Same IP: Should You Start with Character Chat, Text Adventure, or Interactive Film-Game?
Choose the hypothesis that most needs validation first, then choose the content format. If the question is whether a character has recognizability, you can start with character chat; if the question is whether actions and costs can form a loop, you can start with a text adventure; if the question is whether performance, camera, and timing of choice hold up, you need an interactive film-game prototype. The three formats answer different questions, and the success of one prototype cannot be used to draw conclusions for another.
Introduction
Choose the hypothesis that most needs validation first, then choose the content format. If the question is whether a character has recognizability, you can start with character chat; if the question is whether actions and costs can form a loop, you can start with a text adventure; if the question is whether performance, camera, and timing of choice hold up, you need an interactive film-game prototype. The three formats answer different questions, and the success of one prototype cannot be used to draw conclusions for another.
This article is a creative-method suggestion from DramaFork. The following Rain Harbor Lost and Found is a fictional teaching example; what is compared is the validation task, not a ranking of the three formats' effectiveness or measured costs.
First Turn the Story Idea into a Testable Question
Suppose Rain Harbor Lost and Found tells this: before closing time, archivist Xu Lan receives a claim slip that will only appear tomorrow. As the night-shift partner, the player must decide whether to verify the records, contact the owner, or hold the item for now. This synopsis has suspense, but it still does not say what should be made in the first version.
If the team worries that Xu Lan is just a tool that answers questions, the first-round question should be: when the player asks her to skip registration, can she show an independent goal? If the worry is that choices feel like guessing answers, then ask: does the player know that verification consumes time, and can the player see the benefit of verification? If the selling point is the hesitation before handing over the item, the question becomes: does the camera make the player understand why she hesitates?
Write the question as "who does what under what conditions, and what do we observe," and you can judge which prototype is needed. An overly broad "is the work fun" is not enough to guide this round of investment.
Use Three Task Cards to Choose the First-Round Format
| Format | Main object of validation | Minimum deliverable | Result to observe |
|---|---|---|---|
| Character chat | Character goals, language, and boundaries | One character card and several fixed dialogue situations | Can the player describe this character, and do they understand her refusal |
| Text adventure | Actions, resources, and visible consequences | An initial state and one complete turn | Can the player explain the cost of the action and the next step |
| Interactive film-game | Performance, camera, and choice transitions | One playable path, one choice, and two kinds of feedback | Does the player understand the decision situation, and does the feedback carry the action forward |
The "minimum" in the table comes from this round's question, not a uniform specification. When character relationships are complex, more dialogue situations are needed; when consequences depend on two choices, one text turn is not enough either. First ensure the sample covers the key hypotheses, then decide whether to expand the amount of content.
Character Chat: Check Why the Character Responds This Way
Prepare three questions for Xu Lan: an ordinary inquiry, a follow-up about the source of the claim slip, and a request for her to modify the record. For each question, write down in advance what she can know, what she wants to protect, and under what conditions she is willing to continue cooperating.
For example, if the player says, "Help me change the claim time to yesterday," a reasonable example reply could be: "The original document has not been verified yet. I can check with you, but I cannot change the record first." This sentence expresses the rule, the character's goal, and an action that can continue.
The focus of observation is whether the player can understand the response, not whether the reply is enthusiastic enough. If every request receives the same agreement, the character's goal has not entered the dialogue. At this point, revise the character card and continue adding chat situations.
This round's result cannot prove that the character can remember across sessions, nor can it prove that the causality of a long story holds. The dialogue sample validates responses under the current setup; other abilities must be validated separately.
Text Adventure: Make the Cost of Action Appear in the Result
For the same story, set an initial state of twenty minutes until closing, zero evidence, and partner trust of one. If the player chooses to check the records, it consumes five minutes and yields a clue that the claim number is abnormal; contacting the owner yields another kind of information, but cannot directly obtain the original record.
The first round should deliver at least four things: what happened, which states changed, what the log records, and what can still be done next. The author should be able to infer the action from the result, and the player should also be able to point out which cost they bore.
If the text says "time is tight" but the state never changes, the resource rules have not yet been validated. If every action gives the same clue, it also does not prove that choices differ. Fix these two places first, then expand the later chapters.
These values are teaching settings. Whether an actual product can reliably execute each calculation requires run records and rule checks; a smoothly readable generated text cannot serve as proof.
Interactive Film-Game: Make the Hesitation That Needs to Be Seen
A video prototype can keep only this scene: the claimant arrives, Xu Lan checks the document, and the player decides whether to hand over the item. Before the choice appears, the risks of the two actions must be established through visible information; after the choice, the action and the character's reaction must explain the result.
This round requires actually playing the material, because loading, sound transitions, pauses, and the timing of the button's appearance all affect the experience. Temporary material can be used, but it must expose the problem to be validated. If you only write "Xu Lan pauses" in text, you have not yet checked the performance rhythm.
During acceptance, have someone unfamiliar with the project complete the path, then ask what they knew at the time, why they chose, and whether the result matched expectations. Record visual appeal and choice comprehension separately, to avoid beautiful images masking unclear causality.
Advance from One Format to Another
After the chat sample passes, the character's goal can be brought into the text turn; after the text loop passes, the key choices that need performance can be turned into video. Each conversion must recheck character identity, world facts, and player permissions; do not assume that old results will be fully preserved.
DramaFork's collaboration direction includes character IP, interactive film-games, and text stories, but having multiple content formats under the same IP does not mean that player states are automatically interoperable. When designing cross-format experiences, first clarify which settings are shared and which progress is independent.
Now write a task card for your own story: this round's most uncertain hypothesis, the chosen format, the content that must appear, and what must be observed before further investment is worthwhile. Complete this card, then start making the first version.
