Mark Sherry suggested a really interesting idea that would solve a small problem in Do. I had some apprehensions, for a few reasons.
Firstly, and probably most important, we're really late in the development process here. Ryan's already edited the first round of drafts and we are pushing towards and end-of-year deadline to get the text finalized. That isn't even counting how long the game has been informally developing over the years. (Long enough that I had deep concerns about Do being perceived as vaporware.)
Secondly, this solution would call for me to double a portion of the writing workload. My goal is to write one page for each step of play. That is a fairly straightforward model. Write a simple instruction for that step. Give an example of that step. Offer advice and inspiration for that step in the sidebars.
The only place where that model breaks is that one step in particular has a number of branching paths. The first branch is whether your pilgrim is in trouble or out of trouble at the time you take this step. The next branch depends on whether you keep zero, one, two or three stones. That's eight total paths for this step. No biggie. I will just write one page for each. Eight pages, four double-page spreads. Not a ridiculous amount of real estate in the book for what is probably the most critical phase of gameplay.
In short, here are the branching paths as they stand now:
Is your pilgrim in trouble? [Y/N]
How many stones did you keep? [0/1/2/3]
However, if I were to implement this cool idea, I'd add a third branch:
What color were those stones? [Black/White]
Thus doubling again the number of pages devoted to this one step. Sixteen pages, eight double-page spreads. Not necessarily a problem in itself, but there is a certain point where you have to draw the line when you're managing a project like this. You'll see my line tomorrow.