When I started my Joe Ballen series, I had no concept of writing a sequel as such. The book was meant to be a stand-alone, and it was only in editing that I realized I had more to tell about my smart-mouthed cab-driver/space engineer.
One thing I did have though, was a plan for how my fictional universe was going to evolve. Inspired by Robert Heinlein’s “Future History” idea (among others), early on I came up with an outline of development for my setting, starting at the point where Earth has just managed to build a working JumpDrive (as seen in the first Joe Ballen book), to people moving outward from Earth and becoming an interstellar species (nope, I’m not going to give you any spoilers! 😉 )
So, I quickly realized I would need some kind of “series bible” to keep everything straight. Initially, I kept all my character and setting information in the Scrivener project for my book, but anyone who has used Scriv to create a large project will know that there are problems with that approach.
So after the first of the Ballen novels, I separated out all the setting info into a new file that was basically just “Joe’s World.” (In fact, that’s what I call the file–how’s that for originality! 🙂 )
Now it contains details from the first thee Ballen novels, plus details on two of the new projects I’m working on–so five books, including characters, locations, details on theories and technologies, plus star system and planetary data. It’s a fair amount of information–roughly 277 MB currently, though not enough that you’d imagine it being a problem.
Except it is. I have no idea of the internal workings of Scrivener, but I suspect the file handling code is not the best or most efficient. This manifests itself, for example, in why it’s so hard to use Scrivener in a portable manner and why the default autosave settings are so problematic.
The upshot of this is that it now takes several minutes every time my Joe’s World project is saved, or closed and automatically backed up. While 270 MB is relatively large as a word processing file goes, it’s not that big by modern computing standards, and to my mind the process shouldn’t take that long.
The problem I have is how to tackle this issue. I have another tool, Treepad, that provides a similar tree/file arrangement to Scrivener. I believe it’s more efficient as it’s designed to handle large files. The difficulty is that I can’t test it without importing all the data from my Scriv project, and so far, I’ve not found an easy way of doing that.
An awkward conundrum at the moment. Is anyone else facing similar issues with their writing–or other–projects?