October 18, 2019

Random Gibberish Design Thoughts

One of the prevailing strategies in interactive, cyoa-style narrative is the tracking of 'storylines' through a progress meter. The idea is to reduce the effect of the bottlenecked storytelling that cyoa-style is restricted to, which is a problem. As one makes choices that branch the main plot, other story variables take place to help add to the depth of choice, or at least the illusion of it.

As a very simple example: imagine when you were playing Shepard from Mass Effect. As you made more Paragon or Renegade choices, how your character looked and how others reacted to them changed over time. Also, what choices you could make.

So definitely more of this happening, except every variance should modify the whole world, not just appearance or reactions. How others change their behavior as they deal with you, say going from cold to warm. Or how the landscape is transformed by choices you've made in chapters prior. That sort of thing. I don't know if this makes sense. Sorry about that!

Back to the problem of bottlenecking. It's meant as a way to anchor the narrative so that the branching doesn't reach a point of massive incongruity. Imagine attempting to juggle 16 different plot lines across multiple chapters. It would be chaos, and it wouldn't be hard to get there at all.

But perhaps I can still deal with a couple of different branches throughout, each one branching out once or twice more up to the third arc. I can then start weaving them back together over the course of the last two arcs.

Er, apologies for not mentioning it earlier - I plan to write this story in five arcs. I have a pretty good idea of what each arc is going to touch on, but not exactly sure how many individual chapters it will take to say so. I imagine that the third arc will be the largest and most cumbersome. Each chapter will have a number of branching plot lines, which will eventually weave their way towards the three main endings.

Though it seems fairly rigid, the base plot line is really just the framework. I imagine that a good amount of the content within will be generated based on the aforementioned progress meter. I also want to experiment with procedural generation in this. Just a little bit for now.

Ok, gotta go for now. More soon.