I’ll add more comments as time permits [these were written in August 2018, about four months before the game came out — but I didn’t know that at the time], and Ben will probably do the same on his blog at some point. Whatever other comments we add on about our work on the Cragne Manor project, I’ll link it back to this page.
We wrote up a design document to serve as a reference in writing the characters and their situation. It was particularly helpful in putting together a consistent time line. Initially we stuck closely to the design document, but as writing progressed, the story and characters took their own directions, and we ditched some of the design elements — there is no iron golem, for example in the final story. In some instances, we redacted portions that would either not have worked as IF or that were unnecessarily cumbersome in terms of mechanics relative to their narrative contribution.
While Ben started thinking about coding and integrating standard parts of the Cragne Manor project, I began writing the transcript. Perhaps not the best way to approach a project based on dialogues because it tends towards the linear, but given the time constraint versus volume of text needed to tell the story, we though it would be efficient because Ben could review and implement behind me as I wrote. That mostly worked for this project, particular because we kept the dialogue and NPCs relatively simple — not much in terms of forking dialogue or variability based on earlier knowledge, emotional state, etc. I worked within GoogleDocs and for the sake of loading quickly split the model transcript into part 1 and part 2. The final game resembled these transcripts pretty closely.
As usual, we wrote the code collaboratively, using version control to fold our efforts together, in this case the whole project lives on github.