1. Rethinking IF: The Problem and Promise of Parser Fiction
Before this class, I mostly played Japanese-style visual novels, so I assumed most IF took a similar form. Learning about parser fiction and other text-only IF showed me that very minimal interaction can still let players build images and spaces in their own minds, which I found fascinating.
When I actually played a few parser games, though, I didn’t enjoy them. They promise “freedom”: instead of picking from options, you can type whatever you want. But what the system understands is still a finite list of commands. Compared with the clarity of explicit choices, parser fiction forces players into constant trial and error, and most of that frustration and lost time comes from syntax, synonyms, and phrasing, not from the story itself. Having grown used to talking naturally with AI, I found it hard to accept such a rigid way of making a system understand me. Looking at the market, parser fiction remains niche while choice-based fiction stays popular, which suggests its interaction model has a real barrier.
That said, LLMs may offer this genre a way forward. If a model could interpret players’ natural-language input, parser fiction might keep its sense of openness while losing its most frustrating parts. For this project I chose the visual novel format because it felt more familiar and approachable, but in the future I’d like to experiment with combining parser fiction and LLMs.
2. The Power of Visuals: Consistency and Restraint
My own project showed me how much visuals matter in a visual novel. Even simple background images can greatly reduce the effort players need to understand the context, as long as they are consistent with the theme and content of the text.
To make visuals, writing, and music build a shared vibe rather than clash or overload the player, I used shaders to stylize real photos I had collected from various sources. This unified the art style and blurred away unnecessary detail. The images were only meant to set a mood, not to carry more visual information.
3. The Designer as Conductor
This made me think of game design as conducting an orchestra. Every musician has to be skilled, but the sections also have to sound harmonious together in each passage. Deciding when the experience should be calm or tense, and how to combine different players to create that shape, takes careful design and endless iteration. At its core, it is designing for the player’s feelings: a kind of design built on empathy.
During playtests, I broke down players’ reactions section by section. I noticed that the long, flat opening used to establish context could feel boring. Without weakening its setup function, I moved a small fragment of the climax to the beginning as a hook, giving players immediate interest and the patience to get through the slower start. Then, before that interest ran out, I added a hair-braiding minigame that changed the interaction and narrative pacing and gave players a genuinely exciting moment.
My story itself doesn’t have dramatic ups and downs, and player agency is limited. Through pacing techniques like these, I tried to compensate for those weaknesses from other angles and shape the player’s emotional curve.
(The Mood Curve on the graph is slightly exaggerated so that each point is clearly visible.)
4. Keeping the Author’s Voice
Finally, I believe keeping a personal style is important. Compared with genres like roguelikes or management sims, interactive fiction without fancy visuals is already niche, and like novels and films, it depends heavily on whether the author and audience are on the same wavelength. A distinctive writing style or atmosphere helps a work stand out. That’s why I didn’t cut every passage that might slow some readers down. They may not be strictly necessary, but they carry what I want to express and represent my style.
