240,000 Miles. A Million Reasons.
Moon Taxi is an experiment in video game production methodology and gaming narrative
focus. The project was designed to produce an experience with a unique focus on audio narration
and to demonstrate the feasibility of novel approaches to two common problems in game
development. First, Moon Taxi's design aims to minimize the effort required to meaningfully
integrate the work of game designers and subject matter experts. Second, Moon Taxi tests a method
for increasing the potential pool of user created content. This paper will discuss the project's
design, approaches utilized to address these issues, solicitation of content creators, the game
development process, playtesting and future steps.
Moon Taxi is a fanciful space flight simulator. In its most basic form, the gameplay consists
of navigating a space taxi in a two dimensional plane as that plane moves forward to the Moon,
through an asteroid field. The player must avoid asteroids and intercept words in their path. In this
manner, the player plays the role of a space taxi driver taking a passenger to the Moon.
The target audience for Moon Taxi is anyone who enjoys science fiction stories and has any
interest in games. Each level of the game includes a two to seven minute audio recording of an
original work of science fiction told from the perspective of a passenger traveling to the moon in a
taxi. The gameplay is designed to engage the player with the story in a relaxed but focused manner,
without interfering with the player's ability to listen to and comprehend the narrative. Furthermore,
the game emphasizes the narrative and visually connects it to the game world. At times during the
narrative, 3D representations of key words from the audio are introduced into the game
environment. The player is rewarded for navigating the ship towards the words and collecting
One of the unique aspects of Moon Taxi is that the narrative is more important than the
gameplay. Every player hears the complete story, regardless of how poorly he or she plays. None
of the game's punishments ever interrupt the narrative. The game's rewards give the player a greater
selection of narratives to enjoy. Only a moderate level of success is required to get to the next level
(and the next story). However, playing well yields more choices for the next narrative.
One question the game's design raises is whether combining listening and playing will
increase the enjoyment of both aspects or whether the combination will degrade the experience.
Some studies have demonstrated that divided attention generally reduces individuals' capacity to
encode memory (Craik et al., 1996) in addition to diminished reaction times (Brouwer et al., 1991).
However, Moon Taxi's goal isn't for players to remember the story, but rather, for them to enjoy it.
Similarly, Moon Taxi's gameplay should is never so difficult so as to significantly interfere with the
In the 17th century, Baruch Spinoza made a non-intuitive proposal for which recent studies
have shown support (Gilbert et al., 1991). Spinoza thought that people must first believe something
before they can disbelieve it. In other words, understanding something and believing it are a single,
inseparable process. Only after something false has been comprehended and believed can the
arguments against it be assessed and that knowledge can be marked as untrue. Gilbert's studies
have shown that distracting subjects while they are processing information can increase the
likelihood that a subject will skip the second step of cognition and fail to scrutinize the veracity of a
newly learned fact. In a best case scenario for Moon Taxi, the gameplay's distractions from the
story could translate to an increased suspension of disbelief, the heart of any good theatrical
Bridging the Division of Labor
Serious games projects often require the integration of complex knowledge from highly
trained specialists in disparate fields. Hopelab, the team behind Re-mission, a cancer treatment
training game for pediatric patients, employed separate medical experts and game design experts to
produce a game that was medically accurate, informative, and fun (Cole et al., 2006). Packy &
Marlon, another medical game, employed individuals separately versed in managing diabetes and
developing games for the Super Nintendo Entertainment System (Brown et al., 1997).
Serious games projects in which the game developers are not also the content experts have a
knowledge gap. Bridging that gap can be a significant barrier to producing a coherent, effective
game. The effort required to combine content created by subject matter experts and the efforts of
game development experts depends in large part on the nature of the desired link between the
content and the gameplay.
Malone (1981) claimed that games with intrinsic fantasies, where players use a skill that is
connected to the fantasy employed, are more interesting and can be more educational than games
where the fantasy has a complete disconnect from their actions (such as Hangman). In Moon Taxi,
the intrinsic link between stories and gameplay is enforced by simple guidelines for the subject
matter experts (the story writers and actors). The writers are given 3 main rules:
• Start your narrative with "Take me to the moon"
• Tell the narrative from the perspective of one or more passengers traveling to the moon
• Include why the passengers are going to the moon
By following these rules, content creators are free to make a broad variety of assumptions
and still be assured their content will integrate with gameplay in a compelling manner.
Decoupling Users and User Created Content
Given the right conditions, users will create valuable content which will breathe life into a
game. There are games like Second Life which depend entirely on user created content and games
like Counter-Strike, Gary's Mod, and Red Orchestra that are themselves user created modifications
of commercial games. But the potential pool of content creators for each of those games is limited
to the intersection of people who play that game, want to make content, and are capable of making
content. The learning curve in these examples is large, requiring an understanding of the games'
scripting languages and their deployment mechanisms.
Moon Taxi sought to ensure that non-programmers could contribute content. A more
challenging goal was to eliminate the prerequisite that content contributors from the general public
also had to be players. The rules established for content writers intentionally omitted any
requirement for understanding the nature of the game. This clean separation provided an
opportunity to advertise the joys of the content creation experience separately from the game play
experience, effectively decoupling game users from created content.
Moon Taxi's decoupled model has two major advantages:
• The skill set required to make content is writing, not programming
• The potential contributor audience is not limited to Moon Taxi players; Content creation is
technology agnostic – contributors do not need to have ever played a video game in order to
It was important to the Moon Taxi project to establish a stable, on-line presence early in the
development process. I purchased moontaxi.org and paid to host it on DreamHost. Having had
positive previous experiences with the Drupal content management system, I created a custom
Moon Taxi theme and then focused on content and workflow.
I use Drupal's blogging capabilities to post regular updates about the process to
moontaxi.org's front page. Other pages include submission details, game information and site
Using Drupal's workflow capabilities, I constructed a pipeline whereby authors can submit
stories, the stories then go into a review cue, and then they are either marked as complete or in need
of an audio recording. Once a story has passed review, it is automatically added to the publicly
viewable list of stories. To encourage feedback and a sense of community, visitors can read and
rank reviewed stories from one to five stars.
Once the development of the game had progressed to near feature completion, I created two
promotional videos based on gameplay. The first was targeted towards potential game players and
focused on describing Moon Taxi’s style, aesthetics and music. Having been wisely advised that I
would be better served by first targeting the content creator audience, I created a second video
targeted towards potential writers and actors. The second video is a recording of a full play through
of one Moon Taxi level which I had written and performed. The story, entitled Spornak, meets the
requirements for a Moon Taxi story while simultaneously being a detailed advertisement for writers
and actors to submit their own stories to the Moon Taxi project. Both videos were uploaded to
YouTube and are embedded for viewing on moontaxi.org.
Once moontaxi.org was established and populated with submission rules, blog posts and
videos, I began recruiting content contributors using multiple channels. Initial word-of-mouth to
personal acquaintances and friends of friends yielded some enthusiasm, but few results.
Between October 9th and November 25th I ran a pay-per-click internet advertising campaign
using Google Adwords. Over that period the advertisements were shown approximately 1.5 million
times and attracted 909 clicks for a cost of around $70.
Links to moontaxi.org and a description of the project were posted at multiple sites
dedicated to listing writing contests. A free listing on one site,
http://writingcontests.wordpress.com, resulted in nearly 90 clicks during the same period as the
To boost interest and qualify for more postings, I introduced a Moon Taxi contest. I posted a
new contest page on moontaxi.org announcing $100 for the best story with a performance and $50
for the best written story. A few weeks after updating moontaxi.org and the Google advertisements
with the contest information, I received stories from Massachusetts, India, Australia and Ireland.
Only the story from Massachusetts fully met the story submission guidelines. The contest deadline
is January 31st, 2010, so there is still strong potential for more entries.
Another channel for garnering the attention of players and potential contributors is via
contests hosted by other groups. I entered Moon Taxi in the student category of the annual
Independent Game Festival's competition. Contest results won't be announced until January, but
preparing for the contest was a good exercise which pushed me to raise the quality of the game and
to practice my articulation of the game's design and purpose.
A Class Exercise
In parallel with the advertising efforts, I collaborated with Professor Lynn Scott to include
Moon Taxi as an exercise in a freshman writing course she was teaching in James Madison College
at Michigan State University. Students were invited to write and record a Moon Taxi story to
qualify for optional honors credit. Five students accepted the challenge and Professor Scott met
with them several times during the course to review their stories.
The students' work culminated in a recording and performance session. I brought recording
equipment (and refreshments) to the final evening gathering where each student performed their
story. One student brought two friends to assist with performing the multiple voices in some
The technology choices for Moon Taxi were driven by considerations of optimizing
successful deployment of the project and of the experience's value to my game development
portfolio. The deployment concerns were chiefly: to be able to commercially distribute the project,
to provide a bug free user experience, and to limit development costs. Portfolio considerations
included: deployment platform popularity, the industry value of the development tool set, and gaps
in my existing portfolio.
An unfinished Moon Taxi prototype was initially created for the Android mobile platform.
The Android platform was very new at the time and while it had many positive qualities, it lacked
the industry support, game libraries, entertainment focus and commercial appeal to be the optimal
platform for Moon Taxi.
Nintendo, Sony and Microsoft's game consoles are appealing targets because of their gaming
focus and the likelihood that the game industry would value developers with experience in those
platforms. Of the three current generation consoles, Microsoft's XBOX 360 has by far the greatest
level of support for independent developers. Microsoft's highly popular Visual Studio has a capable
free edition. They also provide a free extension called the XNA game development studio which
provides a variety of game specific developer libraries and has a sizable on-line set of tutorials and
Microsoft sells memberships to their XNA Creators Club Online program for $100 annually.
Membership grants the ability to run XNA developed code on one properly authorized XBOX as
well as access to an extended library of on-line tutorials and sample content. The most appealing
aspect is the access that Creators Club members are granted to Microsoft's peer review and XBOX
publishing mechanisms. Members may submit games to the community to be tested for bugs and
labeled for mature content. Once a game has passed testing, it can be priced and published on the
XBOX Live network, making the it available for purchase via any XBOX connected to the internet.
XNA provides many conveniences on top of raw OpenGL or DirectX development, but it
does not provide the rapid development experience of more all inclusive tools such as the Torque or
Unity game engines. If development speed had been a more critical goal than than technical
experience, a more inclusive tool might have been a more appropriate choice.
NASA/JPL-Caltech's libraries of liberally licensed space imagery were invaluable assets.
Even Moon Taxi's taxi model was based on a freely available 3D model of a NASA space shuttle.
Some moderately complex manipulations were required to adapt the images for Moon Taxi,
including mapping textures over models and creating a nearly seamless "sky box" out of space
Musician Stephen Dranger provided two songs from his existing catalog for the project. I
overlaid one song with a public domain speech by John F. Kennedy to create the Moon Taxi theme
song. Alex Wilson provided a sketch of the pop can monster to use as a developer logo. I also
created supplemental art using Blender, GIMP and Photoshop.
Five playtesters volunteered to test a late Moon Taxi prototype and complete a short on-line
survey. The playtesters were encouraged to talk out loud as they played to reveal their frustrations,
joys and questions to me as I observed. Admittedly, this standard practice was less than ideal for
Moon Taxi because the player's talking interfered with the player's ability to hear the game's talking.
Three of the testers, who we will refer to as Speedster, Waity and Sailor, had very little
experience playing video games. Two, The Pro and Gamer Daddy, were very experienced. All
testers were males in their mid 20s to late 20s. Testers Speedster, Waity and Gamer Daddy tested on
a laptop, Sailor and The Pro tested on an XBOX 360.
Interfaces. All of the testers except Gamer Daddy experienced some level of confusion
operating the interfaces for starting a game and selecting a level. Speedster didn't understand how
to start the game from the title screen and pounded every key until he hit the right one. Sailor
wondered aloud what he was supposed to be doing during the introduction screen and wiggled the
gamepad analog buttons until the screen instructed him to press start. Waity and The Pro didn't
touch anything until the introduction had passed and it instructed them to press enter or start (the
instructions were different for PC and XBOX).
The level select screen asked players to choose a level and then "press start." Speedy hit
space before realizing he needed to hit enter. Waity wondered aloud where the start key was and
then guessed it meant the enter key. The XBOX users had no trouble finding the button marked
start on the XBOX.
Between levels, the players were shown the end of level screen and then the level select
screen. Speedy considered the end of level screen neat on first view and a boring interruption on
subsequent views. Waity and The Pro were confused that the end of level screen announced they
had delivered a passenger. All testers easily found their way from the end of level screen back to
the level select screen.
On the second visit to the level select screen, Speedy, Waity and Sailor all missed the yellow
arrow which hinted they could press right to advance the story menu. They each restarted the
tutorial. Gamer Daddy tried to proceed to the next level, but had not collected enough dust.
Game comprehension. The first level all testers played was a tutorial narrative. In the
tutorial narrative, an experienced taxi driver gives verbal instructions to the players including the
keyboard or XBOX controls, that hitting words will increase the dust collection, that increasing the
dust collection unlocks new stories and that hitting asteroids decreases the dust collection.
Speedy frantically pressed keys and wondered what he was supposed to be doing before
pausing to listen to the narration and declaring "Now I get it. This is training." Waity removed his
hands from the keyboard and listened to the tutorial in its entirety without touching the computer,
even when the narrator instructed him how to move the keys. It wasn't until the level was over that
he realized he was supposed to be controlling the taxi. Sailor listened to much of the tutorial
silently while moving his ship around, then declared: "Am I supposed to collect dust? I guess I don't
get it." He swore many times and collected zero dust. Sailor played the tutorial level 3 times,
dramatically improving each time. On one play through, he cheered the approach of the last word
(hoot): "C'mon 'hoot'!" Gamer Daddy did not understand he was supposed to collect dust until he
was presented with a low dust count at the end of level screen. On his second play through of the
tutorial he reported being much more motivated now that he knew he needed to collect something.
However, he was not entirely certain that hitting words was the way to collect dust. Gamer Daddy
was more relaxed than most testers and wished he could make the taxi do maneuvers just for show
to kill time between asteroids. He also commented that he would be interested in showing the game
to his young child if the few mild expletives in the stories were censored.
The playtesters made several observations which were counter to the actual nature of the
game. Speedy concluded incorrectly that the "dust and the words are not related." Waity assumed
that his speed increased at the second level because it was the second level and not because he had
been more successful in dodging asteroids. Sailor thought some of the words had moved to avoid
him. Survey results showed that everyone initially thought there would be much more serious
consequences for hitting an asteroid.
Story comprehension and Enjoyment. Several testers remarked that they were too involved
in understanding the game mechanics and reporting their thoughts to pay much attention to the
stories. Survey results were mixed as to whether the game increased or decreased the enjoyment of
the stories. However, survey results were equally mixed as to whether the player enjoyed reading
science fiction in general.
Improvements. I implemented improvements to Moon Taxi to address several issues
discovered during testing. There were four main interface improvements. To clarify the transition
from the title screen to the level select screen, I added a menu including a "Start Game" option
which overlays the title screen when one of multiple keys is pressed. To reduce uncertainty on the
level select screen, I added text directing players to press left and right to select stories. Also, for
the PC version, "press start" was replaced with "press enter." After leaving the end of level screen,
the level select screen now automatically places the player at the details screen for the next level if
they have collected enough moon dust to unlock it.
I also made gameplay adjustments to reduce confusion and stress. I edited and re-recorded
the tutorial narrative to put the description of the controls at the very beginning. The algorithms
controlling asteroid quantities were significantly modified. During the tutorial, no asteroids will
appear for the first 30 seconds and the taxi will maintain a constant speed for the first 50 seconds.
For all levels, asteroid quantity will initially start at 0 and increment over the course of the level to
reach full quantity. The maximum quantity of asteroids is higher for each subsequent level.
Publishing the game through the XBOX Live Indie channel will require XBOX specific
refinements that were not essential to the development goals of the project, such as removable
memory card support. Also, the story contest is still accepting submissions until February 2010.
This could increase the quantity and/or quality of the stories composing the levels. Depending on
player and contributor interest, there is potential for a second volume of stories after the first
volume is published.
The project's two main goals were to utilize a simple and meaningful method for bridging
the gap between content creators and developers, and to solicit user created content from non-users.
The success of these goals cannot be completely measured until the content creation contest has
ended and the game's merits have been assessed by a wider audience of players. However, based on
the current state of the project and testers' reactions, Moon Taxi's development process has
integrated content and gameplay with minimal interplay between content creators and the game
developer. Furthermore, only two of the story contributors had played Moon Taxi at all.
Moon Taxi has come a long way from its abandoned Android prototype. The submitted
version includes 6 stories, polished visuals, original music, and refinements to the interface and
gameplay based on playtesting. The project has established and advertised an on-line presence.
Game content creation was even used as a real world, college level, academic exercise. It has been
a fantastic learning experience for myself and it will hopefully inspire others to consider non-
traditional content creation methods as well as bring a few smiles to gamers faces.
Brouwer, W.H, Waterink, W., Van Wolffelaar, P.C., & Rothengatter, T. (1991). Divided attention in
experienced you and older drivers: Lane tracking and visual anlysis in a dynamic driving simulator.
Human Factors: The Journal of the Human Factors and Ergonomics Society, 33, 573-582
Brown, S. J., Lieberman, D. A., Gemeny, B. A., Fan, Y. C., Wilson, D. M., & Pasta, D. J. (1997).
Educational Video Game for Juvenile Diabetes: Results of a Controlled Trial. Medical Informatics,
Cole, S. W., Kato, P. M., Marin-Bowling, V. M., Dahl, G. V., & Pollock, B. H. (2006). Clinical
trial of Re-Mission: A video game for young people with cancer. Cyberpsychology & Behavior, 9,
Craik, F.I.M, Govoni, R., Naveh-Benjamin, M., & Anderson, N.D. (1996). The effects of divided
attention on encoding and retrieval processes in human memory. Journal of Experimental
Psychology: General, 125, 159-180.
Gilbert, D. T., Malone, P.S, & Krull, D. S. (1990). Unbelieving the unbelievable: Some problems
in the rejection of false information. Journal of Personality and Social Psychology, 59, 601-613.
Malone, T. W. (1981). Toward a theory of intrinsically motivating instruction. Cognitive Science, 4,