On the Development of Bottles (Pt.1 - Intro and Pre-Production)
Bottles was my first project as a director of a true team— it was the first time I spearheaded the manifestation of an idea on a deeper scale, it was the first time I had to juggle critique against creative vision, and the first time I truly learned what being a director means to me, and where it means I fit in the context of the creative current of game development. The daily, weekly, monthly, and yearly process is what has stuck with me through the duration of working on Bottles. What does it mean to ideate a project in an early stage, show it to a team, and all push towards the goal of bringing it to life? What does it mean to see something from the highest of the high concept to the smallest of the minute details and keep your sights set on every increment of progress between? If you had asked me those questions back in the early Spring of 2019 I may have had an answer; but it wouldn’t have been an informed answer backed by experience.
So I hope this post-mortem can be useful to someone, anyone who is looking to step into the process of creation and may learn from my mistakes, or be inspired by hard-won little victories in the greater machine of making.
What is Bottles?
Before I get ahead of myself, I believe it would be best to introduce the project in question.
Bottles is a mobile game that fuses the tactile, handheld feel of analog puzzles and the delight of mobile indie magic. You play as a small character inside a bottled diorama world, navigating yourself through lush, delightful spaces filled with challenges and stumbling upon charming surprises behind every obstacle.
The idea for Bottles arrived in two parts— one, the central mechanic of rotating a bottle around a character, and two, the intention to make a mobile game that could be played with one hand.
Origin of the Core Concept
First, the mechanic. This was one of those rare, precious ideas that came in a very finite, very distinct moment that can be pinpointed later as “this is the exact time when the idea occurred.” The moment was in the middle of a session with my therapist in February 2019. I had brought a bottle of soda with me that day and had finished it off over the course of talking, and being a fidget-prone person I ended up playing with it as I continued. Staring at the droplets in the bottle collecting and running along the inside of the plastic was, in a strange little way, mesmerizing. Absently I thought: “this would be a fun game”. It was as simple as that. Never before has a game concept come into my head so politely, and I don’t count on it ever happening in the same way ever again.
Second, the greater game concept. Though I already knew what I wanted to base my general gameplay around, the mechanic of rotating levels with the player inside could be taken in myriad ways. Would it be a big, action-y console title? Would it be something you could manipulate in VR, holding the level in your hands? What did I want to make? When it came down to it, I thought about what games meant a lot to me in a practical, down-to-earth sort of way. This mechanic wasn’t fancy or convoluted; it was simple, and the game should follow suit. So what simple, engaging games meant something to me, both as a player and as a person? I thought of the games I used to play when I got my blood drawn.
In the Summer of 2011, I was diagnosed with Thyroid Cancer. I was 19 and terrified, and even though my prognosis was overwhelmingly non-lethal the spectre of sickness and hypothetical death was enough to shake the foundations of my teenage self-assurance. I went under the knife two weeks after my diagnosis and came out of it dazed, in pain, and without a thyroid. I was required to have my blood drawn every three hours for a day after surgery, and even past that I would have to get it drawn every week, every few weeks, every month, every year— further and further out into the future ad infinitum. It wasn’t great for someone with a needle phobia.
The thing about being medically required to do something is that you are medically required to do it. I had to get my blood drawn, and it was down to me to figure out how to make that situation livable. What I settled into was this: I’d sit in the chair, rattle off my name and birthday, and then take out my phone and hold it in one hand on the opposite side from wherever they were going to poke me with needles. I’d open up a game. Usually it was a Bejeweled clone, sometimes Monument Valley; whichever it was, the game proved intellectually engaging enough that I’d be distracted while the nurses took vial after vial for blood work testing, and made it so I didn’t associate the process with the visceral embodied horror of a needle piercing my soft, fleshy, vulnerable arm. Instead I could relax into a pattern of lines and moves, a puzzle to be solved, or a world to be navigated.
So if I had this mechanic— this rotational level idea— I wanted to create something with, what better way to honor that experience than to contribute a title of my own to the genre of games that helped me through a trying time?
So it was: Bottles would be a mobile puzzle game.
The game went through at least two conceptual iterations before the first prototype was ever made. First was the pitch doc version. In this, the bottle was segmented into sections that each rotated individually, which had channels cut into them that the player would navigate like a 3D rounded hedge maze. In order to complete a level, the player would start at one side, walk through those pathways, and have to move them into alignment in specific patterns to reach the other side of the bottle.
This version was quickly scrapped in favor of something more involved, less directly mechanical-feeling. I wanted to evoke emotion as well as engagement and felt that this dry, procedural version of the idea didn’t lend itself to that. Furthermore, as I talked about this idea with peers and other designers, I discovered that people continually brought up the idea of “messages in a bottle” being the iconic image associated with the bottle as an object. That was a compelling creative direction I wanted to pursue more intently.
One thing worth noting is that Bottles was at least a gyroscopic game from the get-go. The slices of the bottle would be turned by rotating the phone toward or away from the player, which allowed for single-handed play more fluidly than if the controls had been entirely touch-based.
The second iteration that happened before the first prototype leaned a lot more into the central concept of “message in a bottle” than the first one had. In this, the bottle was still divided into slices but the environment inside the bottle was much more like a diorama. The player was a traveler, and each of the levels was based on a message in a bottle they wrote about an experience they had had on their journey. The player still had to work to traverse the space but now in a far more organic way: they would not move in set channels, but would need to manipulate ramps, rocks, branches, and other various pieces of the environment to create a path to the goal. Each thing they could interact with was stationed in a different slice of the bottle.
An early digital prototype of this idea was never finished, but involved ruins sticking out of sand in a desert. The player would reveal these ruins by rotating them out from the sand, discovering their complete shape as they played with the world and then having to use them in more practical ways to create what they needed. For example: rotating one piece of ruin into another, to break the weaker one and let it collapse into a ramp that would take the player to the goal.
This proved harder to implement than anticipated, even after enlisting the help of a talented friend who would later go on to be my Art and Design lead on this project. However, in attempting to make this prototype work and considering the ways in which it forced the player to move, I stumbled on an important design choice: Why not just move the bottle as one unit?
The First Complete Prototype
That decision having been made, a cascade of other choices followed. Though I have a background in game development, creating a prototype of an entirely new mechanic means not having any foundation of code to rely on from developers past. My strength wasn’t to be found in programming, but I do have an extensive background in hobbyist fabrication of props and costumes— so I made a physical prototype.
The challenge was: I had to design a prototype level I could feasibly make a physical version of, and also plan out how to make it. With the magic of thermoplastic, hot glue, and self-drying clay covered in paint, I put together an analog level that used a small glowing ball as a stand-in for the player.
The narrative of this level was a departure from the prior idea of the traveler. In discussing the concept with advisors and colleagues, the recurring question remained: what is a message in a bottle, and why does it matter? At this point in the development process I decided that a message in a bottle was an attempt to connect with someone. Reaching out into the world without certainty that the world would reach back.
Inspired by that idea, I came up with a story about a child who ran away from home and returned to a place they felt happiest: the desert, where their family would take them camping in better days gone by. They set up a tent, ignorant of a storm about to roll through and flood them out; but before they had left home they had written a message and stuffed it in a bottle. The player was someone who found the message and had the power to go back and save the child— they were following the path that the child had left, seeking to find them and assuage their loneliness and take them home before the storm hit.
In retrospect, this was an ambitious tale to tell in the medium I was attempting to tell it in, but at this point I was far more interested in finding the boundaries of what I could do and work inwards from there. The puzzle would have the player turn the bottle so as to see that what appears to be a wall from the side is, in fact, a canyon when viewed from above. Then, they would need to reach a high ledge with a tunnel that allowed them into the next area. This proved a more difficult and finesse-worthy task than originally anticipated, and solidified my decision that this game shouldn’t require flick-of-the-wrist mechanics as they were very frustrating and didn’t lend themselves towards the kind of calm and thoughtful feeling I wanted to evoke. Finally, they would find the child in the tent— another glowing light to match the one in the player character. Though I couldn’t render it in clay and plastic, the finale would be that together, these two characters push the cork out of the bottle and leave.
The feedback I got from players was generally positive. Outside of the difficulty of the second challenge, players loved turning the bottle, putting their eyes up to the plastic, and looking at all the tiny details. Even if it wasn’t representative of a digital mobile experience it evoked the things that I wanted to express about the larger feeling of the game: so I would count it as a general success.
Original Experience Goals
My original experience goals changed with iteration, but the primary set that carried through the first few months of the development process were as follows (taken directly from a pitch document written in the spring semester of 2019):
Feels Good - This game should be like wrapping yourself in a warm blanket, or re-reading a favorite book: a comforting experience that makes players want to come back to it.
Replayable - To make people want to come back to it, not only does the game have to feel good but also provide experiences worth returning to. New ways of solving things, new fun details to explore, etc.
Escapism, With Souvenirs - Being a place to return to for comfort and enjoyment, Bottles is a game of conscientious escapism—a place to go for some time, but with a narrative that stresses the importance of connection and coming out of our own metaphorical emotional “bottles” that drive us to escapism in the first place. You can hang out here for a while, but it’s important to go back, too. And we want the player to feel just a little bit more prepared to face that.
Our experience goals ended up changing later in the production process, but are useful to consider in the light of how development unfolded around them. These tenets didn’t leave the broader aim of Bottles as a game per se, but were refined and sharpened in the crucible of critique and iteration, and since then have found their way into the play experience following the the changes we made to gameplay in service of our newer, more precise experience goals.
With that foundational work under my belt, and my first true team member already committed to the project, I pushed ahead into the important task of taking this idea into a practical development process.
I already had a lead on prior art going into pre-production, which ended up being key to expressing my game concept both to potential advisors and teammates alike. First, I knew what mobile titles I wanted to evoke similar experiences to, and second, I knew how I wanted Bottles to be different— which meant finding pre-existing games that I could use to help explain what set Bottles apart.
First, I looked to Monument Valley by Ustwo Games. This is a mobile game in which the player uses very simple tap and swipe controls to manipulate an M.C. Escher-esque world of mind-bending pathways, leading your tiny princess character through beautifully-designed pastel spaces. It was a favorite of mine, both when I was getting blood drawn or was simply bored on the subway. In my conception, the audience for a game like Monument Valley was similar to the player base I imagined seeking out a game like Bottles. People who enjoy puzzles, who are attracted to lush worlds, who want a more intellectually and artistically stimulating experience out of mobile puzzle games than standard match-three fare. Furthermore, the things I felt when I played Monument Valley were similar to the experience goals I had outlined in documentation.
In a change of pace, another inspiration that was influential in how I conceptualized Bottles was not digital at all: puzzle balls. These are plastic toys in which a clear shell contains an obstacle course and at least one metal marble. The goal is to start your marble at one location, and through careful manipulation of the ball, guide it safely through the course and to the goal. While sometimes being riotous fun, more often than not they are enjoyed more for their infuriating challenge than their ease. In drawing inspiration from these I sought to take the tactile analog feel of moving a puzzle ball in your hands and translate it into a digital object, hopefully removing the infuriation element that is endemic to the original.
Often in describing Bottles to people who had never heard of it, my shorthand was something along the lines of “imagine a mobile game that is like if a puzzle ball was crossed with Monument Valley”. However, there is one more title in particular that became more and more valuable as a point of reference for me as the development process went on.
Windosill is a browser-based game by the studio Vectorpark. In it you move a small toy left to right through a series of vignette-like scenes, each one with a different mechanic. The beauty of Windosill is that the game never has a tutorial. It never tells you what to do— the player is left to explore and experiment and figure out the goal of each scene on their own. Though it became clear that Bottles was not a game that could go without a tutorial, when thinking about the design of our primary level, Shipwreck, I often went back to the experience of Windosill in trying to understand how to make something that is parsable by players while not suffocating their drive to explore with over-sharing exactly what to do when and where.