will be the topic of future blog posts.

Other than requiring an enormous number of new entries, what else changes when you add a multiplayer aspect to a digital interactive novel? It turns out that most of the really big changes come as a product of the one control mechanism you have for directing gameplay: the choices.

First of all, you get a synchronization issue. The game picks up a turn-based nature out of necessity. If we’re all influencing each other’s stories, we need to be roughly in the same place in the story. I can’t be making choices for events that happen significantly later than what you’re reading. That synchronizing offers several challenges and even a few opportunities for enhancing our game. We’ve decided to tackle this by treating it as a game that is turn-based only when it needs to be. If your character and another player’s character are in the same room, your turns are short and come fast. But if you are on different sides of town, you are going to be able to make a lot of decisions without having to wait for me, because the effects of what you do and say on a smaller scale don’t directly or immediately influence the stories of the other players in that case.

Because the three stories need to stay synchronized, some choices will require you to wait after you make them. Because we’re dealing with a 1930’s serial adventure, we can use the waiting as a storytelling mechanic rather than an inconvenience. If we align the synchronization points carefully with dramatic moments in the game, two of the three players will end up waiting at a cliffhanger. Cliffhangers are a staple of the genre, and this affords us the opportunity to regularly use miniature cliffhangers, similar to what a reader experiences at the end chapters in a page-turning thriller. This builds tension and maintains player interest even if they’re waiting on a friend to get off work.

What about the drawbacks of making a decision and then not getting to read the outcome for some time? There’s a real possibility that by the time you learn what happened, you no longer feel that it’s connected to your decisions. To counter this, Mark and I decided to put in a little snippet of the result after any decision where you’re going to have to wait. That snippet will give you the direct results of the choice you’ve made, so even if your buddies aren’t able to play for another four hours, you’ll at least get a teaser to read that takes you to the cliffhanger.

Mark and I have a lot of other ideas for handling the emergent challenges of a multiplayer interactive novel, and we’ll no doubt write about them in the near future.

Questions? Comments? Miscellaneous personal abuse? We would love to hear your thoughts, so please leave a comment below.