Getting into this, here’s the current state of the game.
I’ve been writing and publishing novels for seven years now, and in that time I’ve managed to figure out a lot about how my process works, how fast I can write, how long it takes me to finish each stage of the novel-writing process – pre-writing, drafting, editing, revisions, etc. This helps me plan my production schedule, work up a marketing calendar, and to know approximately when I should start browsing about to find a cover artist.
Games are different. There’s no single formula to determine how long it’ll take to produce one, so before tackling an entire interactive novel in Twine, I decided to start with something novellete length that would let me work out a scale-able workflow adapted from my process for writing static fiction.
I do a lot of pre-writing as an author. I develop characters, I design settings, I beat out the entire story, I make complicated relationship and character arc maps. Interestingly enough it doesn’t take me much longer to pre-write a novel than a short story – the primary difference is how many story beats I hit and how many reversals there are. The actual writing itself is quick.
When I plot out the story beats, the basic unit of fiction I use is the scene. Each scene defines three things: What the protagonist wants, what’s in their way, and how things end, leading us to the next scene. Branstorming involves listing all the possible ways a scene can end and then picking the most interesting.
When adapting this to interactive fiction, I can select the few most interesting possible outcomes based on player choice, but the process is otherwise the same.
Look, a clue!
Murder by Clockwork being a mystery, I had an advantage in that an investigation does have a certain deliberate structure; there are clues, and when you find them, they lead you to new scenes. Our players’ freedom is provided by giving them multiple clues and letting them choose which to investigate in which order.
Visually we allow this with a map of London.
The story nodes manifest as physical locations our detective Wainwright can travel to, appearing and vanishing as clues lead us or dry up. The map is our first major departure from “bog standard Twine game” and it took a bit to figure out the implementation, but it feels much more satisfying than a textual hub would have been.
Choices within the Scene Structure
Each scene is written with an eye for choices the player might choose to make. As a preference I’ve tried to keep each individual passage short, ending in either 2-3 choices or a prompt to continue. For the most part I define these inter-scene choices as I write them rather than during pre-writing, simply because they get fairly tangled.
Where We’re At
- Pre-Writing: Done
- First Draft: 90% complete
- Line editing and proofreading
- Add more footnote popups
- Polish CSS look and feel
- Maybe scrape up some royalty free art