Over a period of three years three authors wrote a complex book, The Joyful Sandbox, based on the experience of the first author, later turned editor, Thompson Morrison when he wrote The Dayton Experiment.
Our black-box print script struggled to make the right decisions in the right order. Here instead we run one tier at a time and the turn the results over to the author to review and refine before we ask automation to take the next step.
This is a sketch of what happened:
Authors agreed to write in wiki. New wikis were made for each author.
A narrative style was adopted. This was captured in a page of References to named and ordered sections of each story.
A personally meaningful story emerged for each author. Weekly review via video chat encouraged quality.
CODE: We wrote a story printer that would make pdf documents of the three stories to support the editing process.
CODE: We could reimport copyedited text from pdf but this was a variable success at best.
Stories were supported by what would become shared garden pages. These were forked between authors, sometimes refined, but not always.
Editorial review begins by plotting stories with gateway pages in the super collaborator. Three shared core concepts were identified. These guided further collective editorial decisions.
CODE: Did the gateway pages lead to more pages that meet the numerical expectation for beautiful pages?
A fourth wiki was created to represent the content that was judged book-ready.
CODE: Can we identify a book-sized subset of the garden? Can we indicate where these have the expected three links without leaving the book?
A process was developed for finding and revising outbound links and confirming the book was still complete.
CODE: stories and garden pages three deep were exported to the super collaborator so that for each gateway, the necessary context was easily identified in the beam highlighting.
With stories and garden in order then the book was extracted and run through more traditional review and printing.