Creating Lilt — Part 2

Now that I’ve been working on this game for a month or so, it’s time to share some of my progress. I knew that writing the story for this game was going to be a challenge, but I had not realized that I’d first have to overcome a equally challenging prerequisite — figuring out how to write this story.

Not to downplay the challenges of writing a good novel — but at the most basic level, books are a long string of carefully curated words. There are tons of tools out there for organizing, formatting, and correcting these words, but not much out there for what I was planning to do. My story would be 3D, 4D, or even RealD when compared to the flatness of a book.

So, without further ado,

Attempt #1:

I’m clumping together a few similar attempts I tried here: lists. I had lists in Google Docs, lists in Clear, lists in a notebook… but none of these seemed to suffice. It became obvious that for a multi-dimensional story like Lilt, I’d need a multi-dimensional tool.

Attempt #2:

Lists in lists.

I had the above list hosted on my site so I could view it as I updated the html. This allowed me to map out the branches a player could take as they made their way through the game. I also attempted to use Markdown, which was a little better, but both ended up being to0 unwieldy and neither could scale easily. I discovered over and over again how much I had underestimated the amount of content I’d have to generate for even a small, simple puzzle in this game.

Attempt #3

For this attempt, I decided I needed to start defining the space of the first level of this game, and then see if a good system came to light through that.

The Room

To kick off this process, I had a friend over and quickly threw together an illustration of a room in Google Docs, then we started playing through some ideas. We compiled a list of all the objects in this room, and generated a short list of possible interactions. We also came up with the general path a player would need to follow to advance.

It was this brainstorming session that lead me to the surprising solution that I’ve been running with so far… a spreadsheet.

This allowed me to account for a vast majority of object/interaction combinations with relative ease. The top row also includes an entire section at the end for, “use X with Y,” for instance, “use X with ants,” which would then be followed by a column of responses for “use blanket with ants,” “use bucket with ants,” “use coin with ants,” and so on.

Next Steps

Part 3 of this series will be coming soon, as I’ve started working on a prototype, and it’s already playable. Hit me up on Twitter, @mknepprath, if you’d like to give it a shot!