I’m back
My github activity shows I’m out of my December/January productivity slump :)
One outline to rule them all
In February I combined all of my small single-arc plot tables into one massive table spanning the entire book. While this was challenging, I now find the story exploding in my mind, a non-stop chain reaction of ideas. I can see the book in its entirety at a high level, and I can dig down to subtle details.
I gave myself a deadline: finish this outline by end of month, and start drafting again in March. While the outline is still a living document that will evolve, I’m happy to say I met the deadline successfully.
Single massive table pros and cons
With one massive table, I can see the book in its entirety. If I sort by scene number, I can see what happens in any given chapter and in adjacent chapters. I can see the pacing: this POV slows and breathes while another grows in tension. I can add reveals in one POV that add context and complexity to another POV’s arc. I can create cross-POV arcs that start in one POV and complicate and/or resolve in others.
I had held off creating a single massive table because it presents low signal-to-noise when I’m trying to follow a single POV’s story, a single side character, or a single arc. I might not notice if I leave a side quest unresolved, if a side character disappears for large swaths of the book, if my pacing is rushed or draggy, or if I resolve a mystery without giving the appropriate clues in the right order.
(In addition, I worried about screen size and my editor bogging down. I resolved these last two by artificially limiting the length of my descriptions and switching to neovim on my laptop. I still use my iPad a lot, but I use Shellfish to ssh to my laptop over Tailscale.)
parse_beats.py
I wrote a script that lets me parse a massive markdown table outline and easily filter and reorder:
-
I can sort by book order, Chapter 1 before Chapter 2 before Chapter 3, etc. This is my default. I can output this as a single massive table, or a table per scene so I can easily visualize what’s happening in each scene.
-
I can split by POV, so all the POV1 scenes in order, then POV2 scenes in order, etc.
-
I can also split by arc, so all the beats in the Murder Mystery arc, then all the beats in the Way Home arc, and make sure I’m revealing the appropriate clues and progress markers for each of those arcs to resolve properly.
This gives me all the benefits of multiple small tables and all the benefits of a single massive table, through tooling.
An explosion of arcs
This led to wanting to track everything. Not just the major arcs, but also side characters and subplots. As of this writing, I have 3 main POV characters (plus one chapter in a 4th), and 890 beats over 51 arcs. Most of these 51 arcs have 7 plot points.
This does seem like overkill, and I may need to do some massive trimming to get down to 120k words. But some of these beats just let me keep track of things: at this point in the story, this other character has gotten this far in their side quest off-screen. I may be able to recap 3 or 7 of these beats in a single paragraph. Other beats will take an entire chapter.
Many of these arcs are sub-plots. For example, if I track the murder weapon beats as part of a huge Murder Mystery arc, I may not notice if the weapon goes missing and then no one thinks about it again for the rest of the book. By tracking the Bloody Knife arc as its own arc, I can make sure I resolve that subplot in a satisfying way.
Breakthrough ideas
Even though all of this seems like busywork, on Tuesday I had an idea that so vastly improved one of the POV’s main arcs that I don’t know how the story would work without it anymore. I don’t think I would have had that breakthrough if I hadn’t been looking at the story as a whole at a high level.
Plans for March and beyond
I’m going to get back to my draft. I’m calling this my Draft 2.5, since I’m starting my 2nd draft all over.
With 60 chapters and 98 scenes outlined, I can tackle a chapter a day and finish by end of April, or I can tackle a scene per day and finish in early June. The outline is going to be a living document that lets me see the ramifications and side effects of any changes I want to make along the way.
Stats
The word count is a little bit cheat-y because I have 4 copies of the main outline. (The main outline is sorted by scene number in a single table. Then I split it by POV in a povs.md, by arc in arcs.md, and by scene in scenes.md. I have a shell script to generate each of these files from the main one.)
… These stats have been pretty meaningless for a while now, and I may be rewriting manuscript files for the next month or so. But sometime in April they may become meaningful again :)