University of Southern California Dissertations and Theses

Civic games with 'local fit': embedding with real‐world neighborhoods and place‐based networks

(USC Thesis Other)

Civic games with 'local fit': embedding with real‐world neighborhoods and place‐based networks

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content i

CIVIC GAMES WITH ‘LOCAL FIT’:

EMBEDDING WITH REAL-WORLD NEIGHBORHOODS
AND PLACE-BASED NETWORKS

by

Benjamin Stokes

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(COMMUNICATION)

June 2014

Copyright 2014 Benjamin Stokes

ii

DEDICATION
To those who see play in the system
and organize for justice

iii

ACKNOWLEDGEMENTS

I am forever indebted to the generosity of my mentors. Two members of my
committee joined me in early conversations at Games for Change, and I followed them to
USC: Henry Jenkins and Tracy Fullerton. Tracy’s lab for game innovation is justifiably
world-renowned; not only did it support one of the case studies, but the courage to tackle
real-world games was only possible for me with Tracy as a thought partner. Henry is my
co-chair; if this dissertation succeeds in thinking outside the traditional frame on civic
life, it is only because of Henry’s mentorship – including his weekly think tank on new
civic trajectories and participatory culture.
Half the ideas in this study emerged in the field, especially in South Los Angeles
– and under the mentorship of François Bar and Sandra Ball-Rokeach. Sandra’s
ecological and multi-level approach to studying neighborhoods is at the core of this
dissertation, and was hugely influential for my own thinking. On the ground, her
leadership of the Metamorphosis Project provided a vital interface to working with
organizations in South LA – especially for the participatory mapping case study.
François co-chairs this dissertation, and anchors my approach to bottom-up design. Not
only was the ParTour case study a collaboration with François, but the broader analytic
framework for ‘local fit’ came through our design research explorations into how civic
engagement can be situated in live activities and place-based communities.
I would never have pursued a Ph.D. without the encouragement of Connie
Yowell. Her visionary leadership at the MacArthur Foundation taught me the value of
research to help build fields, especially across levels and institutions. The encouragement
iv

of James Paul Gee has also been transformative, including for his sustained cultivation of
a young scholars network, and for helping me to see how my civic past can build distinct
kinds of theory. Bill Tierney, who served on my qualifying exam committee, also
provided key mentorship for how a non-traditional scholar might still have a role at the
university.
For the case study on Macon Money, the analysis was only possible due to the
generosity of the original evaluation team. Particular credit is due to Madeleine Taylor of
Network Impact in partnership with Anne Whatley of Cause Communications. Their
excellent report is cited throughout this study. Additional thanks are due to the Knight
Foundation, which invited me to join their advisory group on assessing their social issue
games (including Macon Money), and later helped me to secure access to the original
data; particular thanks are due to Mayur Patel, Beverly Blake and Jessica Goldfin. Kati
London, one of the game’s lead designers, deserves great thanks for her incredibly
grounded yet thought-provoking insights into the game’s philosophies and pragmatic
realities.
For the case study on Reality Ends Here/Xposure, incredible access to the data
was provided by Jeff Watson, Tracy Fullerton, Simon Wiscombe, Ioana Literat and
Alison Trope. More importantly, each and every person named was also profoundly
generous and insightful in collectively analyzing the functioning of the game and its
implications across schools; the chapter would never have happened without their
collaboration and insights. The original game was created by Jeff Watson as his doctoral
research and design project, and was designed by Jeff, Simon and Tracy together. A debt
v

of gratitude is also owed for the game insights and student access provided by Holly
Willis and Tara McPherson.
For the case study of ParTour, the collaborators on research and design come
from many organizations with deep roots in Los Angeles. From T.R.U.S.T. South LA,
particular thanks go to Tafarai Bayne and Sandra McNeill; Los Ryderz; the East Side
Riders, especially John Jones; Community Services Unlimited, especially Neelam
Sharma and Dyane Pascall. On the research side at the University of Southern California,
the field work was a joint venture with François Bar, Karl Baumann, George Villanueva,
Otto Khera, Maren Brombeiss, Cesar Jimenez and Theresa Gonzalez. The support of the
Metamorphosis Project and the USC Annenberg Innovation Lab were critical. Funding
support is acknowledged from the Guggenheim Foundation, the USC Neighborhood
Outreach grants office, the Innovation Diploma program, and the USC Annenberg
Innovation Lab.
The foundational tools of Vojo were only possible through the dedicated and
sustained work of Madelou Lourdes Gonzales, Manuel Mancía, Adolfo Cisneros, Crispín
Jimenez, Marcos & Diana, Alma Luz, Ranferi, and all the community correspondents of
VozMob.net, as well as Amanda Garces, Natalie Arellano, Pedro, Raul Añorve, Marlom
Portillo, Neidi Dominguez, and the whole IDEPSCA extended family; and my fellow
community engaged researchers at USC including Sasha Costanza-Chock, Carmen
Gonzales, Melissa Brough, Cara Wallis, François Bar, and Veronica Paredes, not to
mention Mark Burdett, Brenda Aguilera, and Troy Gabrielson. This amazing list of
vi

individuals is testament to the amazing mutual learning that took place over five years
and which has influenced this dissertation in countless small and enormous ways.
For the analysis of Re:Activism Atlanta, I am deeply grateful for Colleen
Macklin’s invitation to join the design team and help run the game for PETLab/MODA.
Similar generosity was shown by the rest of the design team and hosts, including Adam
Rafinski and Ben Pincus. Additional thanks for the contributions of Timothy Crimmins,
John Sharp, and the National Center for Civil and Human Rights. The amazing Celia
Pearce gave me a home at Georgia Tech that was all-too-brief, incredibly generous, and
vital for the space and collaborative community to develop many of the ideas in this
study.
The game insights of this study owe tremendous gratitude to a decade of
conversations with Eric Zimmerman, Colleen Macklin and Kurt Squire. Deep thanks are
due to my colleagues at Games for Change, who encouraged my scholarship as a way to
advance the field of civic games, especially Dave Rejeski, Suzanne Seggerman, Barry
Joseph, and Asi Burak. More recently, Dmitri Williams has pushed the rigor and breadth
of my understanding of games and the multiple methods by which they can be analyzed,
including in this dissertation. Many of the network analytics used in this study were
explored under the guidance of Tom Valente, and his encouragement helped me take
them further.
Funding and support for this research came from several invaluable sources. I am
particularly thankful for an early collaboration grant from HASTAC/The MacArthur
Foundation to organize a convening at USC on games and mobile civic learning. The
vii

support of Intel Labs’ Interactions and Experiences Research (IXR) Group provided the
initial support to theorize real-world civic games with Jeff Watson and Susana Ruiz,
building on a panel with Tracy Fullerton and Stephen Duncombe. Most recently, the
Morkovin Fellowship at USC provided me with incredible flexibility in this past year to
simultaneously visit field sites, follow-up with earlier interviews and focus on the
writing. The invaluable support of Imre Meszaros, Anne Marie Campian and Sophie
Madej, among other dedicated staff at USC, provided a foundation that was not only
secure but flexible and even inspiring throughout the research process.
Saving the most important for last, I would never have made it here without my
family. My partner Lily Siegel has been the best cheerleader I could hope for, believing
in me when I became lost, and keeping my sanity in adventures and music and cooking
and road trips, from Los Angeles to Atlanta, from San Francisco to DC. My parents,
Christine and Walter, made possible the spirit of this dissertation by cultivated a deep and
consistent love of learning; between the lines of this study, a careful reader will see the
constructivist play of my mother’s classroom, and my father’s political debates about
collective action. The grit to finish this dissertation is owed to the support of my full
extended family, living and deceased, living nearby or in Peru, with special thanks to my
siblings Caleb, Alexandria and Madeline. I love you all.

viii

LIST OF TABLES

Table 1: Gender homophily in the first year ................................................................... 113
Table 2: Participation by discipline (academic major) ................................................... 113
Table 3: Highest scoring players, alongside their centrality rankings ............................ 117
Table 4: Summary of game cases in terms of ‘local fit’ ................................................. 222
Table 5: Process for articulating mechanics with a socio-cultural lens .......................... 225

ix

LIST OF FIGURES

Figure 1: Differentiating games for real-world cohesion and community ....................... 12
Figure 2: A sample network graph.................................................................................... 17
Figure 3: Block-matching as conceptual filter .................................................................. 23
Figure 4: Dimensions of ‘local fit’.................................................................................... 30
Figure 5: Promotional map of the College Hill Corridor .................................................. 37
Figure 6: Bonds and players of Macon Money ................................................................. 38
Figure 7: Screenshot of the Facebook page for the game ................................................. 42
Figure 8: Winning totals per player .................................................................................. 43
Figure 9: Network graph of matched pairs from Macon Money ...................................... 44
Figure 10: A framework for studying games using three interlocking schemas .............. 51
Figure 11: Soap Box Derby graphic, as part of a social media campaign ........................ 64
Figure 12: Second Sunday concert in Macon, Georgia .................................................... 65
Figure 13: Business earnings by location, midway through the game ............................. 69
Figure 14: Advertisement for a Macon micro-loan fund .................................................. 91
Figure 15: Cards used in the original game ...................................................................... 94
Figure 16: Playing the game Reality Ends Here in person ............................................... 96
Figure 17: The “Letters of My Lai” project on the game website .................................... 97
Figure 18: Contributors (red circles) and projects (blue squares) in the first year ........ 101
Figure 19: Team size frequency distribution (multi-year) .............................................. 103
Figure 20: Members of the Tribe, from a film they made in the game........................... 104
Figure 21: Players' scores as a function of their network position (year two) ................ 107
x

Figure 22: Network evolution over one game (a semester). ........................................... 109
Figure 23: A mentorship encounter in the game ............................................................. 111
Figure 24: Suggested groups based on participation in the first year of the game. ........ 119
Figure 25: Famous alumni Steven Spielberg and George Lucas .................................... 127
Figure 26: Cards for Xposure, based on Reality Ends Here ........................................... 136
Figure 27: A mysterious postcard for Xposure (left), and a video (right) ...................... 137
Figure 28: Social media buzz for Xposure ...................................................................... 139
Figure 29: Comparing communities: media production experience ............................... 150
Figure 30: “I walked this street every morning” ............................................................. 161
Figure 31: Riders pose for a group picture in front of the Watts Towers ....................... 162
Figure 32: Printed map of the Watts Ride ...................................................................... 166
Figure 33: ParTour across storytelling levels ................................................................. 168
Figure 34: Communication asset mapping in South LA ................................................. 170
Figure 35: Bike-mounted tablet used to provide feedback loop ..................................... 175
Figure 36: Re:Activism players reenact a historic speech ............................................... 180
Figure 37: Selected materials from Re:Activism Atlanta, including mission map ......... 182
Figure 38: Pictures of Re:Activism Atlanta by players ................................................... 185
Figure 39: Flyer describing a ParTour event with multiple bicycle clubs ..................... 190
Figure 40: To distribute the map, bicycle club members visit a local bike shop ............ 198
Figure 41: Adaptation model for scaling using dimensions of ‘local fit’ ....................... 219

xi

ABSTRACT

This study offers a new way to understand how game-based activities can deepen
local communities tied to place, including how to evaluate some of the unusual risks.
Real-world games go beyond training by attempting to get something done, like raising
funds or deepening trust across race and class divides. By playing games tied to human
networks, participants can build social ties. But can the community deepen its collective
capacity? Real-world games require new frameworks to align with regional models for
economic development and social justice. This study introduces “Locally Situated
Games” as a new category to distinguish and compare games that strengthen place-based
communities. Several game-based activities are analyzed in depth, including Macon
Money, Reality Ends Here, and ParTour South LA. Success for these designs is
hypothesized to depend on fitting deeply to an existing community, including to critique
or change the community. To investigate, this dissertation proposes ‘local fit’ as a
conceptual framework to unite the “how of games” with the “how of communities” in
three dimensions. Each dimension brings a distinct cognitive perspective, including:
network mechanics, group identity, and the local ecology of communication. ‘Local fit’
is used to analyze the case study games, and to explain one case of replication failure.
The analysis integrates multiple methods and data sources, including participant
observation, network analysis of player relationships, interviews with game designers,
and surveys of participating businesses. Findings include a model to localize place-based
games, a process for articulating game mechanics in socio-cultural terms, several ethical
considerations, and consideration of Situated Games as a kind of socio-economic policy.
xii

TABLE OF CONTENTS
DEDICATION .............................................................................................................................................. ii
ACKNOWLEDGEMENTS ........................................................................................................................ iii
LIST OF TABLES .................................................................................................................................... viii
LIST OF FIGURES .....................................................................................................................................ix
ABSTRACT ..................................................................................................................................................xi
CHAPTER ONE: OVERVIEW ................................................................................................................... 1
Introduction ............................................................................................................................................ 1
Defining success in a network mode ........................................................................................................... 7
Which games? ........................................................................................................................................ 7
Community coherence, place and network methods ............................................................................ 13
Proposing ‘Locally Situated Games’ .................................................................................................... 22
Research Approach ................................................................................................................................... 25
Fit as conceptual framework ................................................................................................................. 26
Three dimensions of ‘local fit’ ............................................................................................................. 26
Case study approach and selection ....................................................................................................... 31
Data sources and analytic methods ....................................................................................................... 33
Analytic lens: communication as fabric, and games as activities ......................................................... 33
Limitations ............................................................................................................................................ 34
CHAPTER TWO: MACON MONEY & URBAN POLICY .................................................................. 36
Overview of the game ................................................................................................................................ 36
Statistics in context: one player’s trajectory ......................................................................................... 39
Is it a Situated Game? ........................................................................................................................... 47
Data collection and analysis ................................................................................................................. 57
Investigating Fit: Game mechanics (and local modes of meeting) ........................................................... 58
(A) Mechanic for social ties: “to greet in public space and match” (for mutual benefit) ..................... 60
(B) Mechanic for business ties: “to buy locally” (for self and community) ......................................... 67
Addressing manipulation ...................................................................................................................... 73
Toward ‘meaningful choices’ for civic impact ..................................................................................... 75
Rejecting alternative frames ................................................................................................................. 82
As socio-economic policy (reconceptualizing games)............................................................................... 84
Hints of more ‘local fit’ ............................................................................................................................. 92
xiii

CHAPTER THREE: REALITY ENDS HERE & NETWORKS TIED TO A UNIVERSITY ............. 93
Overview of the game ................................................................................................................................ 93
Forming networks – as an institutional goal in education .................................................................... 98
Visualizing ties: the network emerges ................................................................................................ 101
Toward network methods for community games ..................................................................................... 104
Data for network analysis ................................................................................................................... 106
Game score as network indicator ........................................................................................................ 106
Testing network outcomes tied to fit .................................................................................................. 108
Guiding organizers with network feedback from games .................................................................... 115
Investigating ‘local fit’ ............................................................................................................................ 120
Qualifying as a Situated Game ........................................................................................................... 120
Fit with mechanics .............................................................................................................................. 124
Fit #2: to the local identity/brand ........................................................................................................ 126
Fit #3: to the local ecology of communication ................................................................................... 131
When scaling fails: a counter-example of Reality Ends Here ................................................................. 135
A manual… for replication? ............................................................................................................... 140
Investigating localization via fit ......................................................................................................... 142
Comparing incoming students (and their practices) ........................................................................... 148
Summary of local fit opportunities ..................................................................................................... 153
Seeking fit bottom-up ......................................................................................................................... 154
CHAPTER FOUR: PARTOUR IN SOUTH LA, COALITION STORYTELLING &
PARTICIPATION GAPS ......................................................................................................................... 156
Overview ................................................................................................................................................. 156
Confronting negative stories of place ................................................................................................. 158
Mapping with a storytelling parade .................................................................................................... 160
A transmedia campaign emerges ........................................................................................................ 165
Technique: selecting physical locations using “communication asset mapping” ............................... 168
Compatibility as a Situated Game ...................................................................................................... 172
An Atlanta game comparison (Re:Activism) ........................................................................................... 176
Playing Re:Activism ........................................................................................................................... 179
Partially qualifying as a Situated Game .............................................................................................. 186
Missing: the group identity (for fit) .................................................................................................... 187
Missing: local media (for fit) .............................................................................................................. 192
Optimizing non-games via mechanics ..................................................................................................... 196
A thought experiment: distribution as a game .................................................................................... 200
CHAPTER FIVE: CONCLUSIONS ....................................................................................................... 203
Overview ................................................................................................................................................. 203
Data collected ..................................................................................................................................... 206
Models for impact beyond the players ................................................................................................ 209
Essential variance in Situated Games ................................................................................................. 210
xiv

Findings and discussion of fit ................................................................................................................. 215
Scaling to new sites: replication versus adaptation ............................................................................. 215
Analytic power of ‘local fit’ ............................................................................................................... 221
Risks, knowledge gaps, and open questions ............................................................................................ 231
Ethics and empowerment .................................................................................................................... 231
1. Meaningful choices on social impact (for fit#1) ............................................................................. 235
2. Ethical spectacle (for fit#2) ............................................................................................................ 236
3. Ecological accountability (for fit#3) ............................................................................................... 238
Aggravating the participation gap? ..................................................................................................... 239
Participatory design and bottom-up games ......................................................................................... 240
Moment in time: on the cusp of real-world games .................................................................................. 241
BIBLIOGRAPHY ..................................................................................................................................... 245
APPENDICES ........................................................................................................................................... 261
Appendix 1: Macon Money timeline and data sources ........................................................................... 261
Appendix 2: Data for Reality Ends Here and Xposure ........................................................................... 263
Appendix 3: Data and timeline for the ParTour case study .................................................................... 266
Appendix 4: Recipe for participatory mapping with ParTour ................................................................ 270
Appendix 5: Design checklist for making Situated Games ...................................................................... 273
Appendix 6: Game strategies to strengthen the communication ecology ................................................ 281
1
CHAPTER ONE: OVERVIEW

“No other expertise can substitute for locality knowledge in planning… The invention required is
not a device for coordination at the generalized top, but rather an invention to make coordination
possible …in specific and unique localities.”
– Jane Jacobs (1961, p. 418), The Death and Life of Great American Cities

“When information is put at play, game-like experiences replace linear media.”
– Eric Zimmerman (2013), Manifesto for a Ludic Century

“…socio-spatial agglomeration is today being assertively described as the primary cause of
economic development, technological innovation, and cultural creativity…”
– Edward Soja (2010, p. 14), Seeking Spatial Justice

Introduction
The digital world raises new questions for the design of our cities and for civic life. To
sustain engagement, how can local communities deepen social connection across digital and
physical worlds? A new approach is emerging, based on game design with real-world actions.
Weighing the opportunity – especially against costs and ethical risks – is difficult with a
traditional notion of games that train or teach. New models may be needed as games are
increasingly designed to support civic participation.
Games with real-world actions require a framework for structuring participation. Even
before mobile phones, game designers have used play on physical streets to help residents to
connect, move through space, and reimagine the city. Now a technology shift is coming with the
rise of games as pervasive media to bridge digital and physical worlds. Designers are beginning
to investigate play that bridges social networking online with street-level purchasing and
connecting face-to-face.
2

By playing games tied to human networks, participants can build social ties. But can the
community deepen its collective capacity? Pragmatic frameworks are needed to guide the design
and evaluation of such games. Applied research has always been part of good game design, and
methods like network analysis and asset mapping can be particularly useful to link game design
with social science principles. Games that mediate networks and seek collective empowerment
represent a seismic shift for communities, departing from the 1950s critical play of the
situationists and psychogeographers.
1
With real-world actions, the stakes are high. Game
designers need methods to understand how they inherit the inequities of geography, even as they
seek change.
This study offers a new way to understand how game-based activities can deepen local
communities tied to place, including how to evaluate some of the costs and unusual risks.

Two recent games demonstrate some of the possibilities:
• Macon Money (2010) is an alternative currency game to encourage crossing lines of
socio-economic segregation. Printed paper ‘bonds’ served as physical game pieces,
with online tracking. Funded prominently by the Knight Foundation for nearly half a
million dollars, the design points to how a game can serve as social policy to connect
strangers – to one another and to local businesses. An evaluation found that

1
For an excellent art history of artists’ locative games, including the situationists and psychogeographers, see
Critical Play (Flanagan, 2009, Chapter 6). A later section will provide more historical context on emerging game
forms.
3

residents met new people and returned at high rates to locally-owned stores.
2

Simultaneously, the game served as expansionary monetary policy.
• Reality Ends Here (2011) is a game for students from a leading cinema school to
form teams and make multimedia, including 30-second short videos, visual effects, or
longhand notes and movie scripts. Points went to more complex projects, utilizing a
collectible card system. For the university that funded it, the game served as a way to
structure team media production, and serve as a kind of institutional policy to foster
group ties in cycles of collaboration.

Unusual games? Neither is a videogame
3
, at least in the sense of centering on digital
screens. But recent shifts in digital games informed both. Research of online games has shown
that they can have real economies (Castronova et al., 2009), and build social capital online as
third places (Steinkuehler & Williams, 2006). Young people are already helping each other
within videogames (Kahne, Middaugh, & Evans, 2009), and social players may actually have
higher rates of peaceful protest than stereotypical Americans (Stokes & Williams, 2012). The
broader field of game studies is rapidly growing, often with a focus on screen-based media, but
increasingly considering mobile media as well.

2
A post-game survey, for example, showed that 92% of players reported returning to a new business, and 63% met
people they would be “very unlikely” to have met otherwise (M. Taylor & Whatley, 2012b).
3
Ironically, the meteoric rise of videogames in our culture may also be obscuring the vision for alternatives.
Videogames only began to appear in the 1970s (e.g., the launch of Atari in 1972, and the rise of the arcade game
industry circa the 1978 game Space Invaders), with the first deeply social games online spreading in the 1980s with
the rise of the Bulletin Board Systems, followed by more graphical multi-user games in the 1990s. Today, the word
‘game’ in the context of media brings up debates about screen-time and anti-social behavior (e.g., see reports from
the National Institute on Media and the Family, 2008). Imagining alternatives for games may thus be blunted by the
rhetorical dominance of a narrow range of commercial games (Fron, Fullerton, Morie, & Pearce, 2007), deflecting
alternative ideas for what can be “done” with games.
4

Of course, folk games are as old as human culture (Huizinga, 1938). The shift comes
with the explosion in digital technology, and the emergence of new types. For real-world games,
the rise of mobile gaming – especially using cellphones – is particularly important, not for the
tiny mobile screens that go anywhere, but for the power to mediate place-based communication.
Real-world games attempt to get something done, beyond helping people meet face-to-
face. Immediate impact might include raising funds, deepening trust across race and class
divides, or building the social ties that improve quality of life. Immediate impact contrasts with
most civic games, which are educational and preparatory. Educational games sustain learning in
order to transfer knowledge and skills for use in a future context. But for real-world games, the
goal is to have a real impact now.
Yet there is currently a lack of predictive theory for real-world games, especially to
deepen networks. As will be discussed, prior work provides some elements, including theories
of pervasive games (Montola, Stenros, & Waern, 2009), locative games in art (Flanagan, 2009,
Chapter 6), ubiquitous play and alternate reality games (Björk, Holopainen, Ljungstrand, &
Mandryk, 2002; McGonigal, 2011a).
What is most lacking for civic games is a focus on collective impact. Few games have
any idea how to target community health, or network cohesion. (Later this chapter will detail
several leading notions of community cohesion and health, including the capacity to protest.
4
)

4
Communities have collective efficacy when they can organize as a group around community issues -- e.g., to
remove a corrupt official from office, or to restore a neighborhood park, or to establish zoning policy for box retail
outlets. As a preview, strong communities are those with the capacity to overcome adversity (e.g., resiliency to
natural disasters) and get things done as a group. For more on this perspective, a good overview can be found in
Robert Sampson’s analysis of Chicago neighborhoods (2012), and later in this chapter.

5

Network effects are at the center of urban life – for good and ill. Economic growth, for example,
is often driven by regional clusters, like the tech growth of Silicon Valley. Meanwhile, the deep
problems of recurrent urban poverty can also be traced to place-based networks and
“neighborhood effects” (Sampson, 2012). Rather than pick a “best” model for network growth,
this dissertation argues that designers must be strategic and select the most appropriate network
strategy for their theory of community change and development.
Must it be a ‘game’? Formalist definitions of a game
5
may actually be an inappropriate
constraint. Formalist games may be a poor fit for the goals of building local networks, where
culture reigns. Real communities may need to integrate with existing nonprofit programming and
funding sources above all else. To complement formalist definitions, more culture-centered
definition may be needed from the beginning. Therefore, this study explicitly counter-balances
with an effort to identify how more minimalist game-based designs might be included under the
same umbrella.
To counterbalance the games above, the third case emerges from a community coalition
rather than professional game developers:
• ParTour (2012) is participatory mapping activity in South Los Angeles that draws on
game principles to sustain engagement. Funded by small investments from
participating nonprofit organizations, the activity shows how lessons from games can

Healthy communities have a Strong communities are resilient to Given the profusion of models and indicators for
healthy communities, this dissertation is deliberate , strong communities are those who can self-organize to address
issues of shared concern
5
For a good overview, see the landmark textbook on game design Rules of Play (Salen & Zimmerman, 2004).
6

provide a coalition strategy to build and rebrand their community as a
neighborhood that is safe, healthy and innovative.

Are games at the heart of optimizing these three cases? This study proposes and defines
a framework for Locally Situated Games (abbreviated LSGs, henceforth “Situated Games”). Of
course, Situated Games are not magic bullets. In fact, their design is extraordinarily ambitious
even when the goals are focused; many will fail. The goal is to serve policymakers seeking to
evaluate which games deserve funding, alongside designers that need actionable theory. In other
words, the goal is to provide a pragmatic framework that could limit risk in several dimensions,
and tie to established theory to improve the odds of success.
To begin, this study next takes an empirical look into how Situated Games function in
practice. Using the three cases named above, this study investigates how success of such games
can depend on ‘local fit’, especially how the game integrates with regional networks, available
communication modes locally, and the community’s history and identity. (Three dimensions of
‘local fit’ will be defined.) Premature hypothesis testing of fit is not warranted, in part because
the genre is still fluid; and more importantly because a laboratory-like set of controls would
preclude the exact sort of social complexity that is at the heart of neighborhood dynamics.
Therefore, this study seeks to find a pragmatic framework to increase the odds of success by
asking the right questions for optimizing impact.
For each case, several assertions are concretely investigated to validate the umbrella
framework. For analysis, an unusually inter-disciplinary approach is needed. The lens of
communication is seized, first and foremost because communication is at the center of forming
7

interpersonal networks, including with communication technology. Second, communication is
useful as a way to look across disciplines and unite disparate levels of analysis
6
. The larger goal
is to look beyond the cases to understand how ‘local fit’ can be achieved in games that seek to
build networks and empower the local community.
Defining success in a network mode
Which games?
Social games and play have long been studied by anthropologists for their ability to foster
social ties, going back to the work of Brian Sutton-Smith in the early 20
th
century, which
demonstrated the development of community, group identity, and a sense of belonging
(Flanagan, 2009). Educational research has a long tradition with games as well, from Lev
Vygotsky on child development in the 1930s, to the cultural anthropology of Clifford Geertz
analyzing games in Balinese culture, to the emancipatory learning in the role-play games of
Paulo Friere and Augusto Boal in the 1960s and 70s.
7

Can games create a social environment, and spur social ties? Psychologists like Jean
Piaget have argued that games have their own social contract, where players quickly learn that
the space of games is about collective negotiation (in Salen & Zimmerman, 2004, p. 489).
Games can provide players with an “alibi for intimacy,” since players can excuse their
socializing as “just a game” (Montola et al., 2009, p. 11; Poremba, 2007). In other words, games

6
Key levels for this analysis include: the neighborhood-level sociology of communication, the group-level
communication of civic associations, and the individual-level as residents greet a newcomer to the community
7
More specifically, consider Mind in Society (Vygotsky, 1978), Theatre of the Oppressed and Pedagogy of the
Oppressed (Boal, 1979; Freire, 1968), or "Deep Play: Notes on the Balinese Cockfight" (Geertz, 1973).
8

create a temporary “magic circle” where different social rules apply (Huizinga, 1938; Salen &
Zimmerman, 2004, p. 94). Within the magic circle, social failure can be more acceptable, and
social risk-taking can be encouraged.
What distinguishes real-world games is partly how they blur the magic circle, and extend
the temporary world into the real world. Montola et al. (2009) point to several important
expansions: (1) Spatial blur: the game can take place anywhere, not just within the typical
boundaries of games, e.g., the rectangle of a basketball court; (2) Temporal blur: the game can
run for hours or months, and it may even be uncertain to players whether a social interaction was
part of the game at all; (3) Social blur: outsiders and non-players
8
may be involved, sometimes
without their knowledge, akin to Boal’s (1979) political dramas that are performed in public
space and masquerade as reality.
Often the sociality around games is somewhat hidden. In everyday settings, the role of
games may only be visible when looking at the social structures around the game. In Robert
Putnam’s landmark study of social capital (2000), it was only when the game of bowling was
tied to broader leagues that the civic implications stood out.
What happens when the game takes on a local focus? Place-based games have a
particular history in urban planning. Gaming for social science purposes first emerged strongly
in the 1960s and 1970s, with ties to the Model Cities program (Duke, 1974; Light, 2008; Schirra,
2013). Much of this early work was carefully conceptualized, with notions about games often

8
Unaware participants are generally not considered players, since they lack voluntary participation and are not
making choices in the game, or even under a playful frame.
9

featuring role-play and face-to-face interaction; at that point, digital facilitation was not possible
and the landscape of social media had not yet emerged.
Advances in digital and mobile media in recent years have allowed games in the city
planning tradition to explore immersive local planning and augmented deliberation (Gordon,
Schirra, & Hollander, 2011; Lerner, 2014; Mathews & Holden, 2011). Explicit training games
have also emerged, such as the Doorknocking Game
9
, to train community organizers with the
skills and confidence for action.
Practitioner movements have gained momentum under names like Serious Games and
Games for Change.
10
Yet such games movements are largely focused on training, learning and
rational planning – not actions to immediately structure participation, let alone to build collective
cohesion. Even civic games of “collective intelligence” rarely empower the group as a network;
more often, they are mobilizing a crowd for labor (aka “crowd-sourcing”), an approach that is
particularly vulnerable to critiques of exploitation as ‘playbor’
11
.
Research into network formation through games has focused on contexts that are far on
the digital end of the spectrum, such as massively multiplayer games and virtual worlds.
12
This

9
The Doorknocking Game (released in 2006) is also called the Organizing Toolkit, and was created by Kinection,
the Center for Third World Organizing, with funding from the Waitt Foundation and others; see the Civic Tripod
(Ruiz, Stokes, & Watson, 2012, sec. http://civictripod.com/games/doorknocking/).
10
Serious Games is an awkward but important catch-all for “games used for purposes other than mere entertainment”
(Susi, Johannesson, & Backlund, 2007) which grew under Ben Sawyer and David Rejeski in the early 2000s.
Games for Change emerged in 2004 with a more specific focus on social issues and civil society.
11
Playbor (play/labor) is a line of critique aligned with what Scholz calls “cognitive capitalism” that exploits players
for their labors; for a good overview, a useful edited volume is Digital Labor (Scholz, 2013).
12
In particular, much of the work has focused in virtual worlds and massively multiplayer games like World of
Warcraft released in 2004 (for example, see Galarneau, 2005; Zhu, Huang, & Contractor, 2013). On one hand, such
digital worlds are increasingly recognized as authentic and meaningful domains of human activity in their own right.
They may even be meaningful spaces for direct civic action (Thomas & Brown, 2009). However, research on team
formation inside digital games is often considered distinct from real-world group formation (Ahmad, Borbora, Shen,
Srivastava, & Williams, 2011).
10

includes work on how digital games can foster the “associational life” described by Alexis de
Tocqueville (Schulzke, 2011). Most network research with games emphasizes global networks
united by shared interests. Interest-driven learning has an increasingly important place in a
digital world (e.g., see the ethnography Ito et al., 2009), but can undermine a focus on place-
based community. In fact, much of the most prominent work on network benefits from civic
games celebrates an “anywhere/anytime” approach (e.g., on protein folding Khatib et al., 2011),
which runs counter to need for local networks to embrace specific geography to anchor the
community.
‘Gamifying’ real-world behavior is controversial (Raczkowski, 2013). Perhaps most
visible is the movement for ‘gamification’ (Deterding, Khaled, Nacke, & Dixon, 2011), which
includes real-world shopping apps. Investment is at least $10 million annually.
13
The techniques
often do not sound like games as they have been described in the scholarly traditions of
anthropology, education or game studies; rather they are described as behavioral incentives. For
example, a Pew Internet report on the “Future of Gamification” described them as designs that
tap into “people’s competitive instincts and often incorporates the use of rewards to drive action”
(J. Anderson & Rainie, 2012, sec. 1).
Gamification backlash comes in part from a perception of profit-driven motivations to
boost marketing and even purchases, with little concern over behavioral manipulation. Counter-
labels like ‘exploitationware’ and ‘pointsification’ argue that the marketing approaches are
shallow and manipulative shadows of games (Bogost, 2011; Robertson, 2010).

13
For example, at least one venture analysis (Zichermann, 2010) reported that 2010 saw $35 million in seed capital
invested in gamification-centric startups and businesses betting on gamification as a core strategy; this is less than 7%
of all investments in gaming during this time, but does indicate some of the profit-driven interest.
11

Although gamification often lacks a solid theoretical basis – it is more a list of attributes
like the use of badges and points – the approach is also spilling over beyond marketing.
Applications are appearing in education (Lee & Hammer, 2011), social networking
14
, and even
local community engagement (Coulson, 2014). A few brave theorists have attempted to give an
alternate future for gamification that respects participants and gives them more meaningful
choices (McGonigal, 2011b; Nicholson, 2012). As yet there appear no frameworks for
gamification that define meaningful choices in terms of local networks.
The games of this study can be summarized as distinct along two axes. (1) In terms of
timing, the games of this study have immediate real-world impact, unlike training games which
emphasize an impact in the future; (2) In terms of empowerment, the games of this study
primarily emphasize the group-level, like collective strength and network cohesion, more than
empowering individuals in isolation. A visual of the two axes is presented in Figure 1.

14
For example, services like Klout and PeerIndex seek to provide a measure of social network “influence” –
although there is debate as to whether their results more accurately indicate online status or perhaps social capital.
Much of this debate is taking place in public blogs, e.g., http://wallblog.co.uk/2012/03/27/brian-solis-says-klout-
does-not-measure-influence-he-is-right-and-wrong/
12

Figure 1: Differentiating games for real-world cohesion and community

Source: based on a graphic previously published by
the author (in Ruiz, Stokes, & Watson, 2012, sec. Reframing Learning)

The field of game studies in the academy is quite new and still fragmented. Professional
conferences are dedicating tracks to games, games conferences are arising on their own, and new
journals are appearing quickly.
15
Now is a valuable moment for consolidation across domains,
especially linking games more explicitly to theories of place-based communities and local
network mechanics.

15
Prominent academic conferences that have featured games tracks in recent years include AERA in Education and
ICA in Communication. More autonomously, the past decade has seen games carve out space on their own,
including conferences that focus exclusively on games (including DiGRA, GLS, G4C, IndieCade, GDC), textbooks
on games that draw from multiple fields (most notably Salen & Zimmerman, 2004), and even dedicated journals like
Sage’s 2006 launch of Games and Culture.
Group-Level Goals
(e.g., Community Strength or Quality)
Goals of Individual-Level Impact
Preparation
for Future Action
Immediate
Real-world
Impact
• Campaign training
for groups
(e.g., Door Knocking)
• Real-world games for
cohesion & community
improvement
• Social issue games
(individuals learn
about civic issues)
• Real-world
shopping apps
(e.g., FourSquare)
13

Community coherence, place and network methods
There is growing enthusiasm to consider “place-based” strategies for community change
and economic growth. Such strategies go beyond reaching out to neighborhoods – they insist
that solutions must be customized or even emerge from localities. The visible drivers of place-
based movements include the field of health (with U.S. foundations like the California
Endowment), civic engagement and journalism (like the Knight Foundation), and economic
development (with proponents like Richard Florida (2003)). For these movements, the rise of
locative media increasingly forces the question: is place-based worth the trouble?
Technology shifts have made the question of place suddenly relevant again. For
example, locative media
16
on mobile devices is the first popular communication form to be
sensitive to the user’s location in physical space. More broadly, mobile devices are bringing the
global Internet into local neighborhood space as a flow of information and culture. The shift is
thus in both directions, as mobile media localizes and globalizes participation (Castells,
Fernandez-Ardevol, Qiu, & Sey, 2006; Gordon & de Souza e Silva, 2011; Ito, 2005).
17
As a
result of this blending, mobile remediates physical space and time (Ling & Campbell, 2009),
changing how we experience neighborhoods and connect with each other.

16
The terms “locative media” and “locative arts” emerged with coherence around 2003 (Hemment, 2006). The
media emphasis extends a longer history of ubiquitous and pervasive computing.
17
For example, mobile devices can increasingly disclose who is talking in a nearby tattoo parlor, which of our
friends are nearby, and what issues are being discussed by a local school board. At the same time, mobile can also
pull us out of physical space, asking for donations in Haiti when disaster hits, or when family calls mobile can pull
us out of local conversations to talk across the country.
14

Community lost and injustice that clusters by neighborhood
Community building – including collaborative and more confrontational tactics like
protest – is often set against implicit fears of isolation and community lost. Fears about local
cohesion are often summarized in the community lost thesis
18
, which posits that the social ties of
urbanities are becoming segmented, transitory and impersonal – accelerating a process of
globalization that is eclipsing local community (Sampson, 2012, p. 4). As Robert Sampson
warns, narratives of “community lost” are perennial and sentimental; community lost is a
recurring fear that has resonance in “every historical period” (2012, p. 4).
Many place-based communities are anchored in a neighborhood,
19
a term which is
explicitly linked to physical place. Neighborhoods are sometimes measured in strict geography,
but increasingly they are recognized in social terms too; sociologists like Sampson going as far
as to claim that the relational aspects of neighborhoods are how they are best defined (p. 55).
Emphasizing ‘place’ helps to push beyond technology-driven measures of geography.
Databases often delimit neighborhoods in raw longitude and latitude, which are easily quantified.
But for building community and defining neighborhoods, it is dangerous to reduce place to
coordinates. Henri Lefebvre’s work insisted that space is created largely by the imagination, and
that community change depends on imagining space (1974, p. 251). A similar case around
national identity was made a decade later by Benedict Anderson (1983), who sought to explain

18
Although the term “community lost” is used here describe a more longstanding narrative, this research also draws
on the sociological theory of the same name that was developed in the late 1800s to describe how urbanites could
(supposedly) no longer rely on the community for their needs, since their networks were weak.
19
Neighborhood can be defined as a “geographic section of a larger community or region (e.g., city) that usually
contains residents or institutions and has socially distinctive characteristics” (Sampson, 2012, p. 56), following
Sampson’s decades-long work in Chicago neighborhoods.
15

the ‘anomaly’ of nationalism as a powerful horizontal fraternity that transcends pure
geographical boundaries. Anderson’s notion of “imagined community” has been applied to
nation states, but increasingly it is also applied to much smaller communities – especially where
participants are unlikely to known each other personally, yet can be drawn into action at the
invocation of their community.
Place is especially important for certain social goals. The idea that neighborhoods affect
social indicators is often called “neighborhood effects” research, arguing that neighborhoods
matter deeply.
20
Beginning in the 1940s, sociologists argued that neighborhoods possess
“relatively enduring features” that can be linked to human behavior (Sampson, 2012, p. 37). By
the dawn of the 21
st
century, there was a “massive literature” on neighborhood effects.
21
The
literature can be summarized in two main points:
1. that there is fundamental inequality between neighborhoods; and that these
differences coincide with geography; specifically, that differences in health and
crime are tied to neighborhood-level conditions, especially poverty;
2. the converse: that affluence also clusters by geography.
Since neighborhoods and place matter deeply for equality, place should also be central to
community-building strategies that claim to focus on equity, including real-world games.

20
There are long roots to this research, which can be traced to André-Michel Guerry’s 1833 analysis of how crime
follows geography, reinforced by John Snow’s influential mapping of the health outcomes with the cholera outbreak
in 1850s London. In the 1920s with the rise of urban sociology the dynamics of urban living were more
systematically theorized. By 1942 some of the first neighborhood characteristics were being used to track juvenile
delinquency by Robert Park and Ernest Burgess (cited in Sampson, 2012, p. 35).
21
As documented through the literature reviews of Sampson (2012, p. 46)
16

Flipside: economic development via place-based networks
City efforts at growth and development have also taken a spatial turn in the past decade,
with implications for local social justice organizing as the city seeks to make its region appear
distinctive and attractive. As Edward Soja defines it, growth strategies in cities are increasingly
organized on the basis that “socio-spatial agglomeration” is the “primary cause of economic
development, technological innovation, and cultural creativity” (2010, p. 14).
At the city level, the hope of “agglomeration” effects for economies can be traced
intellectually to Jane Jacobs (1970); in The Economy of Cities, Jacobs argued that agglomeration
is the spark of urban economic life. Yet practical training is quite recent. Only in the past
decade have graduate school textbooks begun to train city managers to consider local efficiencies
and interdependencies that could spur innovation and development (Blakely & Leigh, 2010).
Often the desired efficiencies address material resources, although increasingly cities are
also following the human capital analysis of Richard Florida (2002), who claims that economic
development in cities depends on recruiting the right kind of human “talent” to cluster
geographically. Sometimes this work is done by city officials, but neighborhood-level coalitions
of business and civil society leaders can also help identify and promote a local business cluster,
such as the high-end art market that has grown in Culver City, Los Angeles.
22

22
Based on correspondence and conversations with Elizabeth Currid-Halkett, following her research on cultural
capital in Los Angeles and New York (e.g., Elizabeth Currid-Halkett & Stolarick, 2010).
17

Network paradigm
For games that seek to strengthen human networks, a powerful perspective comes from
the rise of network science. Networks can be strengthened in very different ways, from raw
growth to cohesion and bridging ethnic groups. Picking the right network goals to match the
game is critical, and network methods and perspectives can be essential for strategy.
Network graphs (see Figure 2) are the visual anchor for a highly interdisciplinary domain,
which emphasizes relationships between people – rather than the capacities of individuals.
23
The
paradigm is particularly useful when analyzing activities with known participants, such as a
game system that tracks all players. Statistical techniques are often used alongside more
qualitative interpretation of the network visualizations. Perhaps most importantly, the network
perspective is one way to push for group-level indicators, like collective cohesion.

Figure 2: A sample network graph

Source: GUESS software screenshot
24

Social capital is a common language to describe network benefits (most prominently by
Putnam, 2000). Individuals have social capital in proportion to the strength and size of their

23
The network perspective draws on several fields, including group structure from sociology, graph theory from
mathematics, and data mining and information visualization from computer science. For an overview from a
communication perspective tied to health, see Valente (2010, Chapter 1).
24
User DarwinPeacock via Wikipedia
18

social ties, e.g., in terms of trust and reciprocity for interpersonal relations. Social capital
extends a long tradition of analyzing the value of networks for individuals, including for
education (tracing to LJ Hanifan in 1916). In sociology, the conceptual roots trace to Pierre
Bourdieu, well before Putnam investigated civic decline using social capital measures.
Yet social capital has several limitations. Measures of social capital vary widely, and
very few of them can accommodate a sense of place. (As a community measure social capital is
typically modeled as the average of people in that geography.
25
) More egregiously, social capital
celebrates the group as valuable to individuals, with few insights into group strength and
community coherence.
Fortunately, the network paradigm is broader than social capital, and network analysis
may be more immediately useful with games. For example, players in a social game might be
given feedback about the amount of clustering in their group, or whether network ties are
generally mutual (e.g., reciprocity), or how hierarchical versus ‘centralized’ the network is. So
far, few games have been designed using network indicators; in the future, there may be entire
game design practices around optimizing for network indicators. More immediately, network
analysis can already be applied to many existing games, as is demonstrated in Chapter 2 of this
study.
Concretely, how can games strengthen networks? Several possibilities are worth
highlighting: (1) More ties: The most obvious way is to increase the number of ties, so that the
typical resident is connected to more people. (2) Deeper ties: Of course, deeper relationships

25
Such averaging is increasingly problematic in a globalizing world, where a local resident might have high social
capital, but not consider themselves members of the local community, and therefore be unmotivated to contribute to
local place building.
19

also matter – and by bringing people into repeated interaction games can strengthen the tie
between two people. (3) Bridging groups: A third way is that games can help to bridge groups,
such as isolated ethnic enclaves, by first identifying groups and then encouraging cross-group
ties. (4) Key people: network methods in games can identify individuals who are already
serving as key bridges between groups, and shift activity in their direction, such as to help funnel
conversation and resources to them. This is just the beginning – there are many more, but these
few help to establish the terms of engagement.
Conceptually, the network paradigm is a useful bridge between the broad notion of
‘community’ and the quantitative models of most digital games. In the past, the gap has been
large between the quantitative code behind online games and the face-to-face tactics of building
community (like hosting a neighborhood concert). However, as the gap narrows with mobile
media and hybrid places, the network paradigm points to how game designers can create live
models that are simultaneously concrete and aligned with informal face-to-face modes of
building community. Although civil society has long relied on networks for its organizing,
applied handbooks are just beginning to circulate broadly with network techniques (e.g., Plastrik
& Taylor, 2006); such applied handbooks have not yet considered digital games.
Which terms: place-based community and networks
Many of the terms above are ambiguous. For this study, some clarity is needed – but
with caution. Perhaps surprisingly, the embrace of ambiguous terms like ‘building community’
20

can foster a useful ambiguity across diverse academic and professional fields.
26
Of course
empirical research necessitates specificity for measurement, and case study analysis must often
align with field-specific language. Otherwise, the following terms summarize the theory
arguments made above.
In general, a specific kind of community is presumed: place-based communities. Such
communities are unified, in part, by a shared experience and social construction of physical
geography. Building community often includes fostering a sense of belonging, and the belief in
the group’s ability to get things done (i.e., collective self-efficacy; more on this later).
Community members benefit by knowing some other members, but do not need to know
everyone.
The complementary language of networks is used in this study to emphasize more micro-
level actions and interpersonal ties. Networks are only formed when people meet or begin
communicating. Thus network formation points to the social actions people take, like meeting
strangers and greeting neighbors. When social networking is tied to place (in what this study
calls local networks
27
), the place-based community can benefit. Although meeting is typically
voluntary, a benefit of place-based communities is that geography can push people into
proximity and even force some involuntary network formation.

26
‘Community’ is frequently the most accessible (and sometimes only) modifier to emphasize the collective and
local simultaneously – from strategy to entities and measures – for example: community empowerment, community
change, community organizing, community resilience, community-based organizations (CBOs), etc.
27
‘Local networks’ serves as shorthand for networks that are relatively localized by geography or tied to a concept
of place; the terminology “community networks” is generally avoided as confusing here since the two are defined
somewhat in opposition; of course communities encompass numerous local networks.
21

Activities (not information) to structure social interaction
To intentionally form and strengthen networks, social activity is needed. Approaching
games as social ‘activities’ is a useful departure from the tradition of studying games as mass
media using theories of media effects. As activities, games can be understood as helping to
mediate participation – not just mediating flows of information and messages.
Games can thus be defined as activities that structure playful participation. Often, games
structure participation by deliberately presenting challenges.
28
In contrast to the design
philosophy for tools (which minimize challenges in order to accomplish tasks with maximum
efficiency), the design approach for games calibrates challenges to the player’s ability. No
challenge is a poor game; real-world challenges mean that players have a chance to get better at
the skills of civic life. The focus on appropriate challenge means that games are better
understood as cultivating engagement, not ‘fun’.
29

To detect the impact of a game centered on social connection, the ‘game’ itself may need
to be defined broadly. Social connection is notoriously hard to measure, let alone to measure in
real-time with game rules and points. So if we only consider participation in terms of the rules
and points, the social meaning can easily be overlooked. To describe the broader social picture,
game theorists have emphasized studying the ‘ecology of the game’ (Salen, 2008), as well as the
‘big-G Game’ (Gee & Hayes, 2010).

28
The participation or process that a game supports is often described as an “artificial challenge” or conflict,
according to Salen and Zimmerman (2004, p. 80), based on the premise that all games embody a contest of powers,
even if that conflict is with the system, as when a team must cooperate to overcome an obstacle.
29
Although games are popularly described as ‘fun’, they are better understood as being engaging. In fact, the
assumption that games are ‘fun’ is misleading, since games that sustain participation regularly cause enormous
frustration – from tennis to chess.
22

Outside of game studies, a useful way to define the game-in-context is Activity Theory
(Nardi, 1996). The theory is more of a set of perspectives and terms to look beyond technical
artifacts to understand the big picture as an activity system.
30
The benefit is the ability to
articulate how a game “mediates an activity, thereby connecting a human being, not only to the
world of objects – his or her physical surroundings – but also to other human beings” (echoing
Fjeld et al., 2002, p. 155). Throughout this study of real-world games, a similar effort is made to
describe games as part of social activity systems, often with distinctly local characteristics.
The value of taking a broad social view can first be verified with non-digital games.
Again, recall the famous bowling leagues of Robert Putnam (2000): rolling of the bowling ball
and tracking points was not where the socializing happened. In other words, Putnam never
claimed that civic life emerged from the act of rolling the bowling ball. Rather, the vital social
structure tied to bowling is in the conversations between and behind games, as part of the
sustaining community of play – and in how the bowling leagues pushed strangers and friends to
interact. Thus it is important to quickly define ‘game’ broadly, to look for social dynamics at the
edges of the game, and to ask how the broader activity system is interfacing with the local
community.
Proposing ‘Locally Situated Games’
For a deeper understanding of real-world games, it is now necessary to define their type
more precisely. Collections of real-world games exist (e.g., see the collection in Ruiz et al.,

30
Activity theory has been applied in a number of interdisciplinary settings including Human-Computer Interaction
(HCI) and Education. Its roots are in Soviet cultural-historical psychology (especially Vygotsky and Leont’ev).
23

2012), but as this study will argue, they are more often based on a technological criteria (like
“mobile games,” or being ‘pervasive’). Categorizing by more than technology is necessary to
studying how games can build local cohesion.
One contribution of this research is to provide a framework for a type (or analytic
category) of games: those that seek to build local networks and community cohesion. The
problem is hard, in part because games that target local networks must operate at so many levels
simultaneously: as technologies, as cultural objects, as structures for conversation, as street
theater – just to name a few.
To be rigorous and coherent, the category definition must exclude some games – and
leave only the ones that share fundamental properties. The goal is to develop a pragmatic
conceptual filter, based on similar underlying mechanisms (not technology), and thus allow for
coherent study and optimization. The conceptual filter is like a children’s toy box, admitting
only games with the right conceptual ‘shape’ for controlled analysis (see Figure 3).

Figure 3: Block-matching as conceptual filter

Source: a children’s toy advertisement (screenshot from DHGate)
24

Definition and criteria for inclusion
Locally Situated Games (LSGs) are defined as game-based structures for real-world
participation that seek to strengthen a place-based community. These activities integrate
principles of game-based engagement with place-based communities and building network
cohesion.
The conceptual filter for Situated Games only includes designs that have the following
properties: (1) real-world actions as part of the activity; (2) the design is game-based; and (3) has
local network building goals. Each is explained in more detail below. The filter immediately
excludes most educational videogames (which are single player simulations on screens), as well
as most civic activities (which only occasionally encourage playfulness within civic action).
31

In more detail:
Real-world actions: To qualify, games must involve real-world actions, not just
simulations. In general, progress in the game is congruent with affecting the world.
They collapse the frames of “training” and “doing.” If the game has learning goals, they
should follow Gordon’s suggestion that the learning be “not for some future action, but
for the now.”
32
Social action has long-term benefits, and predicts civic participation.
33

31
Exclusion is a key property of any pragmatic definition. Situated Games are different from (a) other games –
since many games are either screen-based or board-based but not both; even location-based games rarely seek
network effects; and from (b) other community-building programs – which rarely combine playful activities with
their larger designs – such as for business improvement districts, or community news organizations.
32
See interview with Eric Gordon, by the author of this study: http://civictripod.com/interview-eric-gordon/
33
For example, research on political engagement has shown that being a member of extracurricular groups is more
predictive of voting and volunteering than knowledge of civic facts (Hart, Donnelly, Youniss, & Atkins, 2007).
25

Game-based. A game-based
34
design is one that structures participation through playful
challenges
35
, feedback loops and uncertain outcomes.
36
In civic terms, this definition contrasts
with many of our stereotypes about civic activities that are filled with a sense of duty and
predictable rules of order. Yet it is also a constraint, since games are typically poor at many
things, like disseminating text-heavy messages.

Network-building goals, especially tied to place-based communities. Participating in the game is
social and seeks to build network connections for collective benefit, such as increasing social
capital, or strengthening the group identity, or deepening relationships across zip codes. The
network scale emphasized in this study is generally much smaller than social movements for
macro-level social change; often the “neighborhood” is an ideal scale – smaller than a metropolis
but larger than a few city blocks.
Research Approach
The premise of this study is that the success of a Situated Game may depend on its “fit”
to the local community. If the fit is right, a game is expected to be better able to help local
networks to strengthen or grow, and thus have community impact. Similarly, a poor fit is
hypothesized to undermine the potential of a game design. Thus the central question for this
study is simply, how does ‘local fit’ matter for success? (Again, success is primarily
considered in terms of human networks and collective capacity, including collaboration – but
also more confrontational forms of coordination like civic protest that invokes the public good.)

34
By emphasizing game-based, this definition is deliberately less formalist. There is currently a messy proliferation
of projects claiming to be games that have some game features, yet fail formalist tests. Analytically, part of the
contribution of this dissertation framework is answer the hunger for clarity about which game features are
hypothesized to be important, by answering for Locally Situated Games in particular. Fewer constraints apply in the
case of game-based designs. One constraint in particular was loosened; now game-like can include activities that do
not have a clear winner or strict “win condition.”
35
Good games are often profoundly challenging and can demand intense work, even commercial games that are
purely for entertainment. Part of what makes the setbacks in games palatable and even appreciated, as compared
with everyday life beyond the “magic circle” that encompasses the game (Huizinga, 1938) is that failure in games is
productive. Productive failure is often the best way to develop insights into the game, and thus improves the odds
that the player will do better next time (Gee, 2005; Juul, 2013). Getting better at something is at the heart of
empowerment. Thus the productive failure of games can be linked to empowering the community.
36
This definition owes a particular debt to the framing work of Salen and Zimmerman (2004) and Fullerton (2004).
26

Fit as conceptual framework
A conceptual framework is proposed to begin rigorously linking the “how of games” to
the “how of communities.” Specifically, I am proposing ‘local fit’ as a conceptual schema or
framework
37
, with three dimensions. The dimensions are detailed below; in brief, they
emphasize fitting to the local modes of meeting and greeting, to the community’s distinct
identity (and history), and to the local communication ecology. The conceptual framework of
‘local fit’ will be used with the more rigid definition of Situated Games to: (a) differentiate what
exactly is new with these games, and how they can be compared; (b) to empirically describe
some of the community impact that has been achieved, including by comparison with non-game
strategies; and finally (c) to provide a framework for optimizing future designs for impact.
Throughout this dissertation, ‘local fit’ in single quotes indicates the specific fit detailed below.
Three dimensions of ‘local fit’
To conceptualize an umbrella notion of fit, three lenses are proposed. Each dimension
(or lens) emerges from the theoretical background above. The first addresses game actions,
since games are distinct for how they structure participation by requiring players to take action.
A second lens considers the collective identity of the local community. A third lens considers
neighborhood communication flows, and how the game integrates holistically with local

37
I deliberately avoid seeking a single “best” measure of local fit, given the profound interdisciplinarity needed to
make games, and the many ways to measure ‘community’. Rather, my goal is to find a schema that can be
somewhat more universal, providing conceptual architecture for optimizing and studying this class of games. The
use of flexible schemas alongside tight definitions can be traced to Plato, and schemas have also been used by game
and cognitive theorists as a powerful way to help the human mind (including researcher minds) to make sense of
concepts (Salen & Zimmerman, 2004, p. 102).
27

organizations and community-focused media. Subsequent chapters of this dissertation will
further explicate each dimension in turn.
Specifically, ‘local fit’ according to this theoretical framework must include:

1. Mechanics: Fit of the game mechanics to local modes of meeting and greeting, especially the
actions that local residents often take to find one another and grow the place-based community.
In brief, a game can be designed to emphasize the connecting and meeting actions that are already
established in the community. The term ‘mechanic’ is used in game studies to describe a core
action that participants can take (often expressed as a verb).
38
For example, a core action in the
game Hide and Seek is to “find people” who are hidden. A good fit with a local community
might be with a church congregation and its membership network – especially during the holidays
when newcomers attend services but hide in the back rows, and the church would like to find and
engage these participants. In general, human networks have established mechanisms for outreach
and growth, like meeting new arrivals at a block party, greeting neighbors when you encounter
them on the street, or gathering angry residents around an injustice. One justification for “fit”
with mechanics is to keep the game simple, since complexity in public space and games alike
tends to dampen participation, and new social mechanics are especially hard to grasp.
Additionally, a mismatch between game and community modes might lead to: (a) little
sustainability beyond the game for ties/relations; (b) resistance from community leaders to
support or repeat the game. The chapter on Macon Money will explore this dimension of fit in
more detail.

38
Game mechanics (or ‘mechanisms’) are how games structure participation – they are what players do. They are
based on the game rules and implicit constraints. Mechanics describe the core processes of a game, and are often
“best described with verbs” (Järvinen, 2008, p. 263; Sicart, 2008). The mechanic(s) is determined partly by the
game rules for what a player is allowed to do in seeking their goals. Additionally, there are implicit constraints that
come from social norms, physical space, etc. Gameplay is often described as emerging from the mechanics.
28

2. Identity: Fit of the game’s performed narratives to the local narratives of community identity
and neighborhood brand. This dimension considers the narratives of group identity, the essential
narratives of “who we are” that make communities coherent and real.
39
Games in public space
enact a kind of visible performance of community that also highlights what the space means, and
what the participants are about. Local organizations tend to only support games or other projects
that resonate with their image for the community.
40
However, conformity should not be
presumed; dissent and deliberation is important to many communities, and many civic
organizations are founded on the basis of challenging the status quo. The limits of acceptable
dissent are part of what defines a community. Residents will also participate only if the game
feels consistent with “who we are.” The risk of a poor fit is akin to cognitive dissonance at the
group level – it is uncomfortable, and risks backlash. Games are like other media and activities,
and can be perceived as telling “our story” or being inappropriate. Games have special properties
in addition, drawing on the tradition of locative games.
41
Real-world games may particularly
need to fit to the community narrative, both to avoid being manipulative with real-world actions,
and to secure buy-in from players who wish to go from being playful to also making a difference.

39
The idea that communities are imagined into being is at the heart of creative place-making; the idea is also at the
heart of theories about neighborhood storytelling networks as necessary for collective action (Kim & Ball-Rokeach,
2006a).
40
Organizations in civil society often depend heavily on their reputation, and through their reputation are held
accountable to the perceived good of the community. This accountability is logically greater in communities with a
strong collective identity.
41
For example, games in public spaces also produce a kind of implicit theater, where bystanders see the commotion
and inquire what is going on. The theater of games is special because it gives longtime residents special permission
to try on different roles, or imagine their neighbors or public space in a different way. Artists’ locative games have
long explored these possibilities, drawing on Guy Debord, Michel de Certeau, and Henri Lefebvre (for an excellent
overview, see Flanagan, 2009, Chapter 6).
29

3. Communication Ecology: Fit of the communication in the game to the local communication
ecology. Communication is often the primary means of forming network ties. It is easy to
communicate in principle, but in practice each community has its own communication ecology –
a set of established channels (infrastructure) and social norms, as theorized by research of the
Metamorphosis Project
42
. For example, some communities rely heavily on organizations like
churches, while others rely on Chinese-language newspapers, and others use online social media.
The consequences of failing to fit into a community’s modes of communication include
minimizing participation, and undermining impact on the broader community. A requirement of
this framework is to look beyond reaching “users” since that is just one level of the network, and
beyond the vagueness of “stakeholders.” Specifically, the communication ecology should include
flows of communication to: (a) organizations with a local focus, such as community churches and
neighborhood nonprofits; and (b) flows to local media, including to journalists focused on
community issues, local media outlets, and online platforms that serve as information hubs for the
local community.

Are the dimensions connected? Not directly: they are conceptually orthogonal, and so
can be evaluated somewhat independently (see Figure 4.)

42
The fabric of communication serves the community as a kind of infrastructure. Scholars can analyze the
communication fabric, much as economists might study how economic infrastructure affects a community, or
political scientists might study how the political infrastructure mediates community participation. Kim and Ball-
Rokeach refer to this approach for studying the community as Communication Infrastructure Theory (CIT); over the
past decade it has been in used and developed in the Metamorphosis Project at the University of Southern California.
See, for example work on the local storytelling network, and how communication infrastructure mediates the
possibility for community storytelling (Ball-Rokeach, Gonzalez, Son, & Kligler-Velenchik, 2012; Ball-Rokeach,
Kim, & Matei, 2001); these concepts will be explored further in Chapter 4.
30

Figure 4: Dimensions of ‘local fit’

One simplistic way to visualize the connection is to tell a metaphoric story about how a
Situated Game works, using the language of the dimensions. (Again, the value of schemas is to
build intuition amidst complexity.) Specifically, imagine that these games build real-world
community in a basic way: through social interactions (fit #1) the community shapes the
narrative of collective identity (fit #2), using the available modes of communication (fit #3). Of
course, the actual process of building community is not linear. But provided the linear story
helps get at the big picture, we can use the visual to conceptualize ‘local fit’ – and thus guide
how games are explored and explained.
A good game might not seek to maximize all dimensions equally. Completely ignoring a
dimension can lead to oversights that alienate local stakeholders, reduce participating and limit
sustainability. Yet for simplicity, a good game will likely emphasize one or two dimensions in
particular. At a minimum, a game should have some basic ‘local fit’ – it should be compatible
Social interaction
Identity

Game mechanics
Performed
narratives
Local ecology
Communication
flows of the game
31

with all three dimensions. As a practical matter, a game might seek to primarily contribute to the
community by making it stronger in one of the dimensions. Similarly, a game that challenges the
community identity might depend for its success on simultaneously deepening local channels of
communication. Thus the goal with the framework of Situated Games is first to give a baseline
for minimum consideration, and then to give concrete examples for how to optimize design along
each of the dimensions. Each case seeks a different pragmatic balance of the dimensions.
In sum, the ‘local fit’ of a Situated Game can be explored and explained in terms of
whether: (1) the game’s actions – known as mechanics – fit to the local modes of social meeting,
especially making new connections; (2) the narratives performed by the game fit to local
narratives of collective identity and neighborhood brand; and (3) the game’s communication
channels fit to the local communication ecology, especially with local organizations and media
focused on community stories.
Case study approach and selection
The approach of this study is to empirically ground the analysis is real-world examples.
In brief, this dissertation will track and model the games and their associated networks.
To triangulate and solidify the proposed framework for Situated Games, three cases are
explored: Macon Money, Reality Ends Here, and Ride South LA. They are embedded in real-
world communities: one is situated in a geographic corridor of Macon, Georgia; one is the
incoming freshmen class at a university; and the last is a group of bike clubs in South L.A. Each
of the following chapters will investigate one case in detail.
Variance between cases is important. Breadth is deliberate, in part to avoid the premature
temptation to promote one “idealized game” for the civic issues in this study. Each game has
32

some specific weaknesses, as with any intervention into local communities given all the socio-
economic and relational complexity. Variance is especially important between specific goals for
bottom-up empowerment (emphasized in the last case study) and the top-down nature of
structured design (emphasized in the first). The cases thus triangulate across civic goals that
particularly rely on strengthening local networks:

Chapter 2: Macon Money. The most top-down approach comes from city planning and policy.
Here the game provides a real-world economic structure, and uses a real-world currency,
addressing socio-economic segregation by seeking to increase flow across zip codes. Games in
this context are a kind of informal socio-economic policy.

Chapter 3: Reality Ends Here. Building professional networks for collaboration, belonging and
institutional affiliation. Games in this context are a kind of social structuring, especially to dive
deeper into collaborative work in small groups and teams.

Chapter 3: ParTour in South LA. Building neighborhood coalitions to overcome stereotypes and
fragmentation. Game-like activity in this context is particularly informal, but play is central;
however, frameworks of game design are shown to be useful as a means to optimize for how the
group can deepen the collective identity and communication channels.

The final chapter of this study looks back across the three cases, reflecting on the value
and validity of an umbrella framework for ‘local fit’.
33

Data sources and analytic methods
The study analyzes each case with a different set of tools to best inform the overall
framework and understand the case. The methods and data for each case vary, as will be detailed
in each chapter. A highlight of methods includes:

• Behavior and attitude shifts according to regression analysis of pre/post surveys of
participants in the Macon Money case
43

• With traditional network analysis in the case of Reality Ends Here using UCINet
software, including tracking the emerging network over time each year, and comparisons
across three years of data and two schools of implementation
• Participant observation in both Reality Ends Here and ParTour South LA. This informs a
qualitative structural analysis of cases which leads to an articulation of their design
principles.
• Interviews with designers: beyond the data for each game showing impact, I conduct a
series of interviews with the lead designers of each case study, investigating how they
linked game mechanics with the desired effects on the network.
Analytic lens: communication as fabric, and games as activities
The metaphor of communication as fabric is useful to help picture how the little threads
of everyday conversation can turn into a garment of collective substance. Communities are

43
The quantitative data for Macon Money is based on data and analysis by Network Impact which I advised in some
capacity, as discussed in that chapter; my own qualitative interviews provided a longer-term perspective from
stakeholders.
34

complex social creations, and so it is vital to adopt a broad perspective that includes how the
game overlaps with the existing architecture and community. The lens of communication is one
such perspective.
The communication lens is due in part to the author’s training at a School of
Communication. Yet communicative acts are also what shape community bonds and collective
identity
44
. From the perspective of community change, a community’s infrastructure for
communication is necessary for any group mobilization (Kim & Ball-Rokeach, 2006a).
At the center of each game analyzed in this study are acts of communication by players.
Tracking participation as communicative acts is often the easiest way to build a corpus of data
for analysis. Communication also helps maintain a focus on people, not just game structures and
incentives. Both are vital, and an effective analysis of games depends on both addressing their
systemic qualities without losing track of their playful center in human sociality.
Limitations
This research is not a double-blind experiment. It does not isolate and reliably predict the
network effects of specific game features. Rather, it seeks to shift the understanding of several
leading games, arguing that their success depends on network effects and ‘local fit’. The goal is
to provide a theoretical framework to unify the study of Locally Situated Games, and to argue
why this framework is necessary to study and optimize their effects in the future.

44
Even scholars outside the field of Communication have implicitly adopted this position, from political scientists
like Robert Putnam (e.g., see Bowling Alone, 2000), to neighborhood sociologists like Robert Sampson (e.g., see
2012). Of course, within the field of Communication, many prominent scholars have reached out to bridge the field
to other domains.
35

Future research will be needed to investigate a number of the implications that arise in
the three case studies. The final chapter will discuss several of these implications in detail, and
point to remaining areas of uncertainty for Situated Games.

36

CHAPTER TWO: MACON MONEY
& URBAN POLICY

“How can we now find modes for interacting with strangers that simultaneously enhance security
and improve the quality of our interactions? …Democratic trust depends on public displays of an
egalitarian, well-intentioned spirit.”
– Danielle Allen (2006, pp. 166–67), Talking to Strangers: Anxieties of Citizenship since
Brown v. Board of Education

Overview of the game
How can games spur social ties and local investment as part of urban revitalization? This
chapter is a case study of Macon Money, a game designed to structure social and economic
activity. The analysis will focus on the central actions or mechanics of the game, and investigate
fit to local modes of meeting and greeting. Along the way, I will argue that Macon Money
provides a kind of informal socio-economic policy, but with a more micro focus than most tax
incentives and the ability to address social ties that preserve local culture.
The game began in October of 2010, when Macon Money launched in the historic city of
Macon, Georgia. The activity lasted eight months, and was designed by highly regarded
commercial game designers
45
. Exchange in the game revolved around a custom paper currency,
also called “Macon Money,” which could be exchanged for real goods at select Macon stores.
The game was funded by the Knight Foundation as part of a broader campaign to energize the
urban core, especially a corridor stretching from Mercer University to the downtown business
district (see Figure 5).

45
Area/Code was acquired by Zynga in the middle of the game (January 2011). They’d been widely acclaimed for
Drop7, and had made games for CSI and A&E (most notably, Parking Wars).
37

Figure 5: Promotional map of the College Hill Corridor
which stretches from Mercer University to the downtown
business district (approx. 2 square miles)

Source: the College Hill Alliance, as of October 2013

Following the money tells the game’s most obvious story: by the end of its run, players
had exchanged game currency for more than $64,000 at local restaurants and retail stores, from
bicycle shops to clothing boutiques. More than 3,500 residents played, with the typical winners
exchanging between $30-50 worth of currency. There were 40 participating businesses.
What made Macon Money a game, and not just a coupon system? A simple twist: the
designers decreed that only matching bonds have value. Players would receive a half bond, and
needed to find someone with a matching half, and together exchange the bonds for spendable
Macon Money.
38

In the language of game design, the requirement of matching bonds served as the primary
formal constraint. Constraints are at the heart of games, including the rules of the game
(Fullerton, Swain, & Hoffman, 2004; Salen & Zimmerman, 2004). Compared to a traditional
coupon system, the matching constraint is inefficient; yet matching adds a social challenge that
can be rewarding to overcome.
For some players, the game was discovered at home when they received a half bond in
the mail. Others attended a community concert or event, were given a half bond, and tried to
make a match before the night’s end. Meeting people – especially strangers – isn’t always
comfortable or easy. But those who successfully joined to exchange a bond were often all
smiles. The businesses were also all smiles when they exchanged the bonds for real dollars. A
fascinating series of pictures were taken when pairs turned in their matched bonds (see Figure 6
and online
46
).

Figure 6: Bonds and players of Macon Money
Bonds can only be exchanged for whole bills of
Macon Money if the halves match. A maximum of
three bonds each month could be cashed per
player.

(B) Two players with their Macon Money
after a successful match (denominations
include $10 USD-equivalent, $20, $50 and
$100)

Source: website for Macon Money, 2011.

46
For the photo stream from Macon Money, see http://www.flickr.com/photos/maconmoney/
39

An evaluation of the game (detailed below) described it as “highly adapted to local
context… carefully designed to respond to local needs and preferences” (M. Taylor & Whatley,
2012b). As this chapter will analyze in more depth, the anchoring comes partly through formal
rules. In particular, the money could only be redeemed at local businesses; additionally, players
had to reside in the center of the city.
Ties to Macon’s identity as a distinctly southern city were reinforced by the game’s
visual design (led by Rachel Morris). The graphics invoked Macon’s distinctive cultural history.
In particular, the paper currency featured the likeness of Otis Redding, who famously grew up in
Macon. Such care was invested in visuals that some players in interviews expressed a desire to
frame the bonds for their walls.
Statistics in context: one player’s trajectory
Since play is emergent in games and rarely linear, a concrete case is needed to understand
the social dynamics behind the game’s structure. Consider the story of Justine
47
, a stay-at-home
mom in her early 30s who has lived in Macon for more than five years. Justine’s story is based
tightly on an in-depth interview conducted shortly after she finished Macon Money.
48

Justine joined toward the end of the game’s eight-month run. She joined at the urging of
her pastor. It happened in Amanda’s Cakery, where Justine was gathered with her church group.

47
Justine’s name and some details of her story have been changed to protect her privacy.
48
A number of similar interviews were also analyzed; these participant interviews and transcripts were produced by
Taylor and Whatley, informing their public report (2012b) on the game. The author served as an advisor to the
evaluation process of Taylor and Whatley, in his role on the Knight Foundation’s Games Advisory; transcripts of the
interviews with Justine and others were provided courtesy of Taylor and Whatley. A full description of research
data sources is provided later in this document.
40

The church outing was part social and partly to support local business. Justine’s church is a big
proponent of downtown revitalization, including a prominent public-private partnership called
NewTown Macon
49
. As Justine describes it, “we’re trying to encourage people to shop
downtown and eat downtown…as much as we can.” Similarly, nearly half of all players
reported playing with a social group rather than on their own, according to a post-game survey
50
.
Common groups included work colleagues, friends and family.
Even with her church companions playing, Justine at first did not intend to join. But after
her meal, Janine discovered that Amanda’s Cakery did not accept credit cards, and she did not
have any cash. “Just play Macon Money,” her pastor urged. The suggestion was perhaps
excessively obvious; at that very moment staff with Macon Money were hosting an event in
Amanda’s Cakery. They distributed the half-bonds to anyone for free, as part of a strategy to
advertise the game in the spaces where the currency could also be used. If you found a match,
you could exchange your bonds immediately with game staff and use the game currency to help
pay your bill, or for use later.
It is worth pausing to consider Justine’s early hesitation. She explained that she “didn’t
really understand; …I didn’t know how to sign up,” even though she recalls hearing about
Macon Money in prior weeks at local events and her church. Similar hesitancy was expressed by

49
NewTown Macon is a non-profit partnership that was created in 1996, and dedicated to “implementing the
community’s vision of a comprehensive, coordinated, and continuous plan for revitalizing downtown Macon,
Georgia.” The group emphasizes its accomplishments in terms of real estate development, creating walking trails,
enabling community events, installing historical markers, and supporting public transit. For more, see their website:
http://www.newtownmacon.com/about/
50
Specifically, 45% of survey respondents indicated that yes, they did “play regularly with other players as a
group … with a group of friends, a group of colleagues at work, a group of people at church”; n=295 people who
answered this survey question. Again, survey data was collected by Taylor and Whatley, as is detailed in the data
overview on Macon Money in a following section.
41

nearly a third of players in the post-game survey. Justine’s confusion points to distinct
challenges for real-world games. In part, games ask busy people to do more than listen; they
must actively participate, and often in ways that are unfamiliar.
As Justine struggled for words, she explained that she had “never been in a situation…
like that.” (By contrast, most community events have a common “situation”; for example,
community concerts consistently structure participation around a performing stage, and people
know that the expected behaviors include dancing and talking.) Novelty can even be threatening,
especially with overtones of money: many participants wondered if Macon Money was a hoax,
and the problem may have been particularly pronounced in the African American community
(M. Taylor & Whatley, 2012b).
For Justine, the challenges become more interesting as she played. She stayed that day at
Amanda’s Cakery for a while looking for a match. In order to match, three symbols on the paper
bond must line up – for example, the peach symbol, kazoo, and a music note (for a picture, see
Figure 6). Unsuccessful in the Cakery, Justine asked others how they had found matches
previously. They suggested she try online.
At home, Justine posted to Facebook in search of a match, using a page on Facebook that
was moderated by Macon Money staff. Her post mimicked what she saw others doing: posting
the symbols they were seeking (see Figure 7). Most players found one or more of their matches
on Facebook according to exit surveys. Additionally there was a message board on the Macon
Money website, which was used disproportionately by older players to find matches.

42

Figure 7: Screenshot of the Facebook page for the game
where players advertised their bonds in hope of finding a match
(three symbols were required to make a match)

Source: Cause Communications, from March of 2011

Within 24 hours, Justine had found a match on Facebook. Her pair was a student at
Mercer University. They had never met before. As Justine says, “I wouldn’t have met him
otherwise – definitely not.” Matching with a stranger was common: two thirds of pairs had
never met, according to a post-game survey. Similarly, a majority of players felt it was “very
unlikely” they would have met the stranger otherwise. A network visualization of the successful
matches is shown in Figure 9.
Justine picked a time to meet with her partner using Facebook. They would meet near
the Macon Money office downtown to exchange their half bonds for the usable Macon Money
currency. This mix of online and offline interaction was appreciated by players, with over 70%
of survey respondents saying that it added “a lot” to the game. For Justine, meeting a stranger in
a public space was straightforward, and not particularly scary or awkward; others, however,
found meeting strangers a bit intimidating. Yet after a match was made, most people did meet –
more than 90% of players who matched bonds reported meeting in person.
43

The value of Justine’s game currency turned out to be $20. Players could only request
one bond per month (and exchange no more than three in a month), but since Justine joined at
the end of the game, this was her only match. Total game winnings for most players were
between $30-$50, as shown in Figure 8.

Figure 8: Winning totals per player
(based on post-game survey, n=261;
largest currency denomination was $100)

Source: Evaluation report on Macon Money (M. Taylor & Whatley, 2012b, p. 12)

The socializing at first glance seems disappointing. Did Justine talk much with her match
while turning in the bonds? “Not really,” she said. Minimal talk between strangers was typical –
under 10 minutes, according to the survey. And most often there was no further contact between
the strangers either; only 20% had more contact online or offline according to the survey.
51

51
Approximately 15% became friends with a previously unknown match on Facebook, and a similar percentage
have talked once or more in-person. Data is from the post-game survey (n=455).
44

Days later on Facebook, Justine heard from her match in a small way when the Mercer
student reached out to become Facebook “friends.” Even this minimal connection depended on
previously hidden network ties; as Justine describes, “he friended me just because we had a
mutual friend” on Facebook (emphasis added). Social norms limited further contact. As Justine
explained, “he’s a single guy and I’m a married woman so that doesn’t really… you know.”
Nonetheless, the Facebook connection means that social updates from the Mercer student
will now appear on Justine’s Facebook feed. She will likely keep an eye out for him; over 70%
of players said they would say “hi” to their match if they ran into them in the street or
supermarket. In a small but important sense, social news is now flowing in a new way between
Justine and her match, a tiny connective bridge across the “town and gown” gap. Closing this
gap was one of Knight’s goals for the neighborhood and the game.

Figure 9: Network graph of matched pairs from Macon Money

Source: Knight Foundation visualization

45

Bridging across diversity was a broad goal of the game, including to students, by zip
code, and by race/ethnic group. For pairs involving a Mercer student, the pair was with a non-
student 25% of the time (n=205). Knight was similarly interested in bridging perceived
segregation by zip code, for mobility and as a proxy for crossing socio-economic boundaries that
tie to geography. Overall, pairs were from different zip codes for more than 45% of all
successful matches (n=244), and from different racial/ethnic categories 26% of the time (n=196).
These numbers are even more impressive considering that players often signed up as a group of
likeminded friends, or at a group event with people similar to them.
It wasn’t over. When Justine sought to spend her Macon Money currency, she joined
with a different friend, offering to take her out for a treat – they would split the $20 value. Most
players similarly reported spending their bill on or with someone else. While official records
would only show one business that received Justine’s currency, the real story has additional
network ties.
First, the two friends visited a local restaurant called the Rookery, which is Justine’s
favorite restaurant downtown. Through her church, Justine had already been recruiting people to
eat downtown; in fact she seemed proud of her accomplishment: “I know a few people that have
started eating different places downtown because I recommended it, but it’s been hard. Most
people are not convinced that easily.”
Justine’s social connections at the Rookery did not appear in the quantitative data tracked
by the game. Why? The waitress accidentally rang up the bill without including the Macon
Money currency, so Justine and her friend decided to try again at a second place – and bring their
46

spouses. But the Rookery story is worth telling, as it shows how the formal actions of the game
can obscure networking at the game’s periphery, outside the official numbers.
The second business Justine visited had some history for her. It was a coffee shop called
Joshua Cup, and Justine had stopped going when the Starbucks chain opened a local branch.
Why the switch away from a local business? “My husband and I just prefer Starbucks coffee.”
But this time, Justine discovered the sandwiches at Joshua Cup on the advice of her friend’s
husband. The result: “I would go there again,” she said, “it did kind of redeem them.”
After her success with the game, Justine “put it on Facebook… I was telling people, some
friends of mine that don’t go to my church… I told lots of people about it.” Similarly, most
players ultimately recommended the game to others – more than 90%, according to surveys.
What made it fun? For Justine the quick answer was “the free money” but she added that
it was also “fun trying to make a match…even though I’m no good at small talk and it takes me a
while to get to know someone.” Predictably, 97 percent of players surveyed afterward liked the
free money “a lot.” Finding a match, which is the central action in the game, was widely
appreciated: 60% “liked it a lot” (the highest level) and an additional 33% reported they “liked it
a little.”
Curiously, more people expressed the highest levels of enthusiasm about supporting a
local business than meeting someone new (86% versus 36%, n=289), although this is for all
players – including those who only played with established friends. Justine also felt that having
residents meet one another was “good for Macon as a whole.”
For the ethics of the game, Justine’s view of the game’s model is important, especially in
terms of the community’s needs. The problem according to Justine is that the residents of
47

Macon are often stuck in their “own little bubble,” and that “it’s hard to convince people that
downtown is a safe and fun place if they’re not there regularly.” (Her assessment matches nicely
to the designer’s goals.) The benefit of the game can be subtle, but Justine still invokes big
possibilities: “Sometimes,” she said, “the little things... open your eyes to the bigger things that
are going on.”
Is it a Situated Game?
The framework of the first chapter proposed three conditions to qualify as a Locally
Situated Game. This chapter explores the criteria in particular detail. The goal is to
demonstrate the kind of analysis hidden in defining the object of study. (Only once we have a
clear object of study can we investigate the value of ‘local fit’.)
Below, Macon Money is verified as a Situated Game in terms of the three baseline
conditions: (a) being a game-based activity, (b) having real-world actions, and (c) having
network-building goals.
(a) Game-based
Transparently calling itself a game is part of what sets Macon Money apart from many of
the experiments with alternative currency. Political uses of alternative currencies are numerous
(e.g., see North, 2007), but rarely have the process and experiential focus of games. In
particular, they are not optimized to be engaging experiences. Even more hand-on experiments
like participatory budgeting (e.g., see de Sousa Santos, 1998) tend to focus on dutiful
participation (shifting power and building trust) rather than tapping into playful engagement.
48

Delineating how Macon Money qualifies as a game-based experience is a prerequisite to
understanding how it can be optimized to build local communities. According to our criteria
defined in the first chapter, a game-based design is one that structures participation through (1)
playful challenges throughout the activity; (2) feedback loops to guide progress; and (3)
uncertain outcomes for participants.
The language of “game-based” provides flexibility that is often needed for real-world
games. Even the game design professionals behind Macon Money had multiple ways for making
progress. In fact, the game’s Senior Producer described winning in terms of two different games,
one for players and one for businesses:

“Players win by meeting new people, learning about their community, earning and
spending currency locally. Businesses win by meeting new community members and
expanding their customer base.” (emphasis added; a full interview with Kati London is
available online)
52

Neither of the win conditions above has a clear quantitative end state, which can
disqualify Macon Money as being a formal game by many accounts
53
. By the end of Macon
Money, no overall winner was declared. Nor were individual players given explicit criteria for
whether or not they personally won, or even a total score. An implicit goal for players was to

52
See (Ruiz, Stokes, & Watson, 2012 – see: http://civictripod.com/interview-kati-london/).
53
There are many attempts to describe what constitutes a legitimate “win condition.” For example, Salen and
Zimmerman (2004, p. 80) insist that all games must result in a “quantifiable outcome” – like a numerical score to
indicate who won or lost; “a quantifiable outcome is what usually distinguishes a game from less formal play
activities.”
49

secure more currency, but there was a quick ceiling since no player could cash in more than three
bonds per month. The driving motivations seemed to be largely extrinsic to the game, with many
players in interviews described their goals as purchasing goods and meeting new people. For
real-world games, flexible win conditions are often needed.
With multiple ways to “win,” real-world games can be more accessible. Residents of
real-world communities have many reasons for participating in community life. Supporting
multiple play styles and goals may be vital when the game is also delivering economic resources,
let alone currency.
Keeping the challenges easy (e.g., finding a match at a large event) was a deliberate
strategy in Macon Money. For Justine, the big constraint was simply time: players were warned
that the currency would run out, and that businesses would only redeem the currency for a few
weeks beyond that. One facilitator of the game admitted:
“It was fairly easy … everybody was a winner …We may have been able to make it last
longer had it not been so easy, but [we] wanted to give a sense that anybody could play
this: you may have a tenured professor meeting someone who doesn’t even have a high
school degree, and that’s what it was supposed to do.”

Of course, some challenge is vital. Identifying the challenge is often easiest by looking
for the most dedicated repeat participants, sometimes called the hardcore players. In Macon
Money, two groups emerged: one kind of player (a) sought to maximize peer participation and
spent great energy telling others about the game and bringing them to events; while a second
kind of player (b) sought to maximize financial gain, and even benefit from securing bonds for
50

family and friends, sometimes illicitly. Each player type had a different implicit strategy for
success, a different way of playing the game. The presence of optimization strategies helps to
verify that the game has sufficient feedback loops.
Despite loosening the notion of ‘game’ to allow game-based activities, recall that “game-
based” is defined explicitly in this study to go beyond including arbitrary game elements. It is a
dangerous and slippery slope to be piecemeal and cherry-pick from what looks like games. (For
a good set of warnings, see the literature review of ‘gamification’.
54
) Instead, the definition of
Situated Games includes a baseline criterion to help ensure that the objects of study are indeed
games and can be credibly optimized as such.
So how can the notion of ‘game’ be broadened yet stay rigorous? The opening chapter
pointed to two useful terms that broaden the frame: investigating the ‘ecology of the game’
(Salen, 2008); and investigating the ‘big-G Game’ that includes the social and cultural
production around it (Gee & Hayes, 2010). However, these terms are somewhat vague. More
specifically broadening ‘game’ to emphasize culture – with all its rich human dynamics – can
provide an important step toward understanding ‘games’ in the broader social context of the real-
world.
Games as culture
To get at culture, a particularly useful approach to studying games comes from Katie
Salen and Eric Zimmerman (2004, p. 102). Specifically, they propose that the study of games

54
As discussed in the opening chapter, gamification is primarily a marketing movement that applies things like
points and badges to “incentivize” non-game activities, without a consistent framework to articulate the underlying
principles for how games work.
51

should explicitly include three distinct perspectives: (a) games as systems of rules; (b) games as
play; and finally (c) games as culture. While rules are controlled quite directly and intrinsically
by the game, culture is at the other end of the spectrum, and brings a set of extrinsic motivations
to the game. These three perspectives are interconnected, as shown in Figure 10, with play
serving as an in-between, structured both by the rules and culture.

Figure 10: A framework for studying games
using three interlocking schemas

Source: Salen and Zimmerman (2004, p. 102)

Pure entertainment games are often analyzed primarily in terms of the rules and play. In
contrast, a primary emphasis on culture makes more sense for real-world games, since the design
seeks to affect the social ties that weave and support local culture. For example, social network
formation may depend on cultural norms of talking to strangers. As Salen and Zimmerman
observe, “it is when we explore games within the realm of culture that the overlap between the
game world and the world at large comes to light” (2004, p. 104). Thus the cultural perspective is
particularly useful for theorizing local games, first in terms of how games are shaped by local
culture, and in turn, how games can affect the local social fabric.

RULES
PLAY
CULTURE

52

Can the playful challenges of Macon Money be understood from a cultural perspective?
The norms of social interaction, for example, tend to go beyond the formal challenge in most
games. Most notably in Macon Money, players had to muster the courage to approach strangers
in public space. As one game facilitator described it, “You had to be bold; you couldn’t be a
wallflower and find a match. … if you want to feel like you’re part of a community, you gotta
get out there and be invested.” Note how quickly the description shifted from interpersonal
hesitation to community belonging and investment.
Feedback loops are the second criteria for being game-based. In Macon Money, feedback
guided players with how to progress in the game, using explicit instructions at each stage. The
primary loop was about getting bonds and cashing them in. Consider each stage: (1) when
getting started, players were guided on how to find a half bond; (2) then they were told how and
where to find the match; (3) with their match, players were guided to meet in person and
exchange bonds for a whole bill of Macon Money; (4) finally, lists of businesses and
advertisements in store windows guided players to select a local business to receive real goods
and services. Then the loop could begin again.
To be game-like the activity must have playful challenges. Play emerged informally at
many places in Macon Money. In fact, one funder of the game felt the playful approach was
what most distinguished the activity, opening new spaces for conversation:

“Macon Money allowed people to be silly. It gave them permission to talk to somebody
they otherwise never would have talked to. It gave permission to match the bond in Otis
[Redding]’s likeness and laugh about the fact that you made a match. It gave people the
53

permission to greet people they never would have met. It’s amazing to see it happen. It’s
just like all these race class barriers, age barriers, just melted away for that small time,
for that moment.”

The play and silliness was structured by the game’s feedback loops and challenges. The
challenges themselves were more serious and goal-oriented, but they also left room for
uncertainty. Supporting uncertain outcomes is the third and final requirement for qualifying as a
game-based design. Uncertainty is one of the best ways to understand how a game is
functioning, especially when uncertainty is expressed in terms of possible failure (Juul, 2013).
Uncertainty in Macon Money was supported at nearly every stage: (1) the player might
fail to get a half bond – either through bad luck, poor timing, or not going to the right events; (2)
the player might fail to make a match – unless they are proactive and mix online searching with
face-to-face; (3) the player might fail to benefit from their match – unless they can coordinate
with their pair, and unless they can find a qualifying business that has something they want to
purchase; and (4) in the long-term, there is uncertainty that the interpersonal relationship will be
socially and culturally compatible.
Analyzing uncertainty also benefits from the cultural perspective on games. Some of the
highest stakes and possibilities for failure are subtle: recall how Justine limited her play of the
game because as a married woman, she could not risk too much contact with a single male
student at Mercer.
Extra time and effort is required to cultivate playful challenges and uncertainty. Why
bother? The overwhelming justification in games is to foster engagement. The goal of games is
54

not to optimize for task efficiency, but to create engaging experiences. Real-world games are
thus better aligned with civic engagement than accomplishing civic tasks efficiently. Long-term
engagement is vital to sustain the political will of civic campaigns for justice, and to shift culture
to support sustainable development.
By contrast, most civic technology design comes from a utilitarian tradition that values
predictability and minimizes uncertainty. For example, the designers of civic tools (e.g., the
VoJo tool for community mapping discussed in Chapter 4) try to give their users a clear picture
of what will happen when they use the tool, and to consistently deliver on users’ hopes.
The playfulness and engagement of games relies on the radical act of deliberately making
the (civic) outcome uncertain. Quite simply, if you know that you will win in advance, the game
is no longer any fun. Formal civic life also has a ‘dutiful’ tone (Bennett, Wells, & Rank, 2009).
But a playful tone is culturally acceptable when building community and engagement. For
example, engagement is central for community events like concerts and open spaces like parks.
Play takes precedence over instrumental tasks and goals as second.
The temptation with real-world actions is to revert to the dutiful tone of instrumental
civic goals. Yet playfulness is vital to social mixing and many community building goals. The
distinction with games is that instrumental goals must come second. As Fullerton et al. describe
it, the role of the game designer is “first and foremost to be an advocate for the player” (2004, p.
2).
(b) Real-world actions
Most real-world games in local communities focus on content and seek primarily to
educate (akin to situated documentaries). The requirement of real-world action is unusual, and
55

part of what makes Macon Money unusual too. Clearly Macon Money takes place in the real
world; as the game’s Senior Producer describes it, “gameplay is not limited to the screen, it
happens where people live: in person at events, on the streets, at work, via Facebook, Twitter,
and the game forums”
55
. There is great potential in educational games in this context, especially
those with augmented reality (Klopfer & Squire, 2008; K. Squire et al., 2007). However,
Situated Games must go beyond training residents for the future – the emphasis is on some
immediate impact.
In particular, the requirement for Locally Situated Games is that progress in the game is
congruent with affecting the real world. If winning in Macon Money follows the designers’
intentions, then progress also results in: (1) economic gains for businesses; and (2) a stronger
community, as measured by new social ties between residents and to community institutions.
Thus the inputs and outputs are both real-world. In the case of Macon Money, the game
“required building or spending your social currency to earn a local currency,” according to the
game’s Senior Producer. The risks were also real for the partner organizations: “Those bonds
were currency,” emphasized a local staff member from the Knight Foundation, “we had to lock
them up, make sure that every single serial number matched, and they all had watermarks.”
(c) Network building goals
The final criterion is that Locally Situated Games must seek to build local networks.
Local in this sense aligns with the geography of place-based communities. For Macon Money, a
similar emphasis is clear beginning with the name of the game, and continuing through Justine’s

55
For the full quote, see my interview with Kati London (Ruiz et al., 2012 – see: http://civictripod.com/interview-
kati-london/).
56

experience. To see the design goals for a game, look to the pre-game funding and design
strategies.
The primary funder was the Knight Foundation, under its portfolio for “engaged
communities.” Macon is one of 26 geographic communities of focus for the Foundation, with
Macon quite prominent among them. The Foundation, in 2009, invested more than $5 million
USD in downtown Macon and surrounding areas. The very name of the “College Hill Corridor”
comes from a community study that was funded by Knight, based on ideas initially proposed by
a graduate student at Mercer University. (Again, the Corridor is a two-square mile area between
Mercer and Macon’s downtown business district, encompassing several historic neighborhoods.)
Most of Knight’s additional funding in Macon has focused on a $3 million dollar “Neighborhood
Challenge,” which “underwrites the best ideas to transform the city of Macon’s first
neighborhoods into the vision of the College Hill Corridor Master Plan.”
According to Knight’s local program officer, Macon Money emerged from the top of the
Foundation hierarchy, where there was interest in the emerging possibilities of real-world games.
She recalls that President Alberto Ibargüen provoked his staff, demanding: “[since] we’re all
about building communities, could we do some experiments with place-based games?”
Knight’s emphasis on building the local community is echoed in the game’s early design
documents. In 2009, more than a year before the game launched, a confidential research
document by the game design company opened with the telling header: “Building Community
Connectedness and a Sense of Place.” From such goals – or at least justified by them – the game
Macon Money was born.
57

Data collection and analysis
Before analyzing local fit for Macon Money, the data sources deserve brief explanation.
The analytic approach in this chapter is multi-method, and relies extensively on collaborations
with several other researchers. My personal involvement began midway through the game’s run,
when I was invited by the Knight Foundation to join the advisory committee
56
for their two first
social issue games – most notably Macon Money and Battlestorm
57
.
As a result of our advising, an evaluation firm was selected. They developed an
evaluation strategy in consultation with the Knight Foundation and our committee. A year after
Macon Money was completed, Taylor and Whatley published their excellent public report
(2012b) on the game, which is cited throughout this chapter. Their report is based on the
following data:
• Surveys of players signing up for the game
• Interviews of eight pairs of individuals who matched their bonds, and a selection of those
who received bonds but did not use them
• Interviews with business owners, stakeholders
• Materials review, including game design documents, the public website, game Facebook
page, etc.
• Participant observation

56
The Knight Games Advisory group was formed in 2010 and continued through 2012. Members included James
Paul Gee, Beth Kolko, Tracy Fullerton and me. As full disclosure: members were not employed for hire, but did
receive an honorarium and public recognition.
57
Battlestorm is based in Biloxi, Mississippi, with goals of building community resilience – especially to mitigate
the effects of natural disasters like hurricanes. Details are at: http://www.knightfoundation.org/grants/20100141/
58

As part of subsequent research, I secured access to most of the original transcripts for interviews,
the raw data from the survey, and raw data on bond tracking. In addition, I conducted follow-up
interviews with key stakeholders. I also conducted participant observation in Macon, visiting
several of the businesses from the game, speaking with former players, and attending a key
public meeting of the College Hill Alliance that included gathering input on renewing the
neighborhood plan, as well as announcing funding for new projects in the Knight Neighborhood
Challenge. For a more comprehensive list of data used in analyzing the Macon Money case, see
the Appendix.
Analysis throughout this chapter seeks to emphasize Macon Money at the level of its
game mechanics (discussed below), and as an example of socio-economic policy for regional
development. Subsequent case studies will emphasize the other two dimensions of ‘local fit’,
including the fit to local community identity and the local communication ecology.
Investigating Fit: Game mechanics (and local modes of meeting)
The first dimension of ‘local fit’ is with local social modes, especially modes of greeting
and meeting. Greeting and meeting – in person and online – are core actions at the heart of any
effort to build community and strengthen local networks. In games, the core actions are called
‘mechanics’ (or mechanisms).
Game mechanics describe what players do when they participate. To stay focused on
lived activity, mechanics are often best described with verbs (Järvinen, 2008, p. 263; Sicart,
2008). For example, a mechanic in Hide and Seek is “finding people” who are hidden; the action
of finding people is repeated throughout the game. Primary mechanics are somewhat shaped by
59

formal structures (like the game rules) but ultimately emerge through play and external cultural
norms.
In a community, local modes of meeting and greeting are established by situated cultural
norms. For example, the situation of greeting neighbors on the street with a quick ‘hi’ is
different than the longer conversation that might be expected at an annual block party. Thus
community life has its own mechanics of meeting and greeting, to use the language of game
design.
The language of ‘game mechanics’ is useful for two reasons. Analytically, the term and
its associated analytics gets at the heart of what makes games distinct as a form, especially how
games structure participation around play. Politically, the term is also useful for its traction with
game designers and the emerging field of game studies (Fullerton et al., 2004; Sicart, 2008), and
thus increases the odds that ‘local fit’ can be used in applied design.
Perhaps unexpectedly, game designers may need to slightly shift their use of ‘game
mechanics’ to better address real-world games. As will be shown with Macon Money, designers
can easily revert to articulations of mechanics in formalist terms and game rules
58
; but shifting
back to emphasize culture may be necessary in order to use mechanics as a common language at
the community level. (Game designers may find it easier to understand this linguistic shift as
more tightly aligning the mechanics with the ‘experience goals’, to use the language of Fullerton

58
For example, rules and play (not culture) are at the center of the mechanics of pervasive games described by Björk
and Peitz (2007), including the CHASE game and the SEARCH game. The authors are seeking ‘patterns’ in
mechanics in search of more universal mechanics. Typologies of mechanics are only possible with rules and
formalism (not culture) when the games are similarly group by technology type (i.e., location-based games). By
contrast, culture is a primary perspective for Situated Games, and so must be more centrally a part of the typology.
60

et al. (2004).) One objective for this chapter is to illustrate the space for common ground,
especially using community-centric language.
To be robust, the cost (or negative impact) of poor fit should also be explicable in terms
of mechanics. As a preview, poor fit can make the game harder to understand (and thus limit
adoption), since players can more easily grasp how established actions can be extended, rather
than being asked to invent new actions. Poor fit can also lead to resistance from community
leaders, who already have established activities, and lack the bandwidth to start new modes for
engaging residents.
Two primary mechanics are proposed to describe Macon Money. Each mechanic is
identified based on interviews with designers of the game, and my own analysis of the data
sources described above including the stories of participants like Justine. By investigating each
of the two mechanics in terms of ‘local fit’, this study seeks to better understand how the success
(and limitations) of the game depend on community integration, and how such games might be
better optimized in the future. Again, the goal is not to prove that the game had fit, so much as to
establish how optimizing for fit helps explain the game’s success and could further improve the
game.
(A) Mechanic for social ties: “to greet in public space and match” (for mutual benefit)
Here is the process used to articulate a mechanic: first, describe the game in a sentence
that explains what players primarily do in the activity, using simple verbs: In Macon Money,
participants must match bonds and redeem them in local businesses. Each primary verb must be
at the core of how players make progress in the game, and the action must be repeated. This
61

articulation of the mechanics draws on what Fullerton et al. (2004) call the experience goals and
player strategies.
59

Additionally, Situated Games emphasize network building goals, so mechanics should be
articulated in a way to highlight the social dynamics that match the theory of change. For Macon
Money, “matching” takes on particular importance, since matching is the moment of outreach
and tie creation for residents. We can use more precise language around the theory of change,
and fully articulate the first mechanic: in Macon Money, participants must greet in public space
and match (private goods) for mutual benefit.
Verifying the mechanic requires a concrete case. Consider how Justine played the game
just once – with just one official match – yet the mechanic of greet-and-match takes place as
many as five times. Each setting is bolded as follows: first, Justine attempts matching bonds in
Amanda’s Cakery, where she greets friends and strangers; unsuccessful, Justine’s second
attempt is on the game’s Facebook page, where her greeting is online in a semi-public page tied
to Macon, and the meeting quickly turns into private messages to plan next steps. Third, Justine
meets the student in the physical world to turn in their half bonds together (at the appointed time,
she greeted the man on the street who came up matching his online description). Finally, in

59
To bring some of the logic of the experience goals and player strategies into the analysis, a new approach may be
needed for Situated Games. The terminology of “experience goals” is not as popular as “game mechanics,” but it is
just as important. In brief, experience goals describe how players will feel in the situations of the game, i.e., how the
experience will feel to them (Fullerton et al., 2004). The experience goals are big picture ambitions, transcending
specific features. One vital difference between mechanics and experience goals is that the designer has more control
over the mechanics, since the game can require players to “match symbols” in order to progress – but players cannot
be required to feel an emotion, or pursue a specific strategy. The proposal of this dissertation is to articulate certain
“umbrella” mechanics a bit more like experience goals. Pragmatically, this is an additional step for designers, since
they will still need to articulate basic experience goals and basic mechanics. The umbrella mechanic is a kind of in-
between that groups primary mechanics in terms of the experience goals. Compared to mechanics on their own,
umbrella mechanics are more transparent to outsiders and are more accountable to the underlying goals.
62

order to use her currency, two more greetings were necessary: Justine met her friend at the
Rookery restaurant; unsuccessful, they met and greeted their spouses at the Joshua Cup coffee
shop to enjoy food, drink and conversation. The mechanic thus helps articulate what exactly is
repeating in the game.
A tempting alternative is to express the mechanic in the literal rules of the game. This
tendency seems particularly natural for game designers, given how mechanics are discussed in
games textbooks and conferences. For example, in an interview one designer described the
mechanic as “dead simple – match icons.” But to capture the experience of Justine requires going
beyond matching icons.
In fact, much of the challenge and reward in Macon Money comes from the social
greeting and negotiation. The most important matching is not the icons but the people.
Validating the social side of the game (i.e., a cultural approach to mechanics) helps to focus
expectations. As a local organizer of the game emphasized in an interview,

“I had to keep saying over and over and over again, Macon Money is about introductions. It is
about introducing people to each other. It’s about introducing people to the neighborhood. It’s
about connecting the people to their neighborhood and … the opportunity to go to businesses in
College Hill and downtown that they may never have gone to before. And as I said, that’s all –
that’s all it’s about. I think these games have way too many expectations on them.”

To begin evaluating fit of this mechanic, consider the other half of the fit puzzle: the
community. Several local activities were already in place to foster actions of social meeting and
63

greeting. For comparison to a game, it is natural to first seek established activities that similarly
have playful challenges, perhaps even with a bit of friendly competition.
The best example in Macon may be the Soap Box Derby, a popular annual event that is
explicitly praised in the master plan for the College Hill Corridor. To compete in the Derby,
approximately 20 teams build cars that roll down a Macon hill, winning if they are “fastest” or if
they are selected as the “People’s Choice.” On the day of the event, hundreds of spectators
gather to watch the derbies compete (see Figure 11).
A matching program exists with the Derby for local businesses: a team can be sponsored,
and receive funds for their material costs from a sponsor who may receive advertising in return.
The branding of the Soap Box Derby has often echoed that of Macon Money, for a good reason.
The Derby was initially funded in 2009 by a Neighborhood Challenge Grant of the Knight
Foundation, which has close ties to the Knight funding of Macon Money. It has since become
self-sustaining (Ramati, 2012). The branding was most noticeable through the endorsements of
the College Hill Alliance, which indicates as least occasional integration of the activity into the
identity building by local organizations.

64

Figure 11: Soap Box Derby graphic, as part of a social media campaign
to promote the Knight Neighborhood Challenge and recruit ideas to improve the Corridor

Source: Website of the Community Foundation of Central Georgia

What are the mechanics of the Derby that build social ties? For participating teams, the
Soap Box Derby can be an intense process of social connecting through building together.
However, the teams seem likely to draw on existing friend circles, with some members forming
new ties – but from a constrained pool.
60
The majority of participants are not in teams, but are
there cheering from the sidewalks akin to a marathon or parade. Cheering is a somewhat one-
sided mechanic. In tight proximity, strangers certainly do greet and meet at events like a Derby,
albeit somewhat incidentally.
A limitation of comparing with Macon Money is that the Derby only happens once a year.
By contrast, community building in the College Hill Corridor happens year-round. To broaden
the Derby comparison, fit should also be compared with more sustained activities. In interviews
about what was “most similar” to Macon Money in terms of network-building goals, a series

60
Institutions like schools may be best positioned to facilitate unlikely teams, bringing students from different
backgrounds together to meet and form teams. However, public schools have a narrow age focus on under-18 year
olds and their families. Third parties often use school grounds, but this is akin to parks.
65

called the “Second Sunday Concerts” was repeatedly mentioned. Macon Money overlapped with
this concert series.
The Second Sunday Concerts take place in the evenings, and are described as a
“signature event” of the Corridor, according to the College Hill Alliance. Between 500-1000
people typically attend. Along with the music, food is available for purchase and picnicking is
encouraged (see Figure 12).

Figure 12: Second Sunday concert in Macon, Georgia

Source: College Hill Macon website

The neutrality of the Second Sunday series is a huge perk, according to interviews with
organizers. One a month, the free concerts in Washington Park are open to the public, March
through October. The neutrality is described in terms of inclusiveness across lines of race and
class, in implicit contrast to other seemingly open events. Churches in Macon, for example,
were described in interviews as places that are often quite segregated by race – and Sunday was
descried by one interviewee as “probably the most segregated day of the week.” The concerts
were held up as more neutral ground. Reportedly, some Macon preachers even encouraged their
Sunday morning congregations to attend the night concerts as a way to build a more integrated
66

Macon. One organizer claimed the concerts have become the “most diverse two hours in
Macon.”
Proximity at the concerts helps spur interaction, as it did at the Derby. The concerts can
serve as a kind of third place (as popularized by Ray Oldenburg, 1989) for social and civic
mixing outside of home and work. Oldenburg even described the importance of a ‘playful’
mood for third places. But even with a playful spirit, the social mixing is fairly passive.
In the beginning, Macon Money approached the concerts of Second Sunday as a way to
recruit participants. A local organizer in Macon who was hired to help with the game recalls
warning that “we did not need to start creating our own events – we needed to work within the
fabric of what we have here in Macon, because we have a lot of events, a lot of social things, and
we needed to use those rather than expending our resources and frankly my time creating
events.”
How can the fit of Macon Money to such local events be optimized? An organic and
flexible approach may be necessary. Perhaps the best organic approach for fit is to rely on an
actual human being, and insist on staff overlap. Evidence for the importance of this tactic comes
from a comparison of Macon Money to a game created by the same designers for a different city
(Battlestorm in Biloxi, Mississippi). The comparison summarized in the evaluation report of
Taylor and Whatley (2012b) found that trusted local leadership was a key reason for the
significant differences in integration with local events and community organizations.
Perhaps most importantly, adding Macon Money also gave something back to the context
of Second Sunday. The game introduced a structure (or excuse) for greeting and meeting. The
bonds themselves with their very physical presence gave players a “ritualized object so that they
67

could walk up to somebody and have something to say,” according to one of the game’s
designers.
Over time Macon Money began to have its own draw, and brought new people to the
Second Sunday series who wanted to play the game. A core group of players recruited heavily,
and their recruiting was vital to helping the game catch on.
61
Ultimately, Macon Money also
brought newcomers to established community events, including a film series. Embedding could
have failed – it could have distracted from the concerts, or been rejected by community
organizers as inconsistent. The relative success of the embedding within Second Sunday, a
series that was pointed to as a model by local stakeholders, indicates reasonable fit to the game’s
mechanics of “greeting and matching” in public space. By this measure, the basic fit was
sufficient, and increases depended largely on convincing existing institutions that they would
benefit enough to help promote and further integrate the game.
(B) Mechanic for business ties: “to buy locally” (for self and community)
A second primary mechanic emphasizes the network ties that players can make to
businesses. These ties are different than the connections among residents, but are almost equally
important for explaining the game (and optimizing for community impact). The mechanic can
be summarized with the simple sentence: to play the game, participants purchase local goods and
services.

61
There was a tipping point in the game’s adoption that occurred when it seemed to reach critical mass, partly
driven by highly motivated networkers who spread the game to a high number of people (M. Taylor & Whatley,
2012b); based on interview transcripts, it seems that several of these power players – perhaps most – did so because
they believed in championing Macon to their networks, not because it helped them to find matches.
68

Progressing in the game thus requires interacting with locally-owned businesses, and
doing a kind of “matching” with them as well. Especially when a local business hosted an event
for the game, a complex set of interactions could emerge around them. In the words of one game
staffer:

“We found that …more than anything [Macon Money] connected locals to our local
businesses. A lot of times the game was played within a local business, and it brought
people into the store that may not have been to that store yet – so now they’re in the
store, playing the game, and shop owners are right beside, also playing the game. It was
just making matches all the way around.”(Emphasis added.)

Many of the business connections were new for participants: nearly half of all players
visited a new business while playing. After the game ended, the returns continued: as
mentioned, more than 90% of the adventuresome players returned to a business they had
discovered through playing. The locations where funds were spent are shown in Figure 13.
Across multiple businesses, the game also seemed to affect where players traveled in Macon.
Two months after the game, visiting downtown “often” was reported by 10% more of the survey
respondents.
62

62
Based on a comparison of means by Taylor and Whatley (p<.01; 304 post-game, n=295 pre/mid-game); since
there was insufficient pre-game data for a direct comparison, the pre-game sample was expanded to include some
mid-game surveys, under the assumption that the bias introduced is conservative, i.e., that it will lead to under-
reporting the magnitude, rather than over-reporting it.
69

Figure 13: Business earnings by location, midway through the game
(after a majority of funds had been distributed;
downtown is located at the top-right of the map)

Source: Cause Communications (2011/12)

Beyond raw purchasing at local restaurants (to help the economy), the notion of “buy
local” can also reflect a deeper civic intentionality. In contrast with boycott movements that
dampen purchasing to punish business targets, buy local is aligned with a growing trend called
“buycotts” (Cohen, 2010), which focus more on the proactive purchasing of responsible brands.
For buy local, the brand is larger than any one company – it is the business side of the local
community. The buy local brand can thus reinforce a community identity, especially when the
buyer is making a deliberate choice.
What should the game be compared to for the mechanic of “buy local?” Like many
cities, Macon has a shop local website – ShopLocalMacon.com – featuring a directory of local
businesses. The directory speaks to the notion of community in largely economic terms; for
example, its homepage declares “for every $100 spent in locally owned stores, $68 returns to the
70

community through taxes, payroll and other expenditures.” The directory was launched shortly
after Macon Money ended – perhaps with some of the game’s momentum – and has since been
taken over by the City of Macon. While the directory embodies the branding of buy local, the
approach is informational – not based in social activities or making connections (i.e., there is
little emphasis on meeting and greeting). For game mechanics, an activity approach is needed.
The best activity-based approach to buying local in Macon may be in “cash mobs.”
According to a Macon resident who participated in one as part of her Rotary Club (more than a
year after the game ended), the goals include building social ties:

“Instead of a flash mob, a cash mob is a conscious decision to visit a certain business on
a certain day, and spend money in that business to help them out. … We all agree that
we’re going to spend our money there and at the same time we are increasing our
fellowship.”

Is there fit? The mechanic for the cash mob is indeed quite consistent with the “meet and
greet” approach to buying locally that is at the heart of Macon Money. Both are social as much
as individual financial decision-making. The challenge is largely a matter of social coordination
in both. At the level of ritual and exchange, the paper currency of Macon Money is a tight fit to
ordinary cash or coupons. A clear benefit of this alignment of tokens is accessibility, as
described by the game’s Senior Producer:
“Money is a fundamental reward that we are all already familiar with (dependent on), no
matter our socio-economic status. Leaning on currency as the primary game piece was
71

something that everyone from the Knight Foundation Board to local stakeholders
understood the value in.”
63

Normal currency has images of national importance, like famous presidents. To optimize
for a more local fit, the game’s currency was visually designed to evoke Macon, including
featuring famous local figures like Otis Redding. As one Macon Money staff person observed,
“People loved it, and frankly it was beautiful… I mean, people wanted it framed!”
The look of the currency was part of a broader campaign of “neighborhood branding” by
the College Hill Alliance. Even the name “College Hill” was invented just a few years prior to
the game as a way to build coherence for redevelopment. The Alliance describes their theory of
change partly in branding terms, especially for creating an umbrella for collective ownership,
and thus their success as a “community-driven” development agency. Said one Alliance staffer:
“to create that community and sense of place, we have a brand, we have a ‘corridor,’ and the
sense of a corridor has definitely helped create that sense of community; having that brand
creates a sense of recognition.”
With a player’s single use of Macon Money currency, the brand of the community was
asserted. More than 40% of players reported their perceptions of the College Hill Corridor had
improved “a lot” since playing the game (and an additional 45% said their perceptions had
improved “a little”). Repeated play seemed to benefit the downtown more
64
, perhaps because the

63
To read more of this quote in context, see my interview online (Ruiz et al., 2012 http://civictripod.com/interview-
kati-london/).
64
There was a positive correlation between the number of self-reported instances of play and the player’s positive
perceptions of the downtown area, according to the study of Taylor and Whatley (2012b, p. 27).
72

area had an unfair reputation for being neglected and unsafe, and each visit may have helped to
chip away at the stereotype.
For store owners, the staffer also felt that Macon Money helped shift organizational
identity to being part of a more collective campaign:
“If you ask any of our local businesses, Macon Money was a big part of being
recognized as a ‘corridor business’ – not just being a storefront, but something that was
a bigger part of the community. You became a local business that was part of
neighborhood revitalization.”

The downside of fitting tightly to modes based on currency exchange is that Macon
Money inherits biases tied to money and purchasing. Heightened fears among African
Americans that the game would turn out to be a hoax are troubling. The evaluation reported
significant under-participation in the game by African Americans relative to the city’s overall
population (M. Taylor & Whatley, 2012b), despite efforts by the game operators to compensate
by mailing extra game bonds to demographics that were joining at slower rates.
The lesson in cultural terms is that fit means a game will inherit more local stereotypes
and history, and without extra effort a game can reproduce or even deepen longstanding
structures of inequity. Yet compared to many Buy Local campaigns, games like Macon Money
may have more flexibility to target and over-sample historically marginalized populations, as
well as to target businesses that preserve historic culture (beyond simply supporting local
ownership). How to make the game accountable to such concerns will be discussed below in the
section on games as socio-economic policy.
73

Addressing manipulation
Do mechanics help to analyze games as potential sources of manipulation? In part, the
concern is with incentives, which are pretty immediate for Macon Money. “When you hear
‘Macon’ and ‘money’ people are going to line up out of curiosity,” observed one of the game’s
organizers. Of course, financial incentives have long been used to reward actions, like tax
credits for installing solar panels.
What is unusual in games like Macon Money is that the incentives are not immediate, but
are woven into a more complex set of social requirements. Games with real-world actions can
raise particular concerns about manipulation, including how incentives can mask a hidden
agenda. Historically, “token economies” to shift behavior have many roots, including the mental
hospitals of the 1960s which used tokens for operant learning (Glynn, 1990). The recent
‘gamification’ movement, for example, uses game features for customer retention (among other
goals), and also has historical roots in the token economy (Raczkowski, 2013).
When token economies are structured largely to manipulate behavior, and the complexity
is beyond the easy comprehension of participants, warnings of manipulation take on real weight.
Such concerns are partly mitigated in Macon Money due to the relative transparency of the game,
including the prominent website explaining the game’s funding, goals, and legal code. (The
formal disclosures and terms of the game were prompted in significant part by liability concerns;
in other countries, this check and balance may be less present.)
Ironically, ‘local fit’ may also be a means to minimize manipulation. The idea is that
local fit improves accountability, especially when fit involves local partners designing the game
and integrating it into their existing modes of community engagement. For oversight, local
74

organizations often derive their credibility from being seen as serving the community, and thus
are more likely to veto a game that makes inappropriate use of their events and existing
networks. Even stronger gatekeeping comes at the level of co-design, where local staff are
involved in creating and operating the game. Of course, local fit as a manipulation check will
only work if civil society organizations are independent and competent.
65

Even with strong local partners, there is some danger of civic designers who justify
manipulation in the name of a greater cause. Is it ethical to pursue civic ends through
weaknesses in human psychology? (Recall that gambling casinos provide vital revenues for
public education in several US states (Brunori, 2005, p. 117).) Games in this sense are
approached as a kind of chocolate sweetness atop the bland necessities of everyday civic life.
However, “chocolate-coated broccoli”
66
is warned as probably less tasty than you think, and
having only short term benefits. When players catch on, a backlash becomes increasingly likely
and the reputational assets of local organizations will be at risk. A strong local civil society
would likely provide a check on manipulation at a basic level.
Finally, literacy is often invoked as a counterbalance to manipulation risks. Should
citizens learn new skills to safely navigate real-world games? Perhaps. There are still relatively
few cases from which to draw conclusions, and the most risqué examples
67
are more global than
local. Fortunately, there are some natural features of games that may provide a more immediate

65
Such litmus tests are similar to the best practices in the foundation world of restraint in giving large grants to
small organizations that have never previously shown the capacity to handle significant sums, without proof of a
significant and active oversight board.
66
The consideration of games as chocolate-covered broccoli is commonly attributed to Brenda Laurel (2001).
67
In perhaps the strangest example, the iPhone game Raise the Village has a curious dual-reality, where participants
“play” to fight poverty, with a recreation of a poverty-stricken Ugandan village. The strange part is that advancing
in the game requires using real cash to purchase virtual items – but the cash is sent to a real-world Ugandan town,
and used on the same kind of goods. For more, see http://civictripod.com/games/raise-the-village/
75

counter-balance for ethics concerns, and build literacy implicitly. Specifically, games can be
optimized to provide “meaningful choices” for the real world.
Toward ‘meaningful choices’ for civic impact
Good game design is about offering choices to players. Games do not play themselves;
they demand participation to unfold. Simply pressing ‘continue’ is rarely sufficient for
engagement. If the choices are too easy or trivial, the game is boring; if it is too hard, the game
is frustrating and players quit. The sweet spot is where choices are satisfying because they have
an impact on the game’s challenges. Meaningful choices (also called “interesting choices”) are
at the core of good games.
68

Ironically, more choices are not always a good thing. Psychological studies of
satisfaction have shown that an expansion of choice can actually undermine happiness (Schwartz
et al., 2002). The initial decision can seem simple (e.g., “I want to donate to fight poverty”), but
when that decision also necessitates selecting between multiple non-profits, the burden of
maximizing impact shifts to the participant. How many of us have the time to fully investigate
each choice? As a society, we cannot simply design more civic tools, without offering
participants more meaningful choices.
69

There are many frameworks to analyze whether choices are meaningful. Real-world
choices are messy, mixing near and long-term, psychology and culture. Games bring a usefully

68
In fact, the full meaning of the game can sometimes be understood in terms of the choices that are offered to
players. Cultivating a sense of fairness is implicit; players must believe that the right choices will pay off. Such
principles go far back in the history of game studies (e.g., Huizinga, 1938).
69
I describe this approach in more depth in an article on “Restructuring Civic Engagement: Meaningful Choice and
Game Design Thinking” (Stokes, 2012).
76

interdisciplinary perspective (in contrast to the more controlled experiments of decision science,
e.g., the “prisoner’s dilemma” in game theory
70
). Consider some of the common “decision
types” that are suggested in a foundational text for game designers-in-training (Fullerton et al.,
2004, p. 320):
• Uninformed decision: an arbitrary choice
• Informed decision: where the player has ample information
• Dramatic decision: taps into a player’s emotional state
• Weighted decision: a balanced decision with consequences (risk/reward)
• Immediate decision: has an immediate impact
• Long-term decision: whose impact will be felt down the road
• Hollow decision: no real consequences
• Obvious decision: no real decision

For a real-world game, it helps to analyze the choices from a social perspective, echoing
the approach to the mechanics. If we were to simply define the mechanic in Macon Money as
“matching symbols,” then the challenge and choices are like a lottery: keep trying to find people,
and hope to get lucky. Certainly, the lottery description has an element of truth, but it is limited
(and misses what makes the gameplay meaningful and interesting).

70
The prisoner’s dilemma is a canonical example of tradeoffs in decision-making. It is part of what is called “game
theory” – which is a branch of decision science focusing on logic. Game theory began by studying behavioral
relations in the 1940s with two-person zero-sum games (Von Neumann & Morgenstern, 2007). Note that the name
“game theory” is somewhat misleading in the context of applied games, since it actually has little understanding of
games beyond formal rules, and is really more of a way of analyzing interactive decisions. The broader study of
games, including their design and cultural modes in society, is game studies.
77

By contrast, the cultural context of “greeting in public space and matching” leads to a
different picture. In the social frame, the challenges include finding the courage to approach a
stranger in public (on Facebook or at an event), or arranging to meet on the street to exchange
the half bonds, or coordinating with friends to cash in on the prize together. Thus the core
choices for the first mechanic are about how players greet and match.
For the mechanic to be meaningful, the choices must affect the outcome of the game.
71

As discussed, the funders and operators desired to build local community and networks. More
importantly, many players also desired to build up Macon, from supporting local businesses to
getting to know the downtown.
72
Did reaching these goals depend on maximizing meaningful
choices in the game?
Two criteria for meaningful choices come from Salen and Zimmerman (2004, p. 35).
They propose that meaningful choices require (a) consequences that are discernable to the
player, so that they can see that their action had some impact; and (b) integration of the choices
in a longer-term understanding of how their decisions affected the big picture in playing. The
first criteria is relatively easy for social outreach like meeting and greeting, since players can
quickly discern that there is an impact from their choices around socializing with neighbors.
The second criterion sets a higher bar. Integration means showing players how their
choices (like who to greet) have consequences for building Macon as a community. The success

71
I am extending the logic of Salen and Zimmerman (2004, p. 37), but emphasizing mechanics as cultural actions.
72
Again, the post-game survey showed that the opportunity to support local business was highly valued by players
(86% valued this “a lot”) – nearly as highly valued as receiving the free money itself. Meeting someone new (which
implies socializing more than community) was not that highly rated, compared to high ratings for “getting to know
downtown” matching the immediate satisfaction of successfully finding a matching bond.
78

of the game depended on participants who recruited intensely.
73
But the game did relatively little
to let players know how their greeting choices will have an impact, like affecting the kind of
community that Macon becomes. In essence, integration means that the game must articulate
community impact, and make visible the progress on its overall theory of change to build
community.
Non-game interventions also rely on integration. For example, ShopLocalMacon clearly
advertises the benefits of supporting a locally-owned business (e.g., they “reinvest in the local
economy at a 60 percent higher rate than chains…”), and gives feedback to the community on
the overall impact of the campaign. Such feedback is a kind of transparency in process and
mechanism, beyond the transparency of intentions. Of course, the choices in Macon Money are
more complex than a simple buy local campaign. But did the game fully support players to make
meaningful choices in the social realm? Consider three social decisions that were theoretically
available to players, but largely unsupported:
• Attempting to meet someone new (rather than just greeting established friends)
• Visiting a business you’ve never tried before (rather than the old favorites)
• Recruit neighbors as players (not just recruiting family)

73
There was a tipping point in the game’s adoption that occurred when it seemed to reach critical mass, partly
driven by highly motivated networkers who spread the game to a high number of people (M. Taylor & Whatley,
2012b); based on interview transcripts, it seems that several of these power players – perhaps most – did so because
they believed in championing Macon to their networks, not because it helped them to find matches.
79

None of these choices are explicitly acknowledged by the game. In fact, the instructions
page for the game on “How to Play” frames all strategies in terms of finding a match with ease,
and gives no tips for advanced social play
74
.
Did the designers miss an opportunity? Quite possibly
75
. There are no recommended
strategies for deliberate social bridging, building the identity of Macon as distinct city, or
increasing community trust. “We were flying the airplane while we were building it,” admitted
one staff person from the Knight Foundation in an interview. Such willingness to take risks is
necessary for innovation; it is also an opportunity to reflect on how to deepen the impact in a
subsequent version of the game.
Perhaps the best evidence of shortcoming in Macon Money as a game is that a number of
players found it too easy; several suggested in interviews that the game would have been more
engaging if there were greater challenges on the social side
76
. Of course, there are always costs
of doing something additional (e.g., adding advanced social play), but there may have been some
low-cost ways of validating the implicit social challenges without affecting the formal game.
Consider two possibilities for how feedback loops can be created. First, players in the
original game were already asked reflection questions as a pair when they turn in their half-

74
At best, the how-to page advises players to be creative in seeking unexpected symbol matches (e.g., “Draw your
symbol set on a piece of paper and tuck it into your car window”).
75
Of course, the craft of game design is notoriously hard – especially with new forms like real-world currency
games. The practitioner community is often surprisingly open about shortcomings of games; the elites behind
Macon Money have excellent professional reputations, and are likely to be some of the first to acknowledge that
their design was experimental and was certain to overlook some areas for optimization.
76
As one player described in an interview, “I feel like it was almost too easy at times. And it was funny the first
time because I was just really excited about, “No, I’m not going to match with you. I’m trying to find somebody I’ve
never matched before.” And I was just very starry-eyed about who my match would be.”
80

bonds and had their picture taken.
77
Yet there are no questions about whether they made any
choices that build a better Macon, like meeting new people, or recruiting others, or celebrating
the city in some way. Simply asking such a question would be a small way to make choices
discernable to other players and validate them in the game’s theory of change. A second
possibility happens at the moment with the currency is exchanged for real goods: why not use the
back side of the currency to invite players to describe why that business is good for Macon.
Their answer could be automatically emailed to the person they originally matched with, and
thus provide a social consequence and feedback loop for their choice. Both of these ideas can be
described as “embedded reflection” – an approach I have described before for real-world games
(Stokes, 2010). In sum, the opportunity prescribed by the framework of meaningful choice is to
make the social choices more discernable and integrated.
Such opportunities can be hidden when the assumption of games-as-rules dominates; by
contrast, the hope of a framework for ‘local fit’ is partly to help to validate the cultural approach
to mechanics, and to include informal social decisions in designing for meaningful choice. The
hope is that the ‘local fit’ framework can help make these opportunities more visible for funders
and evaluators.
Can failure be anticipated? Avoiding likely “failure states” is an important counterpoint
to maximizing for meaningful choice. Three common failures for most games are proposed by
Salen and Zimmerman (2004, p. 65); below, I consider how each indicator could be seen in real-

77
For a live example, here are two winners: http://www.maconmoney.org/news/618/20/Barbara-and-
Toyann/d,winners
81

world games like Macon Money, as well as for civic activities in general that encourage
participant agency:
1. “Feeling as if decisions are arbitrary.” This failure comes when players feel like
their choices make no difference, i.e., that any choice would have been valued by
local organizations. Again, the cultural orientation of mechanics pushes for a
definition of winning or progress that draws more attention to the informal social
decisions made around playing.
2. “Not knowing what to do next.” This failure comes when players are unsure how
to make progress, or when they struggle to make progress and are not sure why.
Again, Macon Money was very clear for the formal mechanics, but could have
easily provided some actions for what else players could do (like form cash mobs,
or invite businesses to put up a “buy local” sticker, or even asking more churches
to recommend events that could help build trust across lines of segregation).
3. “Not knowing if an action had an outcome.” Although this sounds obvious in the
short-term, civic life is filled with activities where volunteers provide many hours
of help but never hear about the longer-term impact of the project. Similarly,
games like Macon Money should assume that players who seek social impact
deserve to hear about the results. The game’s website contains no mention of the
impact study that was done, nor acknowledgement that such a study was intended;
instead, players must visit the website of the national funding organization, the
Knight Foundation. If Fullerton is right that the designer’s job is distinguished by
being an advocate for the player (2004), then real-world game designers may need
82

to see impact studies as part of the game design in a broad sense, and work harder
to ensure that impact is expressed in a frame of interest to participants, not just to
funders and organizers.

In sum, seeking to maximize “meaningful choice” within a real-world game is useful for
many reasons. Most immediately, meaningful choice emphasizes the agency of participants, and
legitimizes their goals in playing the game. As a manipulation counterweight, meaningful choice
places an emphasis on making the impact of actions discernable, and integrated into a longer-
term understanding of cause-and-effect. Since players of Situated Games are often seeking to
build community, the framework of meaningful choice shows the importance of transparency in
process and about the underlying theory of change.
Rejecting alternative frames
To conclude the discussion of mechanics, this study proposes to reject several less-than-
helpful ways that real-world games are often described. In particular, the claims below can
undermine the effort to use games for community empowerment, rather than manipulation.
Consider:

(a) Claim: The benefit of games is that they “add incentives to nudge civic behavior.”
This is a dangerous primary frame, as it encourages the manipulation of citizens. The
frame is also counter-productive because it implies that games are naturally fun –
when in fact, it is incredibly difficult to successfully compete for attention in leisure
time. Moreover, games can be an excuse for people to do things they already wanted
83

to do, like support local businesses and meet their neighbors. The currency in Macon
Money did provide an incentive, but to the extent that participating was a ‘game’ it
also contained deliberately paced challenges to sustain the experience
78
. Reducing
the game to an incentive misses the power of optimizing for meaningful choice, and
seems likely to lead to a less-effective and less ethical game. Mechanics provide a
better way of explaining how games structure participation through challenges and
feedback loops.

(b) Claim: “Winning the game is determined by the rules.” This can be moot for Situated
Games that lack formal win conditions. More importantly, the discussion of
mechanics showed that many players bring their own goals from real-world culture
and the complexities of pre-existing social rewards. Thus it is more accurate to
understand winning in terms of compatibility with community goals. Without this
shift, game designers can forget to tell players how their gameplay affected the
community, and thus fail to give feedback to players on the deeper win conditions.

Yet the greater justification for Macon Money may come from understanding it not as a
game for community engagement, but as socio-economic policy.

78
For example, game operators sought careful balance with the bonds, not simply giving one to each person, but
calibrating the number of likely matches at an event to visible but not universal, thus inviting participation afterward
in online pages to find a match, and to attend future events.
84

As socio-economic policy (reconceptualizing games)
To understand the importance of ‘local fit’, the stakes can be raised by recognizing that
games can provide an informal kind of socio-economic policy. This possibility is a new frontier,
made possible by the hybrid technologies behind games like Macon Money, which can reduce
transaction costs (like managing flows of money) and allow for more interesting sociality to be
layered on top.
79
A policy frame is also useful to reach stakeholders who are struggling to
understand games, including those in social movements who define systemic change in terms of
policy.
Macon Money, for example, has been debated on economics blogs, and like a tax
abatement the benefits can be assessed in terms of the public good. Analyzing the game as a
public program helps break with traditional analysis of games as local media or as public art
80
.
Understanding games as policy also increases the importance of understanding their mechanisms
of accountability, and the related value of ‘local fit’.
Economic activity from Macon Money was undeniable. The mayor’s office in Macon
described the game afterward as “a great local stimulus idea” (Grzeskowiak, 2011). An
economics professor at Columbia University described the money that went to local businesses
in terms that sound like fiscal policy:

79
The technology in Macon Money was minimal, yet still played a vital role. Even as the physical paper played a
vital role, each half-bond had a serial number that was carefully tracked; similar tracking was in place with the
currency that could be exchanged at local businesses. More importantly, the ability to match bonds at a distance was
facilitated by the use of online exchanges that would have been outside the mainstream even a decade prior.
80
For example, Flanagan provides an excellent art history of locative games in urban space (2009, Chapter 6), but
the games are not by or for government, let alone a way of embodying a local community plan.
85

“Macon Money provided an injection of liquidity …an increase in aggregate
expenditure… an increase in production… It was expansionary monetary policy.” (Sethi,
2013)

Government support for Macon Money was largely symbolic, with the mayor playing
himself and generally supporting the game, but not otherwise mobilizing the city apparatus. The
financial capital for the game came from the private Knight Foundation, and much of the ‘local
fit’ was with local nonprofits, like the Knight-funded College Hill Alliance. The Alliance has
won considerable national acclaim for its “creative solutions and inventive ideas” for economic
development.
81

Can there be policy without government? Public policy is – of course – often framed in
government terms (e.g., Birkland, 2001). However, sociology of the modern liberal state makes
clear that much of the applied power and execution is distributed through local nonprofit
organizations and business associations.
82
Private foundations like Knight embody a broader
governing, since their endowments are only possible due to special exemptions in the tax law in
the name of the public good. Knight has thus been delegated a certain amount of policy-making

81
For example, in 2013 the Alliance received awards from the International Economic Development Council in two
categories: Neighborhood Development Initiatives, and Real Estate Redevelopment (“College Hill Alliance
Receives Excellence in Economic Development Award from IEDC,” 2013). Local stakeholders expressed their
belief that Macon Money played an important role in securing these awards.
82
See in particular the writing on governmentality, which describes the “art of government” as extending far beyond
the formal apparatus of power; the term rose in prominence with the Foucault Effect (1991).
86

ability; passing the grant to approve Macon Money was a formal act of policy
83
, albeit one that is
only vaguely accountable to the public.
Master plans of the type that informed Macon Money can also serve as a kind of semi-
formal public policy. The specific Master Plan for the College Hill Corridor was based on an
initial proposal from Mercer University, which then received the support of the mayor and
funding from the Knight Foundation.
84
Formally released in early 2009, the plan claimed
legitimacy in part through a process of public input, including a series of community meetings
that involved more than 1,000 residents. The plan was the basis for the College Hill Alliance as
an organization (with a $2 million grant from the Knight Foundation). Ultimately Macon Money
depended on the College Hill Alliance for guidance, and further asserted legitimacy by invoking
the Corridor Master Plan in press outreach as the basis for intervention.
Live monitoring of the game was necessary to ensure that the currency was balanced –
much like how the Federal Reserve in the United States adjusts the currency supply as part of
fiscal policy. The original design document for the game (which has not been made public) can
be understood as the baseline policy approach. The need for live adjustments further
underscored the importance of local fit. A lead designer for the game described the process:
“We playtested the game by modeling different scenarios, bond and bill distribution and
redemption rates, and attrition rates, etc. Because it’s difficult to playtest this type and
scale of project outside of the community it was designed for we introduced the game

83
The Knight Foundation website lists the Macon Money grant as $492,590 USD, awarded April 1
st
, 2010. Note
that this grant amount does not include the value of donated labor from community organizations (e.g., the College
Hill Alliance). See http://www.knightfoundation.org/grants/20100142/
84
For a history of College Hill projects, see http://www.collegehillmacon.com/index.php/about/project-history
87

slowly, closely monitoring the game when it was live. Bond distribution was tweaked on
an ongoing basis. We closely monitored weekly engagement including matches made
between zipcodes, bonds distributed, redeemed, money spent at local businesses, and
money redeemed by local businesses.” – Senior Producer Kati London (Ruiz, Stokes, &
Watson, 2012 – see: http://civictripod.com/interview-kati-london/)

Seeking balance in games is a central act of game design
85
. Implicitly, the design of
Macon Money became a kind of fiscal policy, including a mix of formal rules and live balancing
of bond allocation across zip codes and demographics by mail, through media marketing, and
through the selection of specific public events for distribution.
The subtle ability to ‘balance’ the currency system of Macon Money hints at the power of
games as vehicles to enact policy. Compared to tax incentives or campaigns like “buy local,”
games can facilitate economic exchange at a more micro level. Technology is part of the
difference, and will play a growing role in making games viable as public programs with
transparent accountability. Future versions of the game would likely track the bond halves with
digital technology rather than by hand. (Staff interviews underscored the demand for technology
that could improve the efficient handling of the money.) Operating the game with more
technology would likely increase the rigidity of the game, and thus its ability to be externally
monitored and held accountable to local priorities. Codifying the game into technical algorithms
is a way of using code to embody policy.

85
There are actually many ways to balance a game; balancing games iteratively with successive frameworks for
balance is one of the most important and insightful kinds of game design. For a useful overview, see Schell’s The
Art of Game Design: a book of lenses (2008).
88

The legal risks of Macon Money underscored the real-world economic frameworks
inherited by the game. Georgia has gambling laws like most American states, and legal
investigations took up considerable time for the game designers and local partners. One result:
the game was determined to be a “contest” requiring skill (not chance) – as echoed in the game’s
extensive terms of service
86
, and in the lengthy contracts that were required of participating
businesses. Other American cities in the future can borrow the legal framework and adopt it to
their states, much as policy is often distributed.
The cost of the game as a standalone was too high, according to Knight and local
stakeholder interviews. Most of the cost was in staff, not currency. The basic currency flowing
to businesses totaled $60,000. At that cost, each participating business might receive $2,500 if
funds had been distributed uniformly. Staffing costs were the greater expense, and involved
outreach and facilitating the game at local events and in the main office. To reduce staff costs,
future versions of the game might integrate into local organizations with aligned missions (like
the visitors bureau, Main Street Macon or the “Gateway Macon” Initiative), who would need
additional funding, yet are also likely to consider matching outside funding, provided the game
helped generate new clients for the long term.
What are comparable local funding options? Pure cultural programming is notoriously
hard to fund. The Second Sunday concert series, frequently mentioned in interviews as a social
analogue to Macon Money, has an estimated price tag of perhaps $20,000 a season
87
for
approximately seven events with live music. But funding is temporary, even for the broader

86
See http://www.maconmoney.org/terms_of_service
87
According to interviews with local Macon stakeholders; this figure is rough, since considerable in-kind
contributions are made to allow the Second Sunday series.
89

College Hill Alliance, which runs the concert series. Knight is only funding the Alliance for a
limited period of time; similarly, Knight has never indicated it would support the game for a
second run. Like many national funders, Knight privileges its role to fund experiments, and
leaves subsequent adoption to the marketplace.
For sustainability, Macon Money may depend on shifting the perception of the game from
cultural programming to being an economic strategy, tied to regional development policy. Keep
in mind for scale that the city government presides over the broader county with approximately
150,000 residents and a median household income of $38,000; the city itself has a much smaller
core of just 15,000 residents.
88

Consider the public-private partnership NewTown Macon
89
, with its more explicit focus
on place-based growth, especially through real estate. NewTown asserts that it has “incentivized
business” on the order of $1 million in investments annually, netting almost nine new businesses
a year, and creating more than 50 jobs a year.
90
The group managed some community events,
such as a First Friday gathering for art and local food.
Compared to many government initiatives, NewTown appears advanced and open in its
economic development strategy. In most cities, economic development is rudimentary. As a
professional arena, economic development is quite new in regional government, and most
professionals are accidental practitioners (Blakely & Leigh, 2010). City staff are often quite

88
According to the 2012 American Community Survey (5-year estimates); see
http://factfinder2.census.gov/bkmk/table/1.0/en/ACS/12_5YR/DP03/0500000US13021
89
NewTown Macon also sponsors a local plan, e.g., see their Strategic Action Plan for 2011 at
http://www.newtownmacon.com/wp-content/uploads/2011/11/NTM_StrategicActionPlan_FINAL.pdf
90
These numbers are based on the claims made by NewTown Macon website in 2013 that since its 2006 launch it
has netted 61 new businesses, leveraged $7 million in investments and created more than 350 jobs.
90

disconnected from academic research on economic planning (E. Currid-Halkett & Stolarick,
2011). Most city efforts are a simple mix of recruiting industry and marketing the city as a good
place to work. Even new technologies like mobile apps are by default focused on marketing and
branding a city.
91
Only recently have cities begun to seek strategies of human capital (Blakely
& Leigh, 2010).
Of course, financing strategies are hard to align with building social capital. For
NewTown Macon, the best example of mixing socializing with business is the aforementioned
First Friday event
92
, where merchants stay open late and the public visits for art exhibits and
dinner specials. But there are no mechanisms for connecting residents across zip codes, or of
promoting specific local businesses. Similarly, the ShopLocalMacon website (operated by the
City via Main Street Macon) offers very little in terms of fostering social relations or local
culture. Their own effort to offer a micro-loan fund (see Figure 14) stands in stark visual
contrast to the localized imagery of Macon Money.

91
For example, consider the iPhone app called GNO, which was developed for the Greater New Orleans area and
launched in 2011. The sponsor was a local economic development alliance. When they released the app, the
alliance claimed that GNO would deliver “easily accessible information about our region's competitive business
advantages” (Smith, 2011). Branding the city of New Orleans in a relatively low-cost way seems to have guided the
technology development process, according to news reports (Sherman, 2011).
92
In 2006, the City Council of Macon reportedly asked NewTown Macon to start managing first Friday; details on
the program are available at http://www.newtownmacon.com/projects/friday-fest/
91

Figure 14: Advertisement for a Macon micro-loan fund

Source: the Shop Local Macon website
93
as of March 2013

What if Macon Money were seen as tax policy? To be sustainable and cost-effective, any
future version of the game “has to be seen as a development tool, so people invest in it,” said a
local funder who helped run the game. So what if instead of a tax incentive, the city put similar
funds into Macon Money. In other words, rather than reducing tax fees (e.g., how the city
currently provides historic and new market tax credits), the city would provide businesses with
bonds from the game. The city’s investment would be a kind of focused investment capital, not
to offset costs but to deepen ties with new customers. In essence, Macon Money would be
understood as a way to help a business to promote, brand and market itself. The approach
matches how businesses already see Macon Money. (In surveys, most businesses said their
principal motive for joining the game was to promote their business more widely (M. Taylor &
Whatley, 2012b).)
Local business groups might also fund part of the game. For example, a business might
contribute 50% of the cost of a bond in order to get new business. Already, Macon has a
Community Improvement District (CID
94
), which operates on the premise that business owners
are willing to tax themselves. Self-taxation makes sense when the collective benefits advance an

93
See http://www.shoplocalmacon.com/
94
Akin to what other cities may call a Business Improvement District (BID) or Special Improvement District (SID)
92

individual business more than the same funding delivered piecemeal. The twist is that every
business that participates is implicitly spreading the idea of the larger community, building the
social fabric for residents, and using a vehicle that can help compensate for the injustices of
geography by over-sampling for specific groups and zip codes.
The greatest challenge may be in aligning funding from a broad set of associations,
redevelopment organizations and local foundations. The role of government would likely be
invaluable, if only to provide basic oversight to prevent abuse (Sethi, 2013). Given the risk,
funding may need to come from a private foundation like Knight. Yet on this second round, the
innovation would need a different focus: identifying how to develop the tax and partnership
structures to embed the game at low cost.
Ultimately, investments from the city and the business community are only viable if the
benefits can be shown to accrue locally, not disperse. Once again, the deep and longer-term
challenge may be in further integrating Macon Money with local redevelopment organizations
and city departments.
Hints of more ‘local fit’
This chapter primarily emphasized fit in terms of mechanics and local meeting modes.
Despite their power, mechanics are only the first way to understand how a game can ‘fit’ to the
local community. The next chapter will explore how the game matches to the community’s
identity and modes of communication.

93

CHAPTER THREE: REALITY ENDS HERE
& NETWORKS TIED TO A UNIVERSITY

“The most natural privilege of man, next to the right of acting for himself, is that of combining
his exertions with those of his fellow-creatures, and of acting in common with them.”
- Alexis de Tocqueville (1848), Democracy in America
Overview of the game
This chapter investigates the game Reality Ends Here over three years. The case study
begins by demonstrating network analysis as an important method for Situated Games. Then the
second two dimensions of ‘local fit’ are explored, including with a contrasting case of a failed
attempt to replicate the game at a second school. As the analysis will show, localizing Situated
Games may require modifying the core game.
The game Reality Ends Here first launched in the fall of 2011at a prominent film school
in Southern California – the USC School of Cinematic Arts. For a university use of games, the
approach was bold since the game was seen as a structure for collaboration, rather than to deliver
pedagogical content or as an incentive for academic courses.
The game started in secret, led by a team of faculty and graduate students, and was never
openly promoted or played in classrooms during its semester-long run. Instead, students were
drawn into the game via rumors, secret websites with countdown timers, and a mysterious black
flag. Each student collected a deck of brainstorming cards, which they pooled when forming
teams. Combinations of cards served as creative prompts, spelling out a specific media-making
challenge. For example, a group prompt might be to create a “Silent Short” film using specific
94

places on campus and specific props. Completed projects were awarded points based on which
cards the team used.
One card featured Mary Pickford (see Figure 15), a silent film performer who is an
industry legend and donor to the school; she co-founded the film studio United Artists. Similar
“collectible” images were on the backs of all cards, featuring key figures from the school’s past
and present. Together the stories gather a shared history for the community, given to players in
discrete pieces and hinting at the alumni network that players can increasingly benefit from – and
even access in the game.

Figure 15: Cards used in the original game
include collectable profiles tied to the school (Mary Pickford, at left);
at right, the cards line up to form the prompt:
“Make a Silent Short film in the SCA Courtyard”

Source: Jeff Watson

The back of the Pickford card indicated that the team’s creation should be a “Silent
Short” film. As shown in the figure above, the card can be paired with a place card like “The
95

SCA Courtyard,” which is a central gathering space for students at the school. Already this very
simple combination prompts students to make a connection to physical and social spaces.
Each player who joined the game was given a packet of 10 printed game cards, selected
from a pool of over 500. Each card had a digital presence, and its use was tracked and celebrated
online. Specific networking instructions were sometimes encoded in the cards, such as requiring
the involvement of specific people.
95

Quickly teams discovered that more complex “deals” of cards were worth more points.
Arrangements of cards began to take place in private and public space, from dorm rooms to
school hallways (see Figure 16). Students filming each other in public space created further
visual spectacle, and raised questions about “what are you doing?” and “How can I play?” in the
broader community.

95
For example, each orange card has a random student’s name printed on its face. Involving that student in a project
where their orange card is a part of the Deal earns every collaborator on the project extra points. Initially, the names
on orange cards were restricted to freshmen peers. However, in year two, the design goals evolved and the orange
cards expanded to also include 25 sophomore students who had participated in the game the year before. The shift
was designed to increase inter-cohort collaboration, and could similarly be modified to target other proximal
networks.
96

Figure 16: Playing the game Reality Ends Here in person.
Teams pooled cards to form creative prompts (left),
often involving video creation (center); to receive points,
players must justify which cards were used by their team (right)

Source: “Reality: The Manual” (Watson & Reality Design Team, 2013)

Most of the community only learned of successful projects when they were posted online.
Discussion quickly followed the more provocative submissions, including a mix of positive
encouragement, debate about what made the project worthwhile and everyday socializing.
Group reflection was structured through an ingenious requirement that teams must visit the
“game office” in person after each finished project and record an impromptu video that explained
their creation.
Video projects were common – it was a film school after all – but students also dove into
many other media. For example, fabricating love letters was part of the creative writing project
that emerged in a deal the students called “Letters of My Lai,” featuring a Vietnam-era cover up
(shown on the website in Figure 17).

97

Figure 17: The “Letters of My Lai” project on the game website

Source: “Reality: The Manual” (Watson & Reality Design Team, 2013)

Keeping the game underground and hidden was celebrated, even as portions of the site
were public and highly visible. Cards for the game were often hidden around the school, but
were sometimes found entirely by accident. Even knowledge of the game itself was often denied
by faculty.
Unexpectedly meeting famous alumni was a highlight of the game for many players. At
one point in the first year, several students with the most points were told to meet at a secret
place and time. No names were provided. Upon arriving, the students were greeted by a
formally dressed Oud player, who played songs while they were driven to a local museum
(Watson, 2012, p. 233). Inside the museum they met two alumni of the school, Jenova Chen and
Kellee Santiago (highly celebrated game designers of Flow, Flower and Journey). For the next
90 minutes, the students had an exclusive conversation with the designers about their field, game
design, art and media. Compared to the usual alumni events, like attending a panel discussion or
a cocktail hour, the access was incredible; yet the reward was for more than just the player: many
98

alumni were excited to be engaged with talented students and the chance to have a meaningful
conversation about their work and the field.
Such tactics of covert experiences and unexpected alumni are foreign to how most civic
and school-based education programs operate, yet Reality had institutional support from the
school. Overtly, the game sought to foster multimedia production. Implicitly, the game had
strong network goals: to catalyze collaboration across focus areas in the school, building a
coherent sense of community tied to place, and fostering networks that would persist into the
future.
Forming networks – as an institutional goal in education
Strong networks are associated with increased life opportunities for students, including
better jobs and greater civic participation (Levine, 2007; Putnam, 2000). Unfortunately, most
schools do relatively little to teach network-building skills or to actively shape network structure.
In the United States, it is easy to blame inaction on the current regime of standardized testing,
which focuses attention on the cognition of individuals (Shaffer & Gee, 2012). Yet the problem
is also structural, since individual teachers rarely have the capacity to build networks beyond a
single classroom, and perhaps a few guest speakers. This case study investigates how games can
structure a different kind of learning environment outside the classroom, explicitly targeting
network growth and quality.
99

The catalyst for creating Reality Ends Here was a strategic review of the school. Initiated
by the dean, a faculty “futures” committee
96
had called for strategies to increase collaboration,
especially across disciplines.
97
Behind the call was a network problem: graduates in film and
media were not sufficiently connected across disciplines. Screenwriting students did not partner
with animation students, interactive media students did not collaborate with production students.
In sum, there were less-than-optimal levels of creative cross-pollination and skill-sharing
across the five divisions of the school (Watson, 2012). The committee asserted that the
professional success of their students depended on their forming interdisciplinary teams to tackle
creative problems. In addition the committee recommended deeper connections to alumni, since
the alumni network can be vital for film sector jobs.
Professional networking and collaboration rely on mechanisms that are often hidden. For
media-making, the collaborative standards are often implicitly set by the practitioner community,
from film to independent games. In general, communities of practice (Lave & Wenger, 1991) are
particularly known to rely on tacit learning and collaboration. Success depends on knowing when
to bend the rules, and under what social contingencies. In other words, success in the media-
making community may depend more on the mastery of cultural tools and “knowing how,”
rather than “knowing what” (per Herrenkohl and Wertsch (1999) in Bagley & Shaffer, 2010).

96
Specifically, the Envisioning the Future Group (EFG) was tasked by Dean Elizabeth Daley, led first by Prof.
Steve Anderson and then by Prof. Holly Willis.
97
For example, the growth of online courses may increase the importance of a strong network strategy for elite
universities, since content is becoming cheap online (relative to network formation, which often still is best face-to-
face). Additionally, the innovation skills increasingly targeted by elite universities may rely on the generation of
ideas at the edges of increasingly siloed academic disciplines.
100

In response to these challenges, Reality Ends Here was created as an extracurricular game
(Watson, 2012).
98
The game is not a network simulation, but rather directly integrates with the
social and media-making lives of university students on a physical campus. The game seeks to
cultivate passionate affinity spaces
99
around core mechanics of social interaction, fostering
know-how that is cultural and situated in localized networks.
Although Reality Ends Here was covertly supported by the university administration, the
designers fought to keep it hidden and separate from the formal power entrenched in course
structures (Fullerton, 2014). Spreading the game by word of mouth and physical clues set a tone
that student agency is valued: to make progress, you must be proactive. Rewarding discovery is
foreign to most school-based education programs, and the obscurity runs counter to
“gamification” approaches that simply reward established learning goals and tasks (Lee &
Hammer, 2011).
Little prior research existed to guide the initial design, especially for games that seek
immediate impact. Educational games are traditionally preparatory, with skills and content to be
applied later. In other words, ‘transfer’ is often assumed to be necessary between game training
and future application. However, the basic premise of transfer may be unnecessary if the future
and present contexts can be collapsed (Thomas & Brown, 2009). Thus network games can go
beyond training and transfer if they can deliver social structures that are immediately useful,
catalyzing network ties that are not so much preparatory as persistent.

98
The game was created by Jeff Watson as his doctoral research and design project (2012); co-design credit is
shared by Watson, Simon Wiscombe, and Tracy Fullerton.
99
The work on ‘passionate affinity spaces’ by Gee and Hayes (2010) shows how even commercial games with no
networking goals can cultivate networks of informal learning. This work shows how the most important learning is
often at the edges of the official game, in cultural spaces of peer reflection and skill sharing.
101

Visualizing ties: the network emerges
What does the game’s network structure look like? A basic network visualization of the
game can simply show the ties between players and projects (Figure 18). The players (red
circles) are shown connected to the projects they worked on (blue squares). The most tightly
connected players are grouped at the center, while the more isolated players in the game are at
the periphery.
100
Such diagrams can be powerful visuals; they result from simulating what it
would look like if the circles and squares followed the laws of physics, where relationships pull
people and objects together.
101

Figure 18: Contributors (red circles) and projects (blue squares)
in the first year of the game; the two pink circles highlight groups of interest

The visual above anchors the mundane statistics: in the first year (or “season”), there
were 119 projects created. There were 103 students participating in the first year and 80 in the

100
A more detailed description of the data used will follow in the more detailed network analysis below.
101
Specifically, the diagram represents what it would look like if all the circles and squares were trying to be far
apart from each other, but every time there is a network pairing then those two items are pulled close to each other.
102

second year (who completed one or more projects). Across two years, the participation rate was
almost half of available students (47%). Curiously, a small number of participants were not
students, since the game was deliberately left open.
Clusters can hint at patterns behind the data. Two patterns are highlighted by hand with
pink circles in the figure above. The lower pink circle shows a group of at least five players who
anchored a series of projects together. The upper pink circle shows two students who
collaborated on many projects with the first pink circle group; these two students were also
influential in drawing in many of the outlier students at top right, who only did one or two
projects. In other words, the two circled students play an important role with peripheral members
of the network. One implication beyond schools: community builders might proactively respond
by seeking these two as allies, asking how they found partners, and encouraging them. (Further
discussion of community building and neighborhood organizing strategies comes in a subsequent
section.)
The typical group in the first year included eight people (per the median). Curiously,
some team sizes were strangely common – in particular, there was a spike for six-person and 24-
person teams, as shown in Figure 19. The six-person group appears to be a natural size for media
making collaboration, especially early in the semester. By contrast, the 24-person group size
hints at a very different phenomenon that took place later in the semester. What was going on?

103

Figure 19: Team size frequency distribution (multi-year)

Behind the spike is an emergent social behavior, not anticipated by the designers.
Midway through the first year of the game, a group of students decided to formalize their group.
Calling themselves MARRA (an acronym for their founding member), the group created an
“exclusivity contract” that prohibited members from working with other players in the game, and
required pooling cards. In other words, they added their own rules atop the game. After the
group’s score rocketed up, another group mimicked the strategy, calling themselves The Tribe
(see Figure 20).

104

Figure 20: Members of the Tribe, from a film they made in the game

Source: screenshot courtesy of Jeff Watson, original by the Tribe

The large number of 24-person projects points to the success of this large and somewhat-
stable team. Stable teams work differently, and may even deserve terms like ‘club’. At this
scale, the organizing is collective.
Toward network methods for community games
As this section will demonstrate, networks around a game can be analyzed – and point to
insights at the collective level. Although the case study takes place in a University setting, this
chapter will repeatedly apply findings for use in neighborhood building and community
organizing.
In related research outside of games, network methods have been applied to investigate
self-assembling project teams that are tracked purely online (Zhu, Huang, & Contractor, 2013).
In terms of games, network analysis has focused almost exclusively online (e.g., Steinkuehler &
Williams, 2006), at best pushing to look for how online games build capital that can transfer to
the physical world (Zhong, 2011).
105

This study shows how some of the same approaches can be applied to place-based
communities. The goal is to deepen the conceptual perspective of ‘local fit’ more than to prove
any single hypothesis (although a few useful tests are demonstrated). Research and analysis in
this study was led by the author in collaboration with the game’s designers.
102
Several formative
questions guided the investigation, based on the framing of ‘local fit’. The guiding questions
were:
1. How did teams form, and how did the network grow? Did game mechanics (e.g.,
incorporating the point system) advance or hinder network formation and quality?
2. How might community organizing and teaching be informed by network analysis? What
decisions might network analysis help teachers to make?
3. Was the resulting network healthy? Was there a bias toward team formation by gender
or sub-group (e.g., by academic discipline)?
This study will show that the game’s mechanics are indeed tied to network effects
(specifically, that score correlates with network centrality). Team formation is shown to be
biased by gender and academic discipline, but appears within acceptable levels. Several
approaches are also discussed for how real-time analysis can inform community organizing.

102
Collaborators include Jeff Watson, Tracy Fullerton and Simon Wiscombe; an early version of these findings was
co-authored with these collaborators at a games conference (Stokes, Watson, Fullerton, & Wiscombe, 2013). This
study was only possible with their support, time and feedback.
106

Data for network analysis
Data was drawn from two years of the game, each year involving more than 100
participants
103
and producing hundreds of team projects. Relationships were inferred between all
collaborators on a group project (bidirectional). Players who completed multiple deals together
were assumed to have proportionally stronger ties.
Concretely, network data was collected by scraping the game website, which listed
collaborators for each completed project. The data was processed using custom scripts and UCI
Net software.
Game score as network indicator
As the formation of the MARRA and Tribe demonstrated, groups often emerge
organically and forge their own strategies to secure points. If the game is designed to reward
collaboration, the points system should be robust and reflect collaboration and networking – even
when teams like the Tribe create their own strategies.
Points in the game are awarded equally to all team members on a project, with more
points awarded for involving more cards. Since cards are scarce, players must recruit peers to
join them if they desire higher scores. Thus the points reflect social collaboration, and

103
Participants in the study were the incoming undergraduate classes at the USC School of Cinematic Arts in the fall
of 2011 and 2012. According to the University, the overall undergraduate mix for undergraduates of all grades is
approximately 60% male, with approximately 10,000 living alumni and 850 enrolled students (“Statistics-At-A-
Glance (2010-2011): USC School of Cinematic Arts,” n.d.).
107

necessitate team formation. To encourage new group and card formations, cards decline in value
with each use.
104

Over many rounds of play, does a player’s score indicate their value in the network? To
measure we can calculate the network influence for each player, also known as their “centrality.”
(The specific measure is called eigenvector centrality; high centrality numbers are akin to being
ranked highly in Google search results: it means that the player is connected to the most highly
connected other people.)
We expect that players with high scores should also have more central roles in the
collaboration network. The result can be determined by look at a graph that lists each player
according to their score and centrality (see Figure 21).

Figure 21: Players' scores as a function of their network position (year two)

104
Once a card has been used three times, it is no longer worth any points at all. Because cards are a limited resource
in the game system, the more a player can connect with other players, the more cards they will have at their disposal.
108

The result: there was indeed a considerable connection between high-score players and
their network influence. More than 50% of the variance in player score could be explained by
that player’s network role (as shown by a simple linear regression of score and centrality that
yielded an R
2
of 0.522, significant at p<.001). In other words, the game’s point system is indeed
substantially linked to the network position of players. Thus the game’s point system does seem
to provide good network feedback to match the goal of encouraging network formation.
Testing network outcomes tied to fit
What are “good networks?” The network analysis approach is next demonstrated with
several common network indicators: (1) group cohesion; (2) fit to external demographics; (3)
performance by existing groups or disciplines; and (4) mixing across groups.
1. Cohesion over time
Over the semester, collaborations may solidify into more established groups – like
MARRA and the Tribe. Can such growing cohesion be seen visually and quantitatively?
Visually, differences are clear between halves of the semester (see Figure 22), especially
the growing density of connections. The second half appears much more tightly connected in the
center, with the peripheral students appearing comparatively isolated. (The semester midpoint
was determined according to project submissions, so that each “half” includes an equal number
of submissions; there were ninety two.)
109

Figure 22: Network evolution over one game (a semester).
At top is the entire semester, while the bottom represents the first and second half.
TOP: A full semester, drawn from 185 group projects submitted, which involved 110
students; node size is proportional to that student’s score for the semester.

First half of the semester:

From the first 92 group projects,
which involved 100 participants
Second half of the semester only:

From the last 92 group projects, with 74
participants (64 returning;10 new )

110

Appearances are valuable as a first glance at network data, but are rarely sufficient to
make conclusive claims since the visuals represent just one solution to representing the data, and
may hide overlaps and alternative solutions.
The visuals can be backed with quantitative measures of the network’s density. The most
intuitive measure is the typical number of teammates for a project, which increased from a
simple average 10.5 participants per project in the first half the semester, to 12.0 participants in
the second half. However, this measure of density does not consider the overall network size –
and there were nearly 20% fewer players in the second half of the semester.
Better yet, consider the network indicator for ‘density’, which indicates whether more
individuals are well connected to one another (relative to a baseline where each participant is
connected exactly once). Looking at the second year, the average density increases considerably,
from 1.0 in the first half of the semester to 1.7 in the second half. (This difference in densities is
unlikely to have occurred by chance.
105
) There is likely an attrition bias behind some of this
density increase.
106

To some extent densities can be compared across years. For the first year of the game, the
average density was 1.8, while the second year had a higher average density of 2.7. However, the
baseline number of participants also shifted across years, as did many social game dynamics, so
this cross-year comparison should be considered with caution; better inferences can be drawn

105
According to a bootstrap paired sample t-test at the p<.01 level. The densities were both calculated using the
same baseline of 110 students who participated in one or more projects during the semester.
106
Attrition during the semester of play is disproportionately likely to leave behind the most motivated and active
student teams. In the early weeks, some dabbling players may give the system a try, and not return or build up a
larger network. Thus it is hard to tell how much of the shift is due to a loss of low-networked individuals (who might
have undermined several projects), or an increase in the networking performance of those who stayed.
111

once the game has stabilized in terms of participation and card content. In these early days of
Situated Games, the primary benefit of ‘density’ for designers may be its well-established
formula as a means to calculate points, and its use as a metaphor to optimize game mechanics.
2. Fit to external demographics
Shortly into the second year, I observed a brief “mentorship encounter” that was
engineered by the game designers. The occasion was a guest lecturer in town to speak to a class.
The speaker-to-be was intrigued by the tales of Reality Ends Here, and was challenged to join
the game and meet some of the students. Agreeing, the speaker went to a designated park bench
and waited. Suddenly, students started zooming up on bicycles and running over. They had seen
a Twitter post declaring that a game card might suddenly be available, and naming the bench
location. “Do you have a card?” They asked him, eying the yellow envelope in his hand.

Figure 23: a mentorship encounter in the game

Source: photograph by the author

“Yes,” the guest replied, “but first I want to hear about one of your recent film projects. I
hear you’ve been practicing elevator pitches – so if yours is decently brief, you’ll get a card.”
112

The students were understandably a bit nervous and flustered. Yet the point was
legitimate: filmmakers may need to give an elevator pitch on a moment’s notice, and it’s
important to be able to describe your work to outside guests. Meanwhile, the guest speaker was
provided with a structured experience to interact with the students outside the classroom, yet still
tied to their work.
As the students took turns giving their pitches, a gender separation became clear. Three
women stood together, while groups of young men stood apart. After the mentorship encounter
ended, I wondered: is there a broader pattern of gender segmentation?
Gender balance is a kind of network health that community organizers may wish to track
in real-time. Consider the first year, when 67% of players appear to be male and 33% female
based on an analysis of their names, profile pictures and internet searches – a slight skew toward
the male beyond the school’s more usual bias of 60% male. Do women bond disproportionately
with women, men with men, or cross-gender? In other words, we might be concerned whether
there is homophily for men and women within the network.
To test for gender homophily, statistics can be used to determine whether there is gender
bias in pairing. Using a method called QAP, we found two results for the first year of play. First,
men are slightly more likely than random to have partnered on projects with other men (p=.044);
and second, men are slightly less likely than random to have collaborated on a project with a
woman (p=.038). The results are shown in Table 1; those in bold were significant (at the p<.05
level; the MD-QAP was conducted for a categorical auto-correlation with 10,000 permutations).

113

Table 1: Gender homophily results in the first year
Node pairings
by Gender
p >= Diff p <= Diff Expected Observed Difference
A. Male-male 0.044 0.957 752.078 871.000 118.922
B. Male-Female 0.963 0.038 752.078 682.000 -70.078
C. Female-Female 0.935 0.069 179.845 131.000 -48.845
3. Performance by existing groups or disciplines
A second kind of healthy diversity is cognitive, aligning with the interdisciplinary goals
of the game. Outside of the university setting, cognitive diversity is needed for many community
issues including neighborhood planning (Page, 2008). In Reality Ends Here, students can
collaborate across the five undergraduate departmental divisions in the school.
To investigate group diversity, ethnic or cognitive, network analysis provides a set of
concrete tools. As Table 2 illustrates, the participation rate for each division varied from 33-67
percent (that is, the percentage of students within the division who played the game). This
represents modest success, in that all departments did participate. At the same time, the least
participating departments (like Critical Studies and Animation/Digital Arts) had rates of
participation as low as half the most active departments (like Interactive Media and Writing).

Table 2: Participation by discipline (academic major)
Department (both years) Partici-
pation
Submissions
(per player)
Performance
in Points
Total
Players
Interactive Media 67% 14 +6% 23
Writing (Screen and TV) 64% 20 -9% 37
Production (Film and TV) 44% 26 +3% 59
Animation and Digital Arts 37% 15 +17% 12
Critical Studies 33% 13 -6% 26
[Average] or Total [47%] [20] –- 157

114

The table above also includes a calculation of the overall “performance” of each
discipline or major, measured here as the point average compared to the overall game average.
For better or worse, some of these trends conform to department-based stereotyping: production
students are supposed to excel in networking and project management (see their high number of
average submissions), while Critical Studies is a discipline that is sometimes perceived as being
more independent. Yet it is also possible that some disciplines do their networking differently.
Critical Studies students may simply be avoiding networking through media making,
preferring instead to network as part of discussing films.
107
Additional research is needed to
tease out a deeper picture for departmental affiliation, including their networking preferences and
media making biases, and whether such differences are heightened at professional schools or
after graduation. Regardless, since at least a third of students in each division participated (and
at most about two-thirds), the game still had substantial penetration across departments.
4. Inter-group mixing
Are groups partnering fairly, or with a bias toward their own type? Consider the second
year: across the five majors/disciplines, there was only evidence of homophily (the bias to
partner with people who are similar) for the Production department students; this is unlikely to
be a statistical fluke.
108
An overall model fit for homophily by discipline was significant, but
was a mild factor.
109
In sum, there is some evidence for homophily among the Production
students, but it not a strong element. The cause remains hard to answer conclusively given the

107
This alternative explanation emerged through my interviews with faculty at the school, and demonstrates the
value of complementing network analysis with qualitative validation afterward, especially with network participants.
108
The bias measured was significant at the p<.01 level in a network autocorrelation using 5000 attributes.
109
The model fit showed an R
2
that only explained 6.5% of the variance.
115

overlapping factors involved: is it that Production students self-identify based on personality
traits to overcome differences, or is it a skill lacked by others, or simply a behavior rewarded in
the Production department?
Guiding organizers with network feedback from games
Network analysis may be especially valuable in a formative mode, providing community
participants and organizers with what game designers refer to as ‘state information’ about the
health of a network. For learners, live feedback about network features can shift behavior
(Gamberini, Martino, Scarpetta, Spoto, & Spagnolli, 2007; Gamberini, Martino, Spagnolli, Baù,
& Ferron, 2011); such feedback is already part of the game, and can be further optimized. For
community organizers, real-time information can be integrated into daily decision-making,
helping to shape targeted learning interventions.
A visual sense for complexity
One benefit of network visualizations (like Figure 24) is in how they synthesize very
disparate types of data. Rarely do educational facilitators or community organizers have the time
to parse rows of statistics, let alone seek out curious intersections in spreadsheet columns. To be
realistic, the data must lure facilitators into asking questions, especially about how they might
increase the performance of network participants.
Several attributes are clearly powerful to visualize, in addition to those mentioned
previously:
• Gender of the student (men can be triangles, women circles, or vice-versa)
116

• Media type submitted most frequently by the student (such as a 30-second short
video, or a screenplay – give the dominant type a color)
• Number of collaborators that students typically work with on projects (this can be
proportionate to their size)
• Attitudes like self-confidence and collective efficacy (which we collected via a
separate survey)
The resulting visuals are colorful, and occasionally quite surprising. Beyond the
educational use as formative assessment, researchers can also use the qualitative visuals to shape
research questions. We might call this approach “mathematical ethnography,” to borrow a term
from Valente at al. (2004, p. 1702).
Identifying leaders and bridge builders
The job of a community organizer depends on their ability to identify leaders. If the
university wanted to unite the MARRA and Tribe groups, which students can bridge the gap?
Can network analysis help?
Leadership can be understood in terms of networking behavior and positioning (as
opposed to personality traits). As with the prior discussion of points, each player can be
evaluated for their network centrality. In the table below we consider the top point earners in the
game and compare their performance with three different measures of centrality. This
comparison is useful in understanding the network implications of the point system, and to
identify unusual participants.
Often the same player has quite different scores for different kinds of centrality (see
Table 2). The three centrality measures used were: (a) eigenvector centrality, which as discussed
117

above is akin to their influence, especially with peers who are highly networked; (b)
betweenness centrality, which measures those who are most in the center, and useful for finding
people who might have the ‘most control’ over the communication of others; and, (c) degree
centrality, which is a somewhat simplistic but intuitive measure defined as the number of links to
others in the network.

Table 3: Highest scoring players, alongside their centrality rankings;
yellow highlights a row with unusual contrasts
Player Score
(and score
rank)
Eigenvector
Centrality
(rank)
Betweenness
Centrality
(rank)
Degree
Centrality
(rank)
21055 (1) 0.17 (4) 117.663 (11) 73 (5)
19150 (2) 0.169 (5) 347.691 (1) 77 (2)
18120 (3) 0.173 (1) 211.348 (5) 78 (1)
17660 (4) 0.163 (9) 104.289 (12) 69 (8)
17630 (5) 0.171 (2) 218.524 (4) 77 (3)
17620 (6) 0.163 (10) 84.322 (14) 68 (9)
16465 (7) 0.171 (3) 154.605 (7) 75 (4)
16445 (8) 0.165 (8) 67.367 (18) 68 (10)
16425 (9) 0.079 (53) 269.026 (3) 45 (30)
16340 (10) 0.16 (11) 69.948 (17) 66 (11)

Disharmony can be particularly instructive. For example, consider the 9
th
row,
representing the 9
th
-ranked point earner. This student places in the top 10 of all point earners, but
has a much lower rating for centrality – they simply do not have many collaborators. Yet the ties
118

they have are particularly powerful in connecting across sub-groups, since they are ranked #3 for
“betweenness.” In other words, they are a rare link between cliques
110
.
For community organizers, a particular value of this perspective is to make timely
interventions during the course of the game. In particular, organizers can look past the
leaderboards of top point earners to identify “influential” community members (e.g., using
eigenvector or betweenness centrality).
To identify emerging groups
While groups like MARRA enjoy their infamy, others may be harder to spot. Network
analysis can complement the observations and intuition of facilitators by analyzing implicit
groups. Using a technique called cluster analysis
111
, players were partitioned into probable
groups. Each group can be shown visually using a unique color (see Figure 24 for the first year
network).

110
Such people are particularly powerful for disseminating new across groups in communities, and more broadly
align with some of the “power of weak ties” discussed famously by Granovetter (1973).
111
The specific partitioning algorithm is known as a K-core analysis, which is designed to find sub-groups in the
network. K-core analysis is a recursive pruning strategy; here the analysis used 24 of 34 possible cores.
119

Figure 24: Suggested groups based on participation in the first year of the game.
Color indicates group, and size is proportionate to their bridging role (betweenness centrality).

When advising community leaders, making sense of such visuals can ironically be easier
with a bit more data. In particular, the names of participants (withheld here for privacy reasons)
can make the graph come alive. Additionally, hints of an individual’s implicit role is also
exciting for facilitators who already know the community members. For example, the graph
above indicates which participants are ‘bridges’ to other groups by making their nodes larger
(i.e., sized per their betweenness centrality).
In hopes of empowerment, the data can also be directly given to community members.
Imagine a community meeting where a spreadsheet is handed out, telling everyone to assemble
into the prescribed groups. Then all participants could look around the room, and see where they
seem to be positioned according to their actual behavior in the game. Such moments for
120

augmented reflection (also called social translucence
112
) could help communities see race and
gender bias and self-correct. In general, such activities can help students and citizens develop
awareness that their play activity is linked to network roles, and so help build metacognition
about how communities can be deliberately structured.
Investigating ‘local fit’
Qualifying as a Situated Game
To investigate fit, the prerequisite is an activity that qualifies as a Situated Game.
Although meeting these prerequisites seems straightforward for Reality Ends Here, the brief
articulation that follows will prove important grounding. Again, the three criteria are that the
design be: a game-based activity, have real-world actions, and have network-building goals.

(a) Network-building goals (and associational life)
The mandate from the Future Committee established the goals in network terms. First
the design would jumpstart “peer discovery,” and then the media making in groups would
“connect students to alumni, faculty, and the broader community.” Empowerment was more
implicit: for students in their careers, and for the community to stay unified. Nearly all network
goals centered on the repeated act of forming small groups. From a civic perspective,
encouraging small group formation has important alignment with democratic philosophies of
collective action, specifically the associational life described by Alexis de Tocqueville (1848).

112
Social translucence seeks to “support coherent behavior by making participants and their activities visible to one
another” with digital systems (Erickson & Kellogg, 2000, p. 59).
121

In de Tocqueville’s description, small groups and associations are vital for fostering
enlightened self-interest, establishing meritocratic norms, and building feelings of efficacy
around collective action. Research on extracurricular groups has shown that practice in small
groups with the skills of civic organizing – like mobilizing others, planning, speaking as a leader,
and building communal identity – produce norms that lead to later political engagement (Brady,
Verba, & Schlozman, 1995; McFarland & Thomas, 2006; Youniss & Yates, 1997).
Games have been proposed as sites for training groups in associational life (Schulzke,
2011), although the research focus has mostly been online. In Reality Ends Here there was
significant discursive reflection about the group, and the identity of teams in relation to the
broader community. For example, a number of prominent student videos discussed group norms,
good behavior, and their vision for a collaborative future.
The lack of explicit social issue content (e.g., on racism, or building community), may
not be vital for civic games in the same way it is crucial for a documentary film or a historical
biography. For games, the issues can be conveyed through experience and reflected on through
immediate conversation. Real-world games can support the direct formation of groups, while
also structuring reflection on the public good as members want to frame it.
Prioritizing experience over issue content has a precedent outside of games. In particular,
research into long-term political engagement has shown that being a member of extracurricular
groups is often more predictive of voting and volunteering than knowledge of civic facts (Hart,
Donnelly, Youniss, & Atkins, 2007). Online, social media activity of the kind used in Reality
Ends Here is increasingly recognized as providing alternative pathways to participatory politics
(Kahne, Middaugh, & Allen, 2014).
122

Decades hence, will the alumni of the film school be more likely to form groups to
address issues of collective concern – such as to raise money from Hollywood, or to mobilize
against censorship? Perhaps, in a small way. The decade-plus time scale for such research is
daunting, but must be considered for games that seek to deepen networks. Certainly the
intervention happens at a transformational moment, when students first enter the place-based
community, and begin to meet alumni.

(b) Actions in the real-world. In the words of the lead designer, the goal was to have “real
stakes, real players, and real outcomes.” Beyond building networks, Reality Ends Here led to
concrete media projects in the real-world. Videos created by participants were posted during
play to global distribution platforms like YouTube, and entered the swirling artist community
that exists online. Media studies have long approached film production as a way of affecting the
real world, alongside the other forms of media created in the game. Thus making progress in the
game is congruent with making a contribution or affecting the real world through broader
cultures of media making.

(c) Game-based. To qualify as game-based, a design structures participation in three ways:
1. Playful challenges throughout the activity: The creativity prompts in Reality Ends Here
were playful, and the challenge for groups was to produce their media project. The social
challenge began with the need to find collaborators, and keep them happy. The game
123

increased the challenge by rewarding complexity and involving more participants.
113
For
some players, the game became a challenge of one-upsmanship to make better media, but
importantly this challenge was issued and owned by the players themselves (Fullerton,
2014).
2. Feedback loops to guide progress: The formal feedback loops came via the point system
and leaderboard. Importantly, the agency of players was allowed by the system, which
supported multiple play styles (e.g., some students vowed in the game to prioritize quality
and avoid “pointsing,” while others clearly sought to maximize points). Peer feedback
came from posting finished projects and receiving scrutiny; videos posted to sites like
YouTube received feedback from a global community of media makers; the institution
provided feedback on the overall endeavor by providing support (albeit with a carefully
indirect tone).
3. Uncertain outcomes for participants: The designers worked hard to keep the game
uncertain, which required extra work as the game became known as a success and
administrators sought to tie the game to assignments. The designers resisted. Said one,
“rather than operating in the open, and creating game-like assignments, we hid the game
from students. We gave out random clues, to only some students” (Fullerton, 2014).
There was no directory of cards, no library or index – which spurred one group of players
to spend hours compiling their own list of cards.

113
As one of the designers describes, “As larger deals were put together, the creative challenge became harder and
harder to meet. It might seem that this would deter players, however, just the opposite. As the game progressed,
teams solidified, and bigger and bigger collaborations were formed, using more and more cards. The highest scoring
deal of the semester included 74 cards” (Fullerton, 2014).
124

Fit with mechanics
The prior chapter gave a detailed explanation of fit in terms of game mechanics.
114
This
chapter continues in that vein, emphasizing a cultural articulation that goes beyond how the term
“game mechanic” is used by professional designers. As in the prior case study, the designers of
Reality Ends Here tended to articulate the mechanics in fairly formalist terms, with one
emphasizing the deck of trading cards which “served as a procedural prompting system.”
Another designer described the mechanic as “constructing creative prompts.”
Going beyond such articulations of the formal system to emphasize social dynamics is
useful when optimizing for local fit. Of course, the designers also recognize that social
dynamics are cultural and core to the game. “The social is the medium,” declared one designer
(Watson, 2012).
Rephrasing the mechanics in social terms can be done as follows:

(a) “Greeting collaborators and pooling resources in order to make more complex
projects.” For players to make progress in the game, meeting collaborators is essential.
Greeting happens as participants show their cards to others, and as groups solidify to
determine exactly what creative project they will tackle. Meeting to pool resources (the
cards, but also personal equipment, technical skills of certain players, etc.) is necessary to
make collaborative media. This mechanic fits very well to the established modes, at least
for the production students who would already meet on their personal time to pool

114
Recall that the prior chapter on Macon Money investigated fit in terms of mechanics. Mechanics are what players
do in the game; the game can fit well – or poorly – into residents’ modes and activities of greeting and meeting.
125

resources for film festival submissions. (For students who do not submit to film festivals,
especially the Critical Studies students, the mechanic’s lack of fit can also help explain
the lower rates of adoption by that major.)

(b) “Greeting the community with a team creation.” After teams create multimedia
projects, they can only make progress in the game by posting them online. The posting
becomes an act of social greeting through the game’s requirement of creating a “making
of” video about each project. This mechanic aligns with some existing modes. In
particular, students were already posting their video creations to Facebook to share with
their local and distant communities; the game made this process more explicit and visible
to the local community.
By contrast, traditional educational environments rarely support reflection on
networking, let alone in public. Even for learning networking skills, reflection on
progress is vital – especially in professional contexts (Schön, 1983). In Reality Ends Here
the audience of peers for the justification raised the stakes, and helped make the choice of
cards more meaningful – including how the cards related to the broader community.

Are mechanics enough to analyze ‘local fit’? Not quite (otherwise there would be only
one dimension for ‘local fit’!). Perhaps the most significant limitation of mechanics is that they
optimize for behavior in a fairly narrow sense. Mechanics emphasize individuals, specifically
the choices of individual players as they progress in the game. While it helps to approach
126

mechanics from a perspective of interpersonal greeting and culture, mechanics are just one
dimension of local fit.
The two remaining dimensions of fit investigate the collective side of integrating with
local communities. The importance of such integration can be demonstrated in the design of
Reality Ends Here, with the help of the framework. More importantly, the framework helps look
beyond Reality Ends Here as a single game, in search of a more generalized language to support
localization and collaboration with community organizations.
In brief, the other two dimensions address fit to the collective identity, and fit to the local
ecology of communication. Each dimension embodies a set of theories for how communities are
collectively empowered. This chapter will address both dimensions in brief, laying the
groundwork for a deeper exploration and bottom-up optimization in the final case study.
Fit #2: to the local identity/brand
Reinforcing the collective identity is one way that games can fit to local communities.
Collective identity matters. For coherence, place-based communities depend on more than
sharing geography. A strong identity is part of what turns geography into places. Establishing
the community identity is an act of collective imagination (B. Anderson, 1983). Storytelling is
one way to describe the collective imagining of the community (Kim & Ball-Rokeach, 2006b).
Narratives and media make clear “who we are,” framing identity from collective
problems and visions. At the heart are stories that frame the group identity – including the tales
told by residents, media portraits of a place, and the branding of community institutions.
For Reality Ends Here, the designers tapped into a deep mythology. For decades, the
USC Cinema School has cultivated a mystique about its origins. The very phrase “Reality Ends
127

Here” is a primary motto of the school. In fact, a Latin translation is engraved in stone behind
the main courtyard (see Figure 25). Famous alumni
115
like Steven Spielberg and George Lucas
have led fundraising campaigns, and buildings bear their names; movie posters made by alumni
adorn the halls. Designers of the game were very aware of the school’s “status as a dreamland”
(Watson, 2012), and explicitly relied on the desire of students to become part of the “storied
history” of the school (Fullerton, 2014).

Figure 25: Famous alumni Steven Spielberg and George Lucas depicted with game cards (while
leaving prints in the cement of Hollywood); at right, images of the motto inscribed around the
school and in archive materials

Source: The Reality Manual (Watson & Reality Design
Team, 2013)

Source: Watson’s dissertation on the game (2012)

The spaces of the game re-embodied historic places. For example, the online discussion
area for the game was called “the Bullpen” after an infamous location in the original Cinema
buildings. As one designer described it:

115
Spielberg evidently did not officially graduate, but is often invoked as an alumnus of the school.
128

“The original Bullpen was a messy, communal editing space with walls covered in
graffiti from students. By naming the online community space after this historical space,
the game lauded a kind of irreverent, playful, disorderly approach ... “This space belong
to you,” the top of the online Bullpen announced…” (Fullerton, 2014).

By elevating historic places, the game sought to inherit and reinforce the community’s
sense of place. Naming places is a kind of branding that is practiced by real estate developers
and fundraisers, often without buy-in from the broader community. Yet the concept is also
integral to the place-making strategies for community redevelopment that have shown recent
popularity across the United States (Nicodemus, 2013), including branding with signs as part of
way-finding (Calori & Chermayeff, 2007).
The game goes beyond linguistic branding by seeking to deepen the sense of place
through embodied activity, including its online version of the historic Bullpen. The collective
imagination of place is thus part physical and part digital. Such blending may be increasingly
necessary in a digital era, where it can be almost impossible to understand a place by accessing
only its physical side (Gordon & de Souza e Silva, 2011); of course, such hybrid spaces raise
equity concerns, especially for residents with limited access to the digital side of their
neighborhood, as will be discussed in the final case study.
Theater is a particularly useful way to describe how games deepen a sense of place-based
identity. Games in public space enact a kind of visible performance of community that also
highlights what the space means, and what the participants are about. Echoing what Erving
Goffman (1990) calls the performance in everyday life, games can structure the theatrics of
129

public identity, and reach bystanders who might notice the commotion and inquire what is going
on. Such theater is place-based; in describing Reality Ends Here, Watson calls it the
“dramaturgy of the local” (Watson, 2012, p. 54).
The theater of games is unusual in terms of agency because it gives residents special
permission to try on different roles
116
, imagining their neighbors and public space in a different
way. Artists’ locative games have long explored these possibilities, drawing on Guy Debord,
Michel de Certeau, and Henri Lefebvre (for an excellent overview, see Flanagan, 2009, Chapter
6). In terms of agency, the designers of Reality Ends Here emphasized that the theatrics were
open and participatory:

“As the game unfolded, it became difficult to ignore. It became a spectacle. This
spectacle was not made for the students as much as it was made by them. It played out
online on Facebook, Twitter, YouTube and other platforms; and offline, in student
dormitories, campus spaces, and the streets of Los Angeles. It told the story of the
friendships, partnerships, rivalries, and dreams forged through the situations of the
game. It was more than just a spectacle. It was a participatory spectacle
117
.” (Watson,
2012)

116
The special space created by games is often described as a “magic circle” (Salen & Zimmerman, 2004) where the
rules of ordinary life are shifted, by the mutual agreement of participants. The origin comes from Huizinga (1938),
who describes how games become “temporary worlds within the ordinary world, dedicated to the performance of an
act apart.”
117
The idea of certain kinds of useful spectacle being participatory (and thus more ethical) has been advanced by
civic theorist Stephen Duncombe (2007), and will be investigated in further detail in the following chapter.
130

Can such spectacle and theatrics have legitimate ‘local fit’? Certainly. In fact, the lack
of fit can lead to the game’s rejection. Games can be perceived as telling “our story” or being
inappropriate. Thus one simple test is participation: students might refuse to participate if the
game feels inconsistent with who “we are.” Such identity checking may be premature in the
freshmen targeted by the game, but more established stewards of the community were also
present in the game and its storytelling. (Recall the Future Committee that requested the game
originally, asking to help students in “writing the next chapter” of the school’s story, and to
“facilitate their telling of that story.”) Support from the administration after the first year was
enthusiastic (albeit discrete, at the urging of the designers). The administration’s most important
vote of confidence was to renew the game. Just as importantly, alumni also gave support
consistently as they were pulled into the game.
Additional techniques are still needed to optimize games for local identity. Yet the most
important strategy is simply to focus on local stories, the kind that differentiate and affirm the
collective nature of the place-based community. Such stories can be called “neighborhood
stories”
118
to emphasize their place-based nature. Often this dissertation simply refers to them as
“community stories,” since ‘neighborhood’ is not the right scale for Reality Ends Here.
Regardless, the power of neighborhood stories only matters if they are heard, which
brings us to the final dimension of fit: the local ecology of communication.

118
“Neighborhood stories” is a term adapted from the Metamorphosis Research Project (Ball-Rokeach et al., 2001),
which studies empowerment at the neighborhood level. More details will follow.
131

Fit #3: to the local ecology of communication
Communication is often the primary means of forming network ties. It is easy to
communicate in principle, but in practice each community has its own communication ecology –
a set of established channels and relational practices that shape the flow of local stories. For
example, some communities rely heavily on organizations like churches, while others rely on
Chinese-language newspapers, and others use social media as a resource (Chen et al., 2013).
As individuals pursue information and relationships locally, they choose which
communication hubs to tap, including physical meeting spots, local media platforms, and even
neighborhood institutions. Over time, individuals shape their communication ecology (Ball-
Rokeach, Gonzalez, Son, & Kligler-Velenchik, 2012). How a communication ecology is formed
at the local level, and implications for collective empowerment, are documented in a decade of
research into the sociology of neighborhood communication at the Metamorphosis Project (Kim
& Ball-Rokeach, 2006b).
119

With the focus on communication where it happens, the ecology frame helps to see
physical spaces as opportunities for interpersonal communication. Design may depend on
understanding the local ecology of communication (Broad et al., 2013). Behind the
Metamorphosis approach, play goals are quietly recognized as a valuable way for individuals to
navigate media systems (Ball-Rokeach, 1985). For Reality Ends Here, the designers similarly
attempted to fit into the physical spaces that are so vital for place-based communication:

119
More on the “communication ecology” approach will be explored in the chapter on Ride South LA; the lens of the
Metamorphosis Project for approaching the local ecology is invaluable, although only some of their formal language
is used in this analysis. For now, it is worth noting the Metamorphosis definition of the communication ecology as
being “a network of communication resource relations constructed by individuals in pursuit of a goal and in context
of their communication environment” (Ball-Rokeach et al., 2012).
132

“The game was designed to capitalize on the fact that it was played in a community that
inhabited the same campus, the same buildings, the same classrooms, sometimes the
same dorms most of their days. The local nature of the community… called for face-to-
face core gameplay … that would provoke players to interact in person and form real-
world social ties” (Fullerton, 2014).

A philosophy of “environmental design” was proposed and used by the creators of
Reality Ends Here (Watson, 2012). Environmental design includes the communication
environment. In fact, communication can be one way of describing the broader environment, via
“spatial, object and emotional communication” (Watson, 2012, p. 2).
120

By contrast, most real-world games have only emphasized the much more shallow
constraints of communication technologies in physical space. Technology assessments often ask
residents about their phones and plans, attempting to understand latent constructs like technology
‘readiness’ (e.g., Parasuraman, 2000). While technology is part of the communication
environment, an exclusive focus risks missing the most important social modes of
communication. Into this trap fell the designers of Reality Ends Here, at first.
In the early days, a smartphone design was initially prototyped for Reality Ends Here.
The designers anticipated players with Smartphones in good supply, and so they tested a design
with “snazzy features” like detecting nearby players (Watson, 2012, p. 69). However, the
prototype was determined to be a failure. The reason was that it missed the fundamental
constraints of the place-based environment for communication. As one designer admitted,

120
The quote here is from the Chair of Environmental Game Design at the Art Center College of Design, as invoked
by Watson in describing Environmental Design as applied in the game.
133

“techno-fetishism had blinded us.” In the end, the paper trading cards were determined to be
superior for the desired modes of communication, despite sufficient access to a more “advanced”
technology of communication.
The emergence of new media actually redoubles the importance of ecological thinking
(Ball-Rokeach et al., 2012), as old and new media increasingly converge and interconnect (H.
Jenkins, 2006). Mobile media in particular is increasing the overlap in communication between
digital and physical worlds (Gordon & de Souza e Silva, 2011; Ito, 2005). Technology baselines
are still valuable, especially as a point of departure, but ecological techniques are increasingly
needed.
An ecological view gives a powerful way to investigate and articulate the local context
for communication.
121
Broadly speaking, ecological models approach neighborhoods as
relational systems, where the sum is greater than the parts. One benefit is that ecological
thinking helps to speak the language of neighborhoods and cities as articulated by many
prominent urban theorists (Jacobs, 1961; Sampson, 2012; Stokols, 1996).
122

From a communications perspective on the urban ecology, the central flow is comprised
of messages and human conversation, including local stories about the community. Analytically,

121
The “Communication Action Context” is akin to the cultural and technical infrastructure available to local actors,
including community-based organizations, local media and residents themselves; for a useful description of the
factors and concepts involved, see Ball-Rokeach et al. (2001).
122
While empowerment and agency is the focus in this study, a deterministic frame was rampant in much of the
early sociology of the 1920s on the ‘natural’ ecologies of neighborhoods (Sampson, 2012). Resistance grew
particularly in the 1970s, with Marxist critiques from scholars like Manual Castells (1979), as well as David Harvey
and Mark Gottdiener. More recent work on ecologies for urban empowerment has been careful to highlight
opportunities for individuals and communities to shape their own destiny (Jacobs, 1961; Sampson, 2012; Stokols,
1996).
134

ecological approaches bring insight into looking across levels, including interpersonal and
organizational, to understand how the community works.
Clarity can come from ecological models, and their relational approach. By comparison,
blunt laundry lists of communication needs and constraints can be overwhelming (this was the
case for Reality Ends Here
123
). Looking across levels can help move beyond the usual focus on
technology “users,” recognizing the importance of local channels that are larger than any one
person. Reality Ends Here depended deeply on the implicit alignment with the university,
integration with Facebook, and with physical space.
Concretely, ‘local fit’ to the ecology requires considering three levels, echoing the three
levels emphasized by communication infrastructure theory (Kim & Ball-Rokeach, 2006b):
(a) Channels between participants, where talk is direct (including physical and digital
spaces), emphasizing actual use patterns, not simply access;
(b) Channels to community organizations, where participants hear from local groups
that are shaping the collective conversation, such as local churches, libraries, health
centers, schools, sports leagues, and business clubs;
(c) Channels facilitated by local media, especially where there is some broadcast power
and legitimacy, as with local TV news, group pages on social media (e.g.,

123
The laundry list is extensive and exhausting to read. According to one design analysis, the constraints ultimately
came to include “the varied interests, tastes, habits, and competencies of the incoming students; the history,
curriculum, and institutional structure of the SCA; the school’s buildings and outdoor environments; the social
media spaces inhabited by the students and faculty; the range of housing situations among freshmen; the proximity
of alumni to the campus; the near ubiquity of smartphone devices and/or laptop computers among students; the
changing landscape of media arts production, and the new demands this places on media arts education; the city of
Los Angeles; the nostalgia associated with collectible card games such as Pokémon; a very tight budget; a desire to
experiment; and many other factors” (Watson, 2012, p. 75).
135

Facebook), and geo-ethnic media
124
. Local media occasionally bridges to national
broadcast media.

Techniques for game designers to describe the local communication ecology will be
increasingly useful for place-based game designers. For example, neighborhood ecologies can
be mapped using “asset mapping” techniques (Villanueva & Broad, 2012), as will be discussed
in the next chapter. Fundamentally, game design for local empowerment requires doing research
as part of game development – and in particular, research about the local organizations and how
communication flows across levels. Such research is time-intensive and sometimes expensive.
The consequences of failing to fit into a community’s modes of communication include
undermining participation, and perhaps even being seen as a threat to self-actualization. In other
words, the game might not work, and there could be backlash. A deeper investigation into the
communication ecology is explored in the following study of failed localization.
When scaling fails: a counter-example of Reality Ends Here
Is fit a condition of success for Reality Ends Here? Failure can be more insightful than
success, including with games (Gee, 2005; Juul, 2013). Particular insight for ‘local fit’ can come
from attempted localization.
In the summer of 2013, enthusiasm about Reality Ends Here inspired an attempt to bring
a version to a second school (“School Two”
125
). School Two was at the same university with

124
Ironically, local and geo-ethnic media can trump a city-wide outlet (Wilkin, Ball-Rokeach, Matsaganis, &
Cheong, 2007).
136

hundreds of its own students. Its area of study also concerned media and culture, but focused on
journalism and communication. In pursuing the game, School Two’s goals included stimulating
multimedia production, fostering interdisciplinary collaboration and networking.
The newly modified game was called Xposure. The primary localization of Xposure was
in brand and content. More than 300 original cards were created (Figure 26), designed to match
“the interests, knowledge and skills of [School Two] students from all three areas of the school:
communication, journalism, and public relations.”
126
As with the original, the backside of the
cards featured famous people from the field, local faculty, and key concepts in the school’s area
of focus. For example, one included a profile on Alexander Graham Bell for his role in the
history of communication technologies.

Figure 26: Cards for Xposure, based on Reality Ends Here

Source: the Xposure design team

125
The name “School Two” is used mainly for simplicity, as shorthand for the School for Communication and
Journalism. The shorter name also provides a little analytic distance, since the author of this dissertation attended
the School while this research was being conducted.
126
According to internal design and review documents; a more detailed description of the data used in analyzing this
second game is provided below and in the appendix on data sources for Reality Ends Here and Xposure.
137

Cultivating mystery was again a focus for Xposure. Before arriving for their first day on
campus, incoming students received a mysterious postcard (Figure 27) that challenged them to
“eXpose the secret.” The summer timing seemed great for the postcards, with student interviews
reporting that the postcards made them look forward to the start of the year. Later on campus, a
mysterious video started appearing on prominent monitors in the main lobby.

Figure 27: A mysterious postcard for Xposure (left), and a video (right)
that appeared mysteriously on screens in the school’s lobby

Source: the Xposure design team

Another student received a fortune cookie containing messages from Xposure staff that
pulled him into the game. He described the process in an interview:
“I got a fortune cookie on the first day of school…at [School Two] orientation. It said
come to so-and-so place at such-and such time, and you’ll get a message. I did. It gave
me another message. That gave me another message. …Before you know it, I found the
Office of Exposure and I got my first stack of cards.”

138

One “student” claimed to have followed the postcard clue to discover a mysterious flag in
the school lobby. The curiosity of participants is captured in Facebook discussion of the flag
(Figure 28); in this case, 84 other people had seen her post, hinting at the desired interplay
between physical space and digital echoes that spread online. However, the student was not real
– her Facebook account and persona were created by the Xposure team to foster a sense of
mystery and experiment with word of mouth outreach. Awareness of the parallel game at the
original School of Cinematic Arts emerges, with one student asked online, “is this a game like
the [Cinema School] kids are doing?” “Si,” another replied.

139

Figure 28: Social media buzz for Xposure

Source: the Xposure design team

140

But after three weeks, Xposure stalled. Signups had already been modest, despite a well-
received early prize for the first winners that made clear the rewards were real and significant.
At first, organizers were advised not to worry.
127
After all, the original School of Cinematic Arts
had similar rates of signups in their first year, and then the game took off later. But not at School
Two.
Only three projects had been submitted. Organizers tried alternate energizing strategies,
including piggybacking a game challenge on an announcement that the school would provide
free Adobe software for students. But nothing seemed to work. Said one faculty member, “the
rabbit holes weren’t really doing it.” Eventually, the game was put on hold for the year.
Yet across campus at the film school, the game continued as before. If anything,
engagement was higher in their third year than ever, even as it evolved. Facilitators observed
that no large super-groups had formed, and that as a result there seemed to be a “more even
playing field” with fluid teams and more than a hundred projects arriving throughout the
semester. What was the difference?
A manual… for replication?
The Xposure design was based on advice from the original designers, and a manual the
original designers had published just a few months prior. The manual was designed to help
replicate Reality Ends Here, with funding from Microsoft Research (Watson & Reality Design

127
A manual created for the original game (Watson & Reality Design Team, 2013, p. 46) explicitly instructed “don’t
worry” if signups are slow at first.
141

Team, 2013). Reading the manual reveals many fascinating elements of the original game, from
its rules to custom software code, and to tips for professionally printing the cards.
Yet the manual omitted advice on localizing the game. Even as the authors describe how
their framework is potentially flexible, the manual describes how to run their game with little
guidance on how to adapt it. Much of this is understandable, and useful. In the manual,
troubleshooting is emphasized in terms of keeping things going smoothly in a context that is
assumed to be similar.
The manual places much of the burden for execution on the primary facilitator called the
game “Director,” who sets the tone and serves as the “alpha player” in the “improvised dance” of
the game. The Director is warned they will inevitably run into “unique circumstances that
require some degree of invention” (p.69). But such advice targets facilitation mid-game, well
after most localization decisions are locked in.
Certainly the manual provides a vital first step to understanding how the game works in
its original context. However, the manual’s implicit model for scaling is based more on
replication than adaptation. Fidelity criteria
128
are important, and such recipe books and
“toolkits” are often necessary (though perhaps not sufficient) tools for scaling.
Of course, the designers abstractly recognized the need to localize. In interviews and
their own published writing, the original designers insisted that the game depended on fitting to
the local environment of their players. But as published, the manual largely emphasizes the
Director as the focal point for localization; early meetings with the original designers also

128
For a discussion of “fidelity criteria” in scaling, it may make sense to focus less on technology scaling and more
on the scaling of social programs, e.g., see Mowbray et al., (2003) for examples from the health and education
literature.
142

emphasized the role of the Director as paramount, in terms of skillset and personality. With such
a focus on the Director, the search for explanations for the failure of Xposure is highly focused
on the Director and their ability to lead the dance on the fly. While the possibility of an
imperfect Director cannot be entirely eliminated, an analysis will show that the root causes may
actually be in the game’s (insufficient) local fit.
Before other schools can confidently rely on the manual, additional advice should be
provided on how to evaluate context. At a minimum, clarity is needed on the conditions for safe
replication, including how to determine whether a target school is sufficiently similar. More
ambitiously, guidance should be given on how to localize the game to other environments like
School Two. The framework for ‘local fit’ aims to help provide such advice.
Investigating localization via fit
The benefit of the ‘local fit’ framework is that it transforms the evaluation of context into
three distinct questions. Each question emerges from a different dimension of fit, which can be
evaluated separately (i.e., mechanics, identity, and the ecology of communication). Without
care, it is easy to overlook one of these dimensions. The risk is that failure in any dimension can
make localization unstable. In a sense, fit is like a three-legged chair, where each leg is needed
to stabilize the game on local terrain.
143

To demonstrate, consider the failure at School Two. First, let us assume the execution of
Xposure was reasonably well-done
129
, and that the original design is somewhat robust
130
. Then
each dimension of ‘local fit’ provides a separate window into the case of School Two as follows:

(a) Fit #1: Mechanics. For this case, the mechanics
131
were not substantially changed. So if
this dimension of fit is to blame, then School Two has “different local modes of
greeting and meeting.” Does it? At first glance, the two schools appear similar. Both
draw on undergraduates from the same university. However, there was a substantial
difference in how students spent their leisure time to meet and greet. Specifically, there
was a deep pattern at the first school of student groups submitting to film and media
festivals using work they had created on their free time. This DIY ethic was echoed in
the game.
132

But for School Two, interviews with students invoked community-level disdain for

129
Of course, poor execution can ruin localization. However, interviews and direct observation showed only
positive signs that the game Director for Xposure did a good job. Some of the assistants at School Two –
undergraduates called “game runners” – were poorly selected at first (one borrowed from the film school was
inflexible and insistent on replicating the original game), but this was later rectified. In general, there is no strong
evidence that staffing was a substantial problem. In general, to investigate fit this analysis assumes that Xposure
was reasonably well executed.
130
It is possibly that the Reality Ends Here game is not robust, which is to say that it only works under guidance by
the most skilled of facilitators. Yet the designers strove to create a robust game that would have the potential to
work elsewhere. While this can only be proven through successful scaling, this analysis hopes to improve those
odds. Therefore, this analysis is conducted assuming that the game is robust enough to scale via localization.
131
Recall that this chapter has articulated the two core mechanics for individuals to make progress in the game as: (1)
“To greet collaborators and pool resources to make more complex things;” and (2) “To greet the community by
sharing our group creations.”
132
The DIY (short for Do-It-Yourself) mentality was explicitly invoked in the first school. Specifically, according
to one of the designers, to receive their first cards players were required to swear an “oath of DIY media” in which
they promised to play safely and yet be fearless in their creativity” (Fullerton, 2014).
144

spending weekends on multimedia. Said one player of Xposure, “we just don’t waste our
free Saturdays on making little videos… whereas the Cinema students would do that –
it’s their way.” More broadly, building media portfolios was uncommon at School Two,
and only aligned with a few of the majors like journalism, where there is a tradition of
producing media for portfolios (closer to graduation, not in the first year). Thus
replicating the first mechanic of the original game so exactly might not have made sense.
As will be described below, School Two players actually do make media on the
weekends, but they do so at social events like Football games, and their products are on
social media not videos for film festivals. Strategies for how to adapt the mechanics
emerge in part by examining the ecology of communication for School Two students
below.

(b) Fit #2: Group Identity. If this dimension of fit is to blame, then School Two has a
“different local identity or community brand.” Does it? Curiously, the identity at
School Two was harder to identify – there was less of a historical narrative, and so less of
a sense of place. There was no mysterious motto of “Reality Ends Here” in the cement
and above the archways. Stories of famous alumni were not invoked as community
mythology to differentiate the school within the university. School Two did have a
strong brand name, but there seemed to be a lack of clarity on what the name actually
145

meant.
133
(Ideally the lack of a clear identity would not be a deal-breaker, since the goal
of the game is partly to build a sense of place; but the lack of a clear identity forces the
design to tackle the harder problem of building the identity rather than matching it.)

Where was identity invoked? Content on the cards was where the design primarily
sought a distinct sense of community and place. Certainly, content does matter.
However, the theatrics of the game are just as important for fitting to group identity. The
theatrics of Xposure were not substantially changed: players still performed the
clandestine creativity of wild film students, with mysterious banners and official denials
of the game. Was this necessary? Certainly for the original film students, it was
important that the game not “smell of school.”
134

Yet for School Two, the game might have needed to smell more like official school –
both professionally and socially. On the professional side, School Two had a stronger
embrace of pre-professional activities and internships; on the social side, School Two
students talked about embracing the umbrella social structures of the school – including
the Greek systems of sororities and fraternities, and closely followed sporting events like
Football. While Reality Ends Here sought secrecy almost in opposition to school-
sanctioned socializing and pre-professional identities, School Two may have needed a

133
The lack of clarity seemed to extend even to the academic majors. One student claimed the perception problem
for incoming freshmen could extend to the degree itself; he observed, “it’s a generic degree – nobody goes into the
program really knowing what the heck [this major] is.” (For privacy reasons, the specific major is not detailed here.)
134
Fears were expressed by the original designers that “the connections we seek are fragile and may be broken if we
do not respect the aspects of games that also make them unlike traditional learning and education” (Fullerton, 2014).
146

different approach. For example, giving extra class credit for participating in Xposure
would have linked the game to pre-professional trajectories (akin to internships). As
several players of Xposure emphasized, “more out in the open it would be better –
especially as you’re establishing the game within [School Two].” On the social side, one
of the prizes had to be postponed at School Two because players would rather attend and
align with the football game than go on secret adventures. Optimizing for collective
identity points to a possible future strategy for School Two: shift the game so that
participants can more openly perform their school’s identity, including as a space for pre-
professionals, and as aligned with the social identities of the overall school (including at
heavily-branded sporting events and in the widely-advertised activities of Greek life).

(c) Fit #3: Communication Ecology. If this dimension of fit is to blame, then School Two
has “different local modes of communication.” Does it? Both groups had access to
similar technologies of communication, and attended the same umbrella university. Yet
the ecologies were quite different. Significant differences emerged, echoing the levels
emphasized by communication infrastructure theory (Kim & Ball-Rokeach, 2006b); two
in particular were echoed in interviews:

Communication between players (in residence): The original game relied on shared
housing for communication; as one designer said, the special housing floor for film
students was “a big driver of play – they are literally right next to each other; that is why
we made the [physical] cards in the first place, because they could bring participants into
147

each other’s rooms.” By contrast, School Two lacked shared housing. Without being
directly asked, shared housing was also brought up by Xposure players as something they
needed for cohesion.
135
As one faculty member at School Two described the feeling,
“Even our coffee cart culture isn’t the same… they have a courtyard… places to hang out
that feel more natural.”

Communication tied to local institutions: All freshmen in the original game shared a
class their first semester. Even though the required class made no official mention of the
game, it had the exact same name: “Reality Ends Here.” More broadly, the institution
played a key role in telling and sharing the collective narrative. By contrast, there is no
unifying course in School Two, let alone a course with the school’s motto.
136
One
Xposure player felt making course ties was especially important at the beginning of the
game: “since we don't have a single class, [the game] needs to be started in a different
way.” (Some majors at School Two have a required class, but implicitly the focus is on
that major and not the collective narrative of School Two. A few weeks into the game,
an effort was made within several School Two courses to recruit, but despite the hopes of
facilitators there were no new sign-ups.)

135
Bemoaned the student, “I think it would be really, really neat if [School Two] was able to create a sort
of…fraternity. Because I think then …there would be a little bit more sense of camaraderie and of brotherhood …
between all the students at [School Two]. Like I just said, Cinema has one and the Business School has one and the
Engineering School has one. Why can’t we have one?”
136
There is a growing move in the direction of a school-wide course at School Two; however for this year only
about 1/3 of the students enrolled. The 100-level course is taught by the Dean and a former Dean and the syllabus
describes how the course seeks to introduce students to the city, the field, and the school.
148

Comparing incoming students (and their practices)
Players have particular influence in shaping the first level of the ecology: communication
channels between participants. Before the game began, players at each school had their
differences, including preferences about communication. Seeking patterns in individual
preferences can be a valuable technique to anticipate the communication ecology, and compare
schools. The prior analysis approached the ecology in largely structural terms (like required
classes and shared living space). Yet individuals and their preferences matter, especially if there
are consistent trends across a community.
To investigate, a survey was distributed to students of both School Two and the original
School of Cinematic Arts. To focus on the students before they joined together, the survey took
place before the semester began in August of 2013. The survey attempted to go beyond
technology alone. (Quite often, such surveys are approached as “needs assessment” with an
emphasis on technology skills and access; such questions are useful but insufficient.) Here the
survey goal was to understand communication practices and preferences more broadly,
emphasizing daily use and the availability to network with new social circles.
For outreach, administrators from each school facilitated the survey distribution via
email. No mention of the game was made in the outreach or the survey, which simply said that
the responses would be used to improve educational offerings. Thus the results point to the
broader community at each school, not the subset that played the game. A decent response was
obtained for each school (n > 110 responses for each school), about 45% of each incoming class.
Predictably, the film-oriented School of Cinematic Arts had significantly more
experience with video editing (89% of students reported having edited a video three or more
149

times, compared to just 54% for School Two). Similarly, the School of Cinematic Arts led in
DIY activities like remix
137
(59% had done this three or more times, compared to 27% from
School Two), and were much more likely to claim “I taught myself” digital media skills (75%
chose this option, compared to 47% for School Two, where students were more likely to report
they learned digital skills in classes). These differences were statistically significant.
138
Students
at the first school had greater confidence in their ability to make media
139
; and looking to their
careers, more from the first school were also more likely to feel that learning more digital media
was “essential.”
140

Curiously, the opposite pattern was true for social media. Specifically, School Two
students had higher rates for daily Instagram and Twitter use (setting aside Facebook – which
was nearly universal). Most starkly, Instagram was in daily use by 74% of students at School
Two (compared to 35% at the original school); somewhat less starkly, Twitter was in daily use

137
The remix survey item asked how often students had “Remixed material you found online (like songs, text or
images) and transformed it into your own artistic creation.”
138
The schools had significantly different means (at the p<.001 level; df=238) using a t-test of independent samples;
the means are actually based on a scale with four levels (i.e., for each activity, participants could say that had done
the activity “Never,” “Once or Twice,” “3-10 times” or “10 or more times”). For the question on where they learned,
the options included “in classes and school,” “from friends and family,” “I taught myself,” and “I don’t really have
any skills.”
139
Confidence in “your ability to make media – like videos, editing pictures…” was significantly higher for students
in the School of Cinematic Arts (28% were extremely confident, compared to 11% at School Two); these
differences were statistically significant.
140
Approximately 15% more students at the School of Cinematic Arts felt that learning more digital skills was
“essential” to their future career; other options for this question included that learning more digital skills would help
their career “a lot,” or “a little,” or “I probably do not need to learn more digital skills for my career.” For this item,
the difference between the means of the two schools is statistically significant (at the p<.001 level; df=238, using a
t-test of independent samples).
150

by nearly half the School Two students, compared to only a third at the School for Cinematic
Arts. The differences in Instagram and Twitter use were statistically significant.
141

Figure 29. Comparing communities: media production experience (left),
and daily social media use (right); While the first school has more production experience, they lag
in daily social media activity.

Gender appears an important way to explain the media use differences. School Two
skewed strongly female (75% female), while data from several years prior showed the original
school with the opposite skew (only 41% female); these statistics show bias beyond a fairly even
split in the broader undergraduate population of the university.
142
In at least one study on media
production, women were less likely to create video content for social networking sites (Hargittai
& Walejko, 2008), with just 17% of women active compared to 27% of men. Conversely,

141
The schools had significantly different means (at the p<.001 level for Instagram and p<.01 level for Twitter;
df=238) using a t-test of independent samples; the means are actually based on a scale with four levels (i.e., for each
activity, participants could say that their frequency with the service was “Never,” “Once a Month (approximately),”
“Once a Week (approximately),” or “Daily (approximately)”).
142
Public gender data from the School of Cinematic Arts website indicated that the class of 2010-2011
undergraduates was 41% female. Data from subsequent years has not been made available by School of Cinematic
Arts. The university as a whole is fairly balanced, with 51% women undergraduates enrolled in the fall of 2012,
according to the National Center for Education Statistics, https://nces.ed.gov/globallocator/ .
151

Instagram appears to be used more by women.
143
Such gender trends likely represent structural
inequity as well as some amount of individual preference. Are there other biases?
One possibility is that students at School Two were simply less socially outgoing. (In
this case, the use of social media like Instagram might be a crutch, hiding a hesitancy to meet
new people face-to-face.) Yet the data does not lend credence to this explanation. Students at
both schools had remarkably similar confidence in their socializing abilities (including the ability
to carry on conversations with others, to make friends of the opposite sex, to work well in
groups, and to form teams)
144
. Additionally, social attitudes toward being outgoing seemed very
similar (the survey focused on introversion – like being inclined to keep in the background on
social occasions; and whether students prefer to maintain a few close friends rather than make
new ones)
145
.
Are students equally available to form group ties? One concern was that the game might
be better suited to communities with new residents who are more open to making local friends.
Yet at both schools, approximately two-thirds of incoming students had very few or no prior
friends in the metro area. A similar concern was that one set of students might be less interested
in the alumni network, or in community service. Yet both schools had similar student interest in
spending time “to support my community.” If anything, School Two may have had more interest

143
For example, a 2013 survey from the Pew Research Center’s Internet and American Life Project showed that
Instagram use was greater by women (16% compared to 10% for men), and that the differences were statistically
significant. Data focused on college students was not available.
144
These questions are based on Bandura’s “social self-efficacy” items in his scale for children’s self-efficacy
(2006).
145
These introversion questions are based on a measure for introversion developed by McCroskey based on
Eysenck’s 1970 Readings in extraversion-introversion: Volume I. See
http://www.jamescmccroskey.com/measures/introversion.htm
152

in the alumni network: more of their students said the alumni network was an important part of
their decision to attend the university.
146

Limited controls for students’ backgrounds were considered, mainly in terms of
educational attainment. The digital divide for creating content like video is associated with
parental education (Hargittai & Walejko, 2008). In terms of parents, the schools did not have
significantly different levels of parental achievement. In terms of students, the differences were
more ambiguous; for student school background, there was no evidence to support the expected
higher rates of private school in players of the original game (in fact, public high school was
more common, as claimed by 64% versus 56% for School Two).
147

In sum, the picture that emerges through the survey includes some telltale differences.
Students at the School of Cinematic Arts had more experience in digital media production, a
background of teaching themselves digital skills, and a belief that such skills were necessary.
Yet School Two students had their own orientation, and it is not accurate to describe them as less
engaged. In particular, their social media use with Instagram and Twitter is significantly higher.
Gender may explain much of the media use differences. At the same time, both incoming groups
appeared similarly outgoing and confident in their ability to form teams.
Overall, these findings help explain the differences between schools in terms of peer-to-
peer communication, an important part of the communication ecology. Additionally, the

146
At the original School of Cinematic Arts there were 14% fewer students who said the alumni network was “very
important” to their decision to attend the University (63% versus 77% for School Two). This difference was only
somewhat significant statistically (p<.03; df=238 for the independent samples t-test on a five-item Likert scale).
147
Keep in mind these equity results are quite limited, based only on the few questions we could justify asking in the
survey’s available space; they are by no means a deep exploration of the differences in privilege between the
communities, though hopefully they provide at least a basic check on some of the education privileges associated
with the digital divide (Hargittai & Walejko, 2008).
153

findings help affirm why the mechanic of meeting for video production was a better fit at the
School of Cinematic Arts. Future research can go further by providing a survey template to help
designers determine the communication ecology in terms that are most useful for optimizing
games for local fit.
Summary of local fit opportunities
The investigation of Reality Ends Here can be summarized rather concisely in terms of
‘local fit’:

(a) Mechanics: Different mechanics may be needed to embed with local modes of greeting
and meeting. School Two students did not have the experience or DIY disposition to
have weekend filmmaking sessions for submitting to festivals; instead socializing is
likely to happen in more university-wide contexts, including sporting events, and around
social media like Instagram.
(b) Group Identity: School Two may have echoed the wrong group identity. They
mimicked the underground brand of School of Cinematic Arts. The alumni network was
valued by students at School Two, but their school is not widely identified with an
underground creative ethos; rather, their brand is more connected to public visibility,
including prominent pre-professional activities, and in the public-facing fields of
journalism and media strategy. Social visibility is also more aligned with the school-
wide identities of Greek life and branded athletic events. It may be necessary to find a
way for the game’s actions to perform a distinct group identity, including one with more
institutional sanction.
154

(c) Communication Ecology: School Two lacked the shared housing that was a vital
communication channel for the original game, as well as universal participation in a
unifying course. To compensate, School Two may have needed to make the game more
visible in school spaces, and be heard in institutional channels of communication.
Individual communication practices at School Two were also strongly aligned with social
media like Instagram, and peer-to-peer media strategies may need to be a central part of
the game, including at school social events like football games and Greek parties.
Seeking fit bottom-up
As the analysis of School Two has shown, localization is non-trivial for place-based
games. Survey techniques are an easy first step, but as this one showed, there are definite limits
to probing the communication ecology with such narrow quantitative approaches. The
framework of ‘local fit’ did bring some focus to the survey, yet it did not yield definitive
solutions for what needs to be fixed.
By implication, a limitation of using surveys to investigate ‘local fit’ is that often the
framework cannot identify which aspect of fit is most to blame for failure. Even network
analysis is not useful if the game fails to gain initial momentum, and the first weeks can be
critical. Of course, there is no “magic bullet” for local fit. If anything, the analysis of Xposure
shows that fit is a puzzle with complex edges, even with a single game.
Researchers in laboratory traditions may be tempted to seek a more controlled
experiment. With sufficient resources (and enough film schools), it should be possible to
identify fairly exact indicators of where Reality Ends Here would work. But that has a cost, and
may not be the most productive focus.
155

When a game is made predictable, it loses some of the ability to fit deeply into local
communities. Just for Reality Ends Here to fit with School Two may require modifications at
the level of mechanics for social meeting, becoming in some ways a different game. Forcing the
same game, even when possible, may flatten the differences between communities, promoting
homogeneity rather than the local distinction celebrated in movements like place-making.
Are there more bottom-up approaches to ‘local fit’? The paradox of games with ‘local
fit’ is that they seek to retain some of the game, while still retaining the flexibility to adapt.
Good games are not made by committee – they require careful balancing and aesthetics that
necessitate countless executive decisions. Can the design process itself be more deeply
embedded in conversations with the community about its own priorities?
The next chapter takes a grassroots approach, testing the limits of the framework for
‘local fit’. Play is central, even though the participatory mapping of RideSouthLA is not formally
a game. Yet the activity can be improved by borrowing theory from game studies. At the same
time, game designers can also benefit from participatory techniques to situate design more
deeply within the local ecology for communication.

156

CHAPTER FOUR: PARTOUR IN SOUTH LA,
COALITION STORYTELLING
& PARTICIPATION GAPS

“Just as none of us is beyond geography, none of us is completely free from the struggle over
geography. That struggle is complex and interesting because it is not only about soldiers and cannons
but also about ideas, about forms, about images and imaginings.”
- Edward Said (2012, p. 7), Culture and Imperialism

Overview
Game design is increasingly visible in civic domains, but remains a privileged practice.
Behind the scenes of high-end games like Macon Money and Reality Ends Here are some of the
most experienced designers of independent games. Yet in historically marginalized
neighborhoods, formal games can be premature. Even when budgets allow, power relations are
delicate and flexibility is vital. Empowerment may depend in particular on activities that are not
strictly games, and fit that comes from the bottom-up.
The use of digital media for empowerment raises new questions of equity. The notion of
equity in a digital age is broadening and benefits from an activity perspective. At first, with the
rise of the Internet, warnings addressed the “digital divide” in technology access across lines of
socio-economic privilege, including between rich and poor countries (Norris, 2001). Today,
157

despite ongoing concerns over technology access, it is also increasingly important to analyze
social practices. Access is a necessary but not sufficient condition.
148

How can equity be investigated around social practices? One useful framing is to
consider equity in terms of participation and the “participation gap” (Henry Jenkins, Clinton,
Purushotma, Robison, & Weigel, 2007). By emphasizing participation, the access question is
more broadly framed as access to opportunity – technical and social. Social supports (like
mentoring and access to helpful peers) matter alongside the traditional institutions of technical
access.
149
For example, underprivileged neighborhoods suffer from a lack of “civic
opportunities,” including in publicly-funded institutions like K12 schools (Kahne & Middaugh,
2008). As the digital and physical come into closer contact, activities can be designed to
increase social opportunity – including by providing opportunities for group formation from the
bottom-up and for collective storytelling.
This chapter investigates ParTour, an activity centered on participatory mapping and
transmedia storytelling in South Los Angeles. The mapping of ParTour is a group activity, with
roots in participatory design and coalition building. As the analysis will show, activities like
ParTour can be optimized with game strategies to improve and situate engagement, such as by

148
For example, even for individuals with similar access to technology, those with higher income have greater
political activity online. More specifically, when considering five prominent online political activities, individuals
with high socio-economic status (SES) had a rate of 65% participating in at least one, compared to just 10% for the
bottom quintile of SES (Schlozman, Verba, & Brady, 2010).
149
For example, the largest national ethnography of the digital media practices of youth (Ito et al., 2009) found that
online practices diverged for the same platforms. Platform access was used in some cases to mirror offline
friendship circles for “hanging out,” while in other cases participation was driven by passions and interests that led
to different social networks for “geeking out.” Not surprisingly, the supports for geeking out were much better in
wealthier households.
158

providing media feedback loops during physical events and to calibrate challenges within social
activity.
As street games can increasingly overlap with networking activities, the framework of
‘local fit’ can encourage equity tactics that are bottom-up. For example, game designers using
the framework will more consciously import practices from community organizing, including
approaches to networking that rely on local storytelling. Integration is not easy; this chapter will
describe a civil rights game in Atlanta that used ParTour, but just as a storytelling tool and
technology. As will be shown, tools only move toward ‘local fit’ in small steps; larger gains
require adopting the broader goals of local empowerment, as is required of Situated Games.
Confronting negative stories of place
The ParTour project began in South LA, a place burdened by an inaccurate and fear-
based image. The negative reputation is partly a historical function of the 1965 Watts Riots and
the 1992 Civil Unrest. Research by Matei et al. (2001) showed that the Watts neighborhood in
South LA persists as an epicenter of perceived fear in the city; yet they also discovered that this
fear did not correlate with actual crime statistics. In other words, feeling misunderstood is
justifiable. More recent policy reports like the UCLA School of Public Affairs' State of South LA
(2008) conclude that the area still suffers from long-term disinvestment and systematic
discrimination.
Yet there is also good news in Los Angeles. The city is home to some of the most
politically prominent work to connect spatial justice with social movements and neighborhoods.
159

In particular, research by scholars like Edward Soja and Manuel Pastor
150
has shown Los
Angeles to be a leading case of “community-based regionalism.” Using examples from bus
unions to Hollywood, they document the rise of spatial thinking
151
, especially showing how
diverse social campaigns have united for economic justice under a regional banner.
Despite the persistence of poverty and disadvantage in South LA (an area southwest of
downtown Los Angeles), the neighborhood has also proved to be a space of creativity and
collective resistance (Hunt & Ramon, 2010). Recent surveys have revealed an impressive range
of organizations working to improve the well-being of residents through community organizing,
advocacy and institution building (Broad et al., 2013).
Emerging media tactics require a delicate touch. In particular, storytelling on the
problems of a marginalized neighborhood can further damage its brand, in what Villanueva and
Broad call a “negative tradition of storytelling” (2012, p. 1). Strengthening the neighborhood’s
own capacity for positive storytelling is an important strategy for empowerment (Chen, Dong,
Ball-Rokeach, Parks, & Huang, 2012). But if funds are scarce, where should an intervention
begin, and how can it track success? Prior work on geo-ethnic media (Kim, Jung, & Ball-
Rokeach, 2006) has emphasized fitting to geography and ethnicity when designing interventions;
this study applies the premise of geo-ethnic media in the context of games. The following case
study of ParTour is based on a multi-year research program of collaborative design, participant
observation, and stakeholder interviews (see Appendix for a complete data and timeline
summary).

150
In particular, see Pastor et al. (2009) and Soja (2010).
151
The approach builds on Lefebvre’s “right to the city” philosophy, echoing Rawls notion of justice and Foucault’s
recognition that space shapes power and information.
160

A qualitative approach is embraced in this chapter, emphasizing the process and
structures of participation, and some of the resulting discourse across networks. Media-based
organizing and networking is profoundly difficult to track, especially across geographic areas as
vast as South Los Angeles.
152
Even when the community identity shifts or strengthens, isolating
the change for a single causal factor is difficult. Perhaps most importantly, change is likely to
result from a convergence of projects, organizations and events that are deeply interwoven.
Qualitative studies like this one can give a glimpse of promising directions. In the future,
comparison with sister interventions in storytelling networks
153
will be increasingly useful, once
frameworks like ‘local fit’ can help to distinguish the most important features of real-world
games and activities.
Mapping with a storytelling parade
Picture a sunny Sunday in South LA, early in January of 2012. A group of 60 bicyclists
pauses in an urban park on their ride, near a few picnic tables. Several push down kickstands
and crowd around a tablet iPad mounted to the handlebars of a bike. The tablet belongs to one of
the ride organizers, and he scrolls through a live feed of pictures. A middle-aged African-

152
The difficulty comes from the raw scope of the significant geography (more than 50 square miles that includes 25
neighborhoods of LA and three unincorporated districts), the timeframe necessary to track changes (since TV stigma
of the area has been shown to last decades (Matei, Ball-Rokeach, & Qiu, 2001)), to the cost to evaluate the
implications across such grand constructs as community power, social cohesion and economic revitalization. As a
point of comparison, Sampson’s book (2012) shows the stunning complexity of tracking neighborhood effects over
multiple years.
153
For example, a multi-year project to build a community storytelling platform called the Alhambra Source is being
tracked quantitatively by the Metamorphosis Project. Early findings and research frames are emerging (e.g., Chen
et al., 2013); such projects increasingly deserve the attention of community-oriented game designers as well.
161

American man points, “there’s mine!” as images from the neighborhood scroll by, each
submitted by a bicyclist over the past hour.

Figure 30: “I walked this street every morning to attend Markham middle school”
(caption as submitted to ParTour, en route to the Watts Towers)

Source: User-13331 for RideSouthLA

The pleasure of the ride was evident in the smiles on the faces of most riders in
attendance, and the conversational buzz between them. The pace was family friendly – a slow
half tour, half parade. When the 60-person crowd arrived at a particular park or intersection,
heads turned. Residents on benches looked up from their weekend conversations, some shouting
out, “What are you all doing? Where are you going?” Then one of the bicyclists would yell
back, “We’re riding to the Watts Towers! Come join us!”
The Watts Towers are a National Historic Monument and source of pride in South LA,
and have been ever since their completion in the 1950s. As one of the nation’s best-known folk
162

art sculptures
154
, the Towers are visually stunning and all the more impressive for the 30-plus
years it took the hands of just one Italian immigrant, Simon Rodia, to craft them. Over the years,
countless visitors made a pilgrimage to the Towers. The vast majority came by bus and car.

Figure 31: Riders pose for a group picture in front of the Watts Towers; many wave

Source: Sahra Sulaiman/StreetsblogLA (reformatted
155
)

The idea of biking to the towers was unusual, even according to many in the
neighborhood. Two stereotypes are directly challenged by emphasizing biking to the Towers.
The first stereotype is the car culture famously associated with Los Angeles, which the event

154
The Office of Historic Preservation of the California State Parks declared the Towers a monument in 1990; see
http://ohp.parks.ca.gov/ListedResources/Detail.aspx?num=993
155
For the original image, see the article by Sahra Sulaiman on “Measuring Success One Rider at a Time,”
http://la.streetsblog.org/2012/06/18/unpacking-the-numbers-measuring-success-one-rider-at-a-time/
163

organizers sought to counterpose with a recently-rising bicycle movement, evoking healthy
family exercise.
Second, the organizers wanted to counter a stereotype of the neighborhood as a
dangerous place, publicizing instead a group ride where South LA is a weekend destination for
families interested in local culture. The ParTour ride began in an urban wetlands park, stopped
by local businesses and even visited a historic trolley line to talk about public transit before
arriving at the iconic Towers.
Dozens of photographs were taken at each stop, each from a different perspective. The
technology behind ParTour automatically assembled a group story online. Specifically, the
photo mapping of ParTour allowed riders to publish a live stream of captions and pictures,
which also populated an online map. By participating, riders transformed their everyday phones
into multimedia tools for locative storytelling, using SMS and MMS rather than smartphone
applications. Unlike many reporting tools, ParTour emphasized group aggregation as the
primary mode.
Images were also taken by traditional journalists. The day after the ride, the Los Angeles
Times covered the event as an effort to produce a “crowd-sourced map”
156
, raising the city-wide
profile of the riders. Many pictures were taken at the Watts Towers. The photograph in Figure
31 is a visual statement that asserts the ride can be done with families – and that it was joyous.
Mapping was part of the broader campaign to assert collective identity, and to share the visual

156
See http://articles.latimes.com/2012/jan/22/local/la-me-bicycle-map-20120123
164

spectacle to fellow residents and the broader city of Los Angeles. (The importance of ties to
hyper-local journalists and media outlets will be discussed further in this chapter.)
The bike-mounted tablet was also used to show a live map of incoming pictures. The
map included any picture from the group that was tagged with a location. Landmarks were
photographed disproportionately, and clusters of similar pictures began to appear at the key spots
along the route. The ride leaders helped to direct the group’s attention, first by selecting the
route and then by stopping everyone for brief lectures on locations they felt valuable for
reflection.
Group mapping had its own theatrics in person – a parade that claimed attention on urban
streets and made public space into a stage. Wherever the group stopped, mobile cameras were
raised like props, implicitly proposing that the visual was worth capturing – that someone else
might want to see the picture later. Bystanders struck up conversations with photographers, as
they stepped back from the melee to frame their photograph. Images of the group and the
Watters Towers circulated online among participants after the ride. Creating media as a
collective act of branding and affiliation is a tactic that is increasingly supported online, with
social media tools and the tactics of participatory culture organizing (Kligler-Vilenchik &
Shresthova, 2012).
Data collection is often presumed as the goal in mapping activities. But here data was
secondary to mapping as a network-building activity. Echoing techniques of community
organizing, each action was seized as an opportunity to build participation. The future glory of a
print version of the map was extolled by organizers, always promising growth. The route itself
provided a linear structure to the narrative and to the activity, an arc that was “to the iconic Watts
165

Towers” and back again. The goal was to unify a coalition of organizations in storytelling their
shared vision for South LA.
A transmedia campaign emerges
Three months after the January ride, thousands of copies of the print “RideSouthLA” map
began to be distributed (see Figure 32). The map featured the route at its center, again echoing
the linear narrative. Paper was celebrated as the primary mode of distribution, perceived by the
designers as better for starting conversations face-to-face than the fanciest mobile application.
Pictures shown on the map were selected carefully through meetings with community leaders
and ride participants. Each picture adds to the sense of a collective narrative for the place and
the community. The overall graphic design reserves plenty of white space, hinting at what might
yet be added.

166

Figure 32: Printed map of the Watts Ride (front), woven from pictures
selected by neighborhood leaders
and ordered by the riding route to tell a group story

Source: map from RideSouthLA.com; photographs courtesy of Tafarai Bayne of T.R.U.S.T. South L.A.

The map emphasizes a “pedal-tested” legitimacy, from safe streets to historic landmarks
and local businesses along the way. The grainy images of the map emphasize their sourcing
from everyday cellphones, alongside the polished lines of a professional designer. The map is
handed out in follow-up rides to local bike and coffee shops, drawing on established
relationships between owners and residents, and aligning with hopes that the map might bring
more business to the region.
167

Much like the transmedia storytelling (Henry Jenkins, 2007) of Hollywood
157
, circulation
of the map happened across multiple devices and organizations. In neighborhoods like South LA
with disparate levels of technology access, print media remains particularly important for basic
access. The goal was to create spreadable media
158
that participants could seize and share across
distribution channels. For example, a printed map might structure an initial trajectory through
space with physical signs and billboards, with subsequent distribution complemented by email
blasts, mobile phone reposting on Facebook, and coordinated physical events to further cultivate
the group story.
A brand was strategically created for the project, in consultation with local bicycle clubs
and community organizing groups. A website was launched for the map with an ongoing blog
(based at RideSouthLA.com), as well as a similarly-named social media accounts (on Twitter
and Facebook). The website drew more than 2,000 unique visitors in its first six months, with
more than 80% coming from the Los Angeles area
159
. Unifying storytelling across digital, news
and organizational nodes was a recurring strategy for organizers.
In the first six months, coverage included a front-page story in the L.A. Times, the radio
station KPCC, and an article in Forbes. Just as importantly, coverage also came from local

157
Jenkins describes transmedia storytelling as a deliberate use of multiple delivery channels, leading to a
“coordinated entertainment experience.” Ideally any channel makes sense independently; various stories are often
told in parallel, contributing to a shared story world. For ParTour, the story world circles around bicycle clubs in
South LA, with frequent appearances by drivers, urban planners and community activists. Games in the Hollywood
model are often presumed to be digital (e.g., the game for Desperate Housewives). Yet theme parks like Disneyland
use place-based storytelling to do some important anchoring of the experience too.
158
The notion of “spreadable media” (Henry Jenkins, Ford, & Green, 2012) provides a vital counterpoint to the
temptations of “viral media”; in particular, the viral meme fails to account for any social agency by individuals or
communities in selectively choosing what to spread and to whom. From an empowerment perspective, spread is a
better framework to understand how communities can select media that tells their story and build collective identity.
159
According to IP address look-ups
168

media that shapes neighborhood storytelling, especially local blogs and cable TV (e.g., local Fox
affiliates). The project cultivated some media spectacles, like pictures with Los Angeles Mayor
Antonio Villaraigosa (see Figure 33). Politicians also showed up for rides, including a local
Congresswoman interested in connecting with the press and residents.

Figure 33: ParTour across storytelling levels: bike clubs, meeting the mayor, and local media. At
left: local cable TV interviews a ride leader about their reasons for making community maps. At
right: the Mayor of Los Angeles discussing a group map (posted by the RideSouthLA Twitter)

Sources: screenshot of local NBC website; photograph by the author

While bicycling appears as the map’s focus, the campaign goals were broader. A host of
social issues were embedded, including fighting entrenched poverty, the obesity epidemic, more
equitable access to open space, and even the housing foreclosure crisis. Across such diverse
challenges the map provided one unifying story, but to be heard it needed to tap into the local
ecology of communication.
Technique: selecting physical locations using “communication asset mapping”
Place-based games often need to pick local spaces to use, or routes to follow.
Simultaneously, they need to locate themselves in the communicative fabric of the local
169

community. Certainly this was the case with ParTour, where organizers picked the basic route
in advance. To avoid negative storytelling, the ride and route embraced a bottom-up tradition
called community asset mapping (Kretzmann & McKnight, 1993). The asset mapping
philosophy seeks to go beyond landmarks to identify places tied to local strength and community
capacity.
The communication ecology can also be approached via asset mapping. In
communication asset mapping, researchers and residents unite to identify locations in the
neighborhood that are vital for communication (Villanueva & Broad, 2012)
160
. These key
locations are where residents particularly engage in conversation and connect to community
organizations. Tattoo parlors and barber shops are important in South LA, according to research
by Villanueva and Broad (see Figure 34 for some of the data). Ethnicity is often a key
determinant in how such informal spaces of communication function.

160
The origin of most developments in CIT come from the Metamorphosis Project at the University of Southern
California. The asset mapping developments were supported by grants from the California Endowment.
170

Figure 34: Communication asset mapping in South LA includes physical spaces
where residents gather to talk about neighborhood issues

Source: The Metamorphosis Project, described by Villanueva and Broad (2012)

At full scale, the methodology described by Villanueva and Broad requires significant
time to analyze a geographically large storytelling network. Fortunately, the method can also be
adapted by community groups on a smaller scale. In the ParTour case, a light-touch approach
was intentionally explored. For the map route, a sub-neighborhood (along the Figueroa corridor
of South LA) was identified by the partner organizations based on their longstanding experience.
To identify communication assets, several community organizations gathered to debate what
their respective members and interests might prioritize. From these priorities, the mapping route
was determined.
A recurring theme in this chapter is the importance of working with organizations,
resisting the temptation in technology design of testing only with individuals in focus groups.
According to Communication Infrastructure Theory (CIT), neighborhood strength is a function
171

of local groups that organize much of the neighborhood storytelling (Kim & Ball-Rokeach,
2006a). Outsider storytelling and data collection can risk undermining community
organizations, which often facilitate the bottom-up assessment of community issues and
conversation on group priorities.
We found particular resonance with community organizations focused on local
knowledge and amplifying resident voices. As one club leader described in an interview
161
, “our
organization’s role is to help get folks… to see and document the community through their
eyes.” A more established nonprofit leader described their work as “community knowledge
development, treating residents like experts… Since [residents] know best what change they need
in their community, and what’s wrong… we develop tactics to draw that out.”
As a result of working with the community organizations, for example, the Watts Coffee
House was identified as a community location for inclusion on the map. The Coffee House has a
strong commitment to building community (it shares space with a school), credibility with
African American activists, and a “hip” and open atmosphere that welcomes conversation. By
including such spaces, the map pointed the cycling community toward “third places” and
conversation hot spots for bridging to the rest of the community.
In other words, the map sought to selectively identify social context, not just points of
interest. By contrast, most city bicycling maps focus just on features. They show bike lanes

161
In-depth interviews were conducted approximately eight months into the project, including all organization and
club staff that were deeply involved in the planning and execution of ParTour. These more formal interviews
complemented ongoing transcription of meeting notes, and more informal interviews and debriefs that took place
after events.
172

without connections to food, community or culture – let alone indicating how residents can have
a voice or get further involved.
Finally, the process of making the map is also part of the project’s transparency. Rather
than method-as-footnote, ParTour legitimizes the map data by introducing the map from the
beginning as coming from a community-based activity. When asset mapping is used, the
community deserves to know – if only to celebrate that process.
Compatibility as a Situated Game
How applicable are insights from ParTour’s activity to real-world games? ‘Game’ was
never a word used to describe ParTour. Yet the premise of this chapter is that the case holds
insights and methods for ‘local fit’. As in prior cases, using the framework depends on some
basic compatibility with Situated Games; a brief assessment of ParTour follows:

(a) Real-world actions. Is progress in the activity congruent with affecting the world?
Yes. Most immediately, ParTour took place within a real-world parade in support of
South LA. Taking pictures with ParTour augmented the bicycle parade, amplifying the live
spectacle (including the theatrics of raised cameras) and reverberating in group storytelling
online (as well as in participants’ personal social networks).
By contributing to ParTour, participants contributed to the joint media story about their
collective identity. The activity tells a real story to a live audience. The data was captured for
future use too, but much of the impact was real and immediate.

173

(b) Network-building goals. Does participating in the activity seek to build a place-
based community?
Clearly. A fundamental goal was to connect and align storytelling actors in South LA,
using bicycling to connect a host of place-based social issues. The network goals were
particularly apparent in the multi-level approach. Echoing the levels of CIT, ParTour sought to:
(1) connect residents to an open-streets and bicycling movement in South LA; (2) connect
organizations who share an interest in building a strong local identity; and (3) bridge hyperlocal
journalism with city-level news media. By aligning the three levels, the activity sought to turn
the storytelling network
162
toward a joint articulation of collective concerns and community
identity.
(c) Game-based. Is participation conducted via playful challenges, feedback loops and
uncertain outcomes?
Mostly no, but ParTour overlaps with games in some important ways. Compared to raw
mapping tools, ParTour was unusual for its emphasis on the activity of mapping, not just the
data. Mapping professionals and civic technologists alike tend to emphasize data quality, but
ParTour deliberately embraced fuzzy data in the first rounds, leaving space to increase accuracy
later via social deliberation.
163
For ParTour the data was often a means to place-based civic
engagement, a goal which we have called “situated engagement.”
164

162
The idea of “turning” a storytelling network has been described informally by the Metamorphosis Project; more
on such strategies will be addressed later in this case study.
163
For example longitude and latitude were often at the block or even zip-code level, with the understanding that the
precision could be increased through iteration. Iteration included group analysis of the pictures and map with
174

Next consider the explicit criteria for “game-based” activities. First, in contrast to many
dutiful civic activities
165
, a sense of playfulness filled the parade and mapping of ParTour. The
same bicycle with the tablet for pictures was also pulling a trailer with speakers blasting popular
music, and a seven-foot flag announcing the group.
Participants were challenged to locate community assets, and to use captions that
articulate the collective benefit. These challenges were especially honed to focus on either (a)
assets worth sharing, or (b) things the group wanted changed. Forcing data collection into these
two buckets – roughly one positive and one negative – made the activity more challenging, and
ultimately more engaging.
166

Mobile media has distinctive power for real-world feedback, in part for the ability to
move between foreground and background awareness, being both pervasive and persistent (Ito,
2005, p. 15). Feedback loops gave just-in-time information about what peers were

representatives of various stakeholder groups, organizations, and the graphic designer. The same “process over
product” was emphasized in distributing the map, and in recurring events that returned to the route.
164
Situated Engagement is a kind of civic engagement that depends on being deeply situated in a place and activity.
The term emphasizes how the experience of civic engagement can be transformative under three specific conditions.
The conditions are that the activity must be: (1) situated here in physical places – not anywhere; (2) situated now in
social experience – not anytime; (3) sustained in building group capacity – not just in the improvement of
individuals. The phrase was proposed by François Bar and myself, and a joint paper to theorize the term is
forthcoming.
165
See work by Bennett et al. (2009) on dutiful participation versus more self-actualizing civic modes
166
The temptation with tools is to offer unlimited choice, and open-ended missions. Yet the principles of games
emphasize that choice is more meaningful when it is constrained. Our testing of ParTour confirmed this principle
with the mapping missions: when we let people map anything, the feedback loops were harder to manage and the
engagement was less than when the number of missions was constrained. Our testing found that two “missions” are
often most useful for community-based asset mapping: one focused on sharing the positive with outside audiences,
and the other focused on identifying areas for improvement to rally internally. Each mission was given a hashtag;
for example, submissions to the ParTour system were might be tagged #gem or #2change. The communication
perspective was retained by talking with the group about how places of social interaction can be particularly useful
for organizing and building social cohesion. When social practices were emphasized by leaders, participants took
more pictures of people in social settings, rather than just objects alone. In practice, submissions often were not
explicitly tagged; there are many reasons for this, including simply that tagging was not a standard practice among
our participants in their personal lives. Thus while the tool was open-ended, the missions were focused by the social
practices and leadership of the group.
175

photographing and saying, and sparked conversations comparing what submissions meant to the
group. Feedback loops included the tablets of ride organizers (see Figure 35), where the pictures
were shown back to the group, sparking conversations that compared submissions.

Figure 35: Bike-mounted tablet used to provide feedback loop of pictures submitted to the group
which can also be seen on the web as a photo stream (at right)

Source: photography by the author; screenshot of VoJo.co for RideSouthLA

The group was also challenged to cross-post their pictures to their personal social media
networks, which started a secondary feedback loop featuring responses from distant friends who
were interested in the ride, and conversations with fellow riders that continued after the ride was
over. Similarly, the print map served as a feedback loop. At a later event when the map was
handed out to a group of bicycle club members, the enthusiasm was often palpable: “that’s us!”
club members would say excitedly when they saw their group pictures on the polished design.
Seeing their images tied to a professional-looking map helped provide local groups with a sense
of ownership of the map, and helped foster a sense that the contributions had been meaningful.
176

Most lacking in terms of games was the element of uncertainty
167
. In general, the
activity was made reassuringly predictable, not deliberately uncertain. Yet there was some
inherent uncertainty in the act of group mapping. After submitting a picture, participants could
not be certain about the reception: would their picture be discussed, reposted to social media, or
even featured in the print version? The desire for a positive reception affected the choice of
pictures, and provided an implicit challenge.
What have we learned? Most importantly, ParTour was already aligned with many
characteristics of games. The alignment seems due to ParTour’s emphasis on being an engaging
mapping activity, rather than a way to collect missing data. At the same time, analyzing a civic
activity in game terms is quite unusual. Implicitly, the analysis makes clear that ParTour is
much more than a technical system for publishing pictures by cellphones. For civic organizers,
the game lens and criteria help reveal the underlying functioning of ParTour as an activity. The
diffusion of ParTour to other under-served communities may require a description of its function
in terms of social practices (not just its technology), ultimately helping more communities with
access to social opportunity.
An Atlanta game comparison (Re:Activism)
Integrating ParTour within a full game might be possible, but would not be simple.
Studying the integration and adaptation of ParTour is valuable, in part because the activity is so
deeply aligned with storytelling of a local ecology of communication. The most simplistic way

167
In general, uncertainty must be resolved with each iteration of the activity (like submitting a picture to ParTour)
so that players know how their contribution mattered; thus the activity needs the feedback loops to show how the
challenges turned out, empowering players to playfully iterate and improve their civic contributions.
177

to add ParTour to a game is to borrow only the technical features – in other words to use the
group storytelling tools. Treating ParTour as a technology would strip out most of its social
practices and game-like features, for better and worse.
In late 2013, an established game called Re:Activism incorporated ParTour when visiting
designers localized the game for Atlanta as part of a museum exhibit.
168
In brief, the game of
Re:Activism is played by small teams that race around the city, competing to reenact events from
civil rights history. Many of the reenactments feature peaceful protests, which provide excellent
public theater – primed for cellphone photography. The original game was made for New York
City in 2008, and has subsequently been adapted for Beijing, Minneapolis/St. Paul, and
Philadelphia. Over two months in the summer of 2013 as a form of design and action research,
the author of this study helped to incorporate ParTour into the first version of Re:Activism
Atlanta.
When the designers brought the game Re:Activism to Atlanta, they hoped that the mobile
media tools of ParTour could make the storytelling more public and thus have impact that
extends to the real-world. Specifically, the activist tools of ParTour would be used for public
photography and persistent storytelling.
The basic tools behind ParTour are deliberately low-fi and were created through a
participatory design process
169
. (The tools are now are available through a free service called

168
The game was adapted for Atlanta by Colleen Macklin with Adam Rafinski and Ben Pincus, as part of an exhibit
at the Museum of Design Atlanta (MODA) called “XYZ: Alternative Voices in Game Design.” Credit for the initial
development of Re:Activism go to PETLab, a public-interest game design and research lab based in New York City.
169
The initial design was called VozMob, which began in 2008 and was focused on immigrant workers around Los
Angeles. The participatory method fits into a broader tradition of community-based participatory design (DiSalvo,
Clement, & Pipek, 2013). The success of the VozMob process is partly evident in the ongoing use of the tools to
178

Vojo.
170
) The tools show how seemingly outdated mobile technologies can be appropriated and
re-invented for new uses to empower a community, echoing the strategies of mobile design
proposed by Bar et al. (2007).
Does the method apply to games? Participatory design has been theorized more with
making tools than making digital games. Of course, game design is a very “user-centered”
process, but one that rarely seeks to share the power to make design decisions with participants.
The limited history of participatory game design includes recruiting a group of target users to
join the process as designers,
171
and empowering users to localize and “mod” a game.
172

However, participatory game design has largely overlooked the possibility of being “community-
centered” and sharing the power to make decisions with existing nodes of power in the
community, especially the community-based organizations emphasized in ParTour.
The Re:Activism designers had a simpler goal: to make their game more real-world by
empowering players with real-world tools. As the analysis below will show, the gap between
individual and group empowerment is deep fault line for Situated Games, and the framework of
‘local fit’ makes clear some of the limits. For Re:Activism to go further – including by aligning

publish grassroots stories from community correspondents (see http://VozMob.net). A powerful case study has been
written on the original participatory design process (Mobile Voices (Añorve R., F. Bar, M. Brough, A. Cisneros, S.
Costanza-Chock, A. Lucía Garcés, C. Gonzalez, M. de Lourdes González Reyes, C. Jimenez, C. Lapsansky, M.
Mancia, M. Rodriguez, and C. Wallis), 2011).
170
See http://vojo.co/
171
For example, see Vanden Abeele and Van Rompaey (2006).
172
Some theorists have pushed for greater recognition of players as already playing a role in the design process (T.
L. Taylor, 2006). On rare occasions, game design has also adopted a participatory stance for place-based scaling,
perhaps most notably with the ARIS project. ARIS is a platform for creating place-based games using a smartphone
toolset; they have used participatory design in powerful ways – see in particular, “Participatory Scaling Through
Augmented Reality Learning Through Local Games” (Martin, Dikkers, Squire, & Gagnon, 2014).
179

with the community identity and ecology – would necessitate a significant shift in the project
goals.
Playing Re:Activism
In brief, a game of Re:Activism takes about three hours to play, rain or shine. Teams of
4-6 players race against the clock to visit historic sites around the city, choosing the order as they
go. Points are earned by teams through location-based challenges. Each challenge involves
creating theater in public spaces to dramatize a historic event in civil rights history. Often the
enactment is on a historic location of civic protest. Mobile phones are used by players to
communicate with game facilitators and to document their activities for game points.
Documentation in the Atlanta version came largely through the ParTour technology (see Figure
38).
Meet the blue bandana team. Teammates included two local college students who had
grown up in Atlanta, and two game scholars who were in town for a conference – an unlikely
group to have met otherwise. The bandanas gave each group a unified look, akin to teams on a
sports field
173
. Visual signifiers like bandanas made it easier for players to approach strangers,
as required for several missions in the game.
For simplicity the blue bandana team began with the nearest challenge, a few blocks from
the starting point (see the neighborhood map for point ‘B’ on Figure 37). The game instructions

173
Of course, colored arm bands have also been used by street gangs. While gangs represent strong place-based
networks, such associations were not the organizers’ intent. The game avoided the issue by staying in downtown
areas where gang rivalries and color affiliations were a more distant concern. Unintended consequences of street
activity are a clear reason for consulting with local experts (i.e., a check provided by ties to the local ecology of
communication).
180

told them to look for the “Cotton States and International Exposition of 1895.” After searching
for a few minutes, a cry of “I found it” was heard from one player who discovered a tall plaque
with the name of the Exposition (background of Figure 36).

Figure 36. Re:Activism players reenact a historic speech, and document with camera phones; at
right, players enact a protest march to protest hate speech

Source: Photographs by the author

To prove they found the historic plaque, the blue team was instructed to locate the last
word in the historic marker’s text. This word was easy to identify in person but would have been
hard to find online. The word was verified by sending a text message to the “game
headquarters.” Then the real challenge began.
Points came from enacting a historic event at that location. By consulting their game
cards (Figure 37), the blue team learned that Booker T. Washington had given a famous speech
at the Cotton States Exposition in September of 1895. Washington’s “Atlanta Compromise”
speech called for racial cooperation. As a warm-up for 100 points, the blue team simply had to
interview a passerby about the Exposition, and ask what they knew about it – if anything.
For the reenactment, a member of the team had to publicly read an excerpt from
Washington’s speech. Video documentation was required (see left in Figure 36) to receive the
181

full 200 points. The team’s actions drew the attention of passersby who paused mid-walk to
listen; such walkers might pay more attention to the plaque on a future walk of their own, and
read it out of curiosity.
The greatest opportunity to attract attention was the 300-point action, which involved
staging a miniature protest at the location on the same issues. The specifics were left to the
group. Recruiting a bystander increased the point value of the protest by an additional 50 points.
Sign-making materials were provided, often used in staging marches (see Figure 36 on the right).
When players recruited a stranger to participate, they had to cross the boundary (or
“magic circle”) between gameplay and real-world. It is often awkward to talk with strangers; the
game gives cover for introductions, and even to try new civic behavior (Macklin & Thomson,
2011, p. 166).
After a few tries, the blue team settled on some standard language of introduction. They
explained to strangers that, “We are playing a game about civil rights history, and we get more
points if we can convince a stranger to help us.” By explicitly acknowledging the game, players
lowered the stakes for strangers too. Some were uncomfortable and declined; others agreed
readily.
To design the game, the various locations were selected in consultation with a professor
of history at Georgia State. The Re:Activism designers sought to be location-specific and address
“public space as a site for the interplay of social issues” (Macklin, 2010, p. 275). For accuracy,
designers sought to verify the locations by inviting community members to a trial run before the
public version of the game.

182

Figure 37. Selected materials from Re:Activism Atlanta, including mission map
and location-based challenges

Sources: materials courtesy the Re:Activism Atlanta team; photograph by the author

Outcomes of the game were often described in terms of participant education. Yet the
learning itself was more about attitude and civic disposition than the traditional knowledge gains
measured in classrooms. An assessment of the game in Philadelphia found little retention of the
historical specifics like where/when/who, but positive signs of an “attitude of activism” (Macklin
& Thomson, 2011, p. 163). In particular, the study found attitudes and strategies that fostered
social collaboration. The evaluation describes one player responding to the question of “what
did you learn” by answering “A lot. All about approaching people, being positive, trying to get
their answer, interact.” The skills were thus interpersonal – including greeting and meeting
strangers in public space.
What was missing from the Philadelphia version – why add ParTour? Philadelphia
teams already used camera phones for documentation, so why not make the stories public with
183

real-world activist tools? The ParTour tools are real-world, used worldwide by grassroots
groups to tell collective stories.
174

Grassroots media is increasingly recognized as an important part of modern protest.
Protest depends on having a strong collective identity (Opp, 2009, Chapter 7), which can emerge
from grassroots media as participants collectively work to make sense of a situation and frame
it.
175
Protest is an act of being noticed, and is increasingly staged as media-friendly spectacle,
drawing attention through tactics in person – but also online and with media strategy
(Duncombe, 2007; Earl & Kimport, 2009). The communication of spectacle is rarely valorized
by civic philosophers, in part because civic life is often described in terms of being informed and
making rational choices.
Yet spectacle can and may need to be aligned with the rationalism of enlightenment
traditions. Rather than oppose spectacle, an important strategy is to seek an ethical kind of
spectacle that builds atop rationalist arguments (Duncombe, 2007).
176
Such dualism has
historical precedent going back to the Ancient Greeks, where being a citizen demanded
participating in two ways: making rational sense of speeches, and the distinct persuasive modes
of emotional appeal (Garver, 1994, p. 114; Pelling, 2005).
In Re:Activism the activity gave access to the drama and performances of civic protest.
The goal is not to turn players into protest leaders, but to give them access to a tacit learning that

174
Groups using the underlying VoJo toolset include the Cambridge Campaign Against Domestic Violence, Sandy
Storyline (a bottom-up documentary on Hurricane Sandy), the Erase the Border campaign on the US-Mexico border,
and Quilambos in Brazil; the basic tools behind ParTour/RideSouthLA are based on the mobile phone blogging
platform of VozMob/Mobile Voices. More on the VoJo examples is available at http://VoJo.co/en/examples
175
Early “rationalist” models of protest often omitted collective identity, focusing more exclusively on resource
mobilization and political opportunity structures. Both perspectives are useful for predicting protest.
176
Stephen Duncombe has advanced a useful preliminary model for ethical spectacle, as will be discussed in the
final chapter of this study.
184

is otherwise high risk. Civic protest is a rare activity for most people, and the designers of
Re:Activism sought to recreate protest in public in a safe way, providing a personal access point
(Ruiz et al., 2012). Simultaneously the activity also accomplishes real civic work by starting
public conversations that are situated in historically important (and often overlooked) locations.
When ParTour was introduced, the Atlanta version of the game changed course in some
subtle but important ways. The stream of pictures (see Figure 38) was still visible to the game
facilitators, but now it was also visible to competing teams and even the Internet public.
Implicitly, ParTour brought an online audience to watch the game. A few savvy players realized
that they could gain hints about the game in real-time by watching other teams around the city.
(Defensive strategies are possible too, such as deliberately misleading other teams.) In other
words, a new feedback loop was added to the game.

185

Figure 38: Pictures of Re:Activism Atlanta by players
using the storytelling tools of ParTour

Source: screenshot of ParTour tools (VoJo) based on #reactivism tag

Taking pictures in a real-world game brought some controversy. One team encountered a
homeless man in a park; they debated that while the image would be dramatic, it could also be
exploitive. Their debate was largely outside the feedback loops of the formal game, for better
and worse. Most teams encountered at least one person who refused to have their picture taken
under any circumstances. ParTour shifted the picture-taking dynamics to be more real-world,
redoubling the potential for exploitation but also giving a pathway for real-world utility. On one
hand, this meant that some residents were less likely to participate, since their image would
appear on a public website. Alternatively, if the players told the person’s story respectfully, the
narrative could spread beyond the magic circle of play to the real world. At least some teams
were quite deliberate and careful with the scenes they enacted, trying to reduce the possibility of
outraging concerned local groups who might later view the pictures and captions.
186

A new kind of legitimacy check emerged with the addition of the online audience.
Accountability is one measure of a real-world game, though the pathways can be hard to
anticipate. For example, one photo was captioned “Statue of Barbara Asher” with specific cross-
streets; this photo might now appear in web searches about the statue. By having the photo
stream persist after the game finished, the consequences of play also persisted with media stories
that could continue to circulate. Very few did, but as with video sharing platforms like
YouTube, a single controversial picture can be taken up by a community and spread (Henry
Jenkins, Ford, & Green, 2012).
The persistence of game-created stories had some benefits for players and communities
too. Players after the game could return to find pictures of their team, reflect on the experience,
and spread stories of the activity to their network. All players received an email the week after
the game, with links to their pictures and to articles to learn how real-world activists were using
the same ParTour/Vojo tools.
For the community, the trove of pictures provided a content repository for collective
benefit. Yet there was not much pickup in Atlanta by local organizations.
Partially qualifying as a Situated Game
Although clearly a game, Re:Activism could hardly be called a Situated Game. The
activity did not seek to build local networks; nor did the game designers greatly attempt to affect
the real-world beyond short-term theatrics and educational outcomes. Geography and history
were used in the game, but the implied community was not particularly local – the sense of place
was more often about “the South” than Atlanta, let alone a more local neighborhood. (One
187

designer attributed the geographic spread to the decentralized nature of Atlanta, and the difficulty
of finding a set of historic protests that took place in walking proximity.)
Of course, the game was educational in a very active way, attempting to give players a
taste of activism, and turning bystanders into temporary participants as well. As the chief
designer described it in an interview, the game might be understood as a kind of emancipatory
theater that pushes participants to think for themselves about locative history and their own
capacity for protest. (There are strong parallels to what Augusto Boal (1979) called the Theater
of the Oppressed.) Such theater can be a powerful educational tool, but largely for individuals in
a social context, as opposed to collective empowerment.
When ParTour tools alone were added, the real-world component increased a bit. The
tools helped turn the documentation process into a media spectacle that could reach additional
audiences. But it was unclear who was watching, aside from competing teams. Incorporating
the storytelling tools of ParTour was not sufficient to engage with real-world audiences beyond
the game.
By contrast, ParTour as a broader activity had empowerment goals that were collective,
and group actions that sought to shift the neighborhood brand of South LA. Engagement was
necessary. Understanding the tradeoffs of incorporating ParTour as a raw tool becomes even
clearer below, as we briefly consider ‘local fit’.
Missing: the group identity (for fit)
Collective identity – which may include the neighborhood brand – is one of the three
central concerns of ‘local fit’. In Re:Activism, no strong identity of place was invoked.
“Atlanta” was more of a geographic boundary, a filter for events in the history of civil rights.
188

Participants were only minimally asked to come together as members of the Atlanta community,
or any other place-based collective. By contrast, ParTour in South LA had a primary emphasis
on performing local narratives of neighborhood brand and community identity. Was the
difference simply a matter of emphasis – or is there a more fundamental difference in strategy
that allowed ParTour in South LA to optimize for group identity?
The most significant difference is that ParTour in South LA focused on existing
community groups who were seeking to build a collective identity already. Around the brand
RideSouthLA, an unusual set of semi-formal associations had emerged around bicycles. From
these groups, the narrative of collective identity emerged.
In fact, the bicycle clubs were themselves a neighborhood story. Groups like the East
Side Riders and the Compton Schwinnmasters for Peace have some alignment with the broad
rise of bicycle culture in Los Angeles
177
. At the same time, these two clubs in South LA are a
world apart from the stereotypes of young hipster bicyclists. They are predominantly middle-
aged African-American men, who ride in groups to “promote peace and a healthy way of life.”
In their spare time, they custom-fit their bicycles – including with stunning chrome and hinges
that require special welding. Their group rides often include community service in an adjoining
neighborhood, or they might join with an anti-obesity campaign to encourage exercise and good
diet.

177
Midway through the study of ParTour, L.A. County released a bicycle plan for the first time in 37 years that will
“vastly expand” the existing 144-mile network, adding 832 miles of new bikeways (Yaroslavsky, 2012). Meanwhile
an open-streets event called CicLAvia has demonstrated unexpected success in drawing more than 100,000
participants – most on bike – to the streets of the city several times a year.
189

As a group, the club members have a collective narrative already. Often the narrative
invokes a shared problem of marginalization, since “Watts gets overlooked – we’re at the tail end
of the money, at the corner of the county, and the corner of Compton – we get forgotten because
[different city officials] represent the pieces.” Their story of the unusual bicycle clubs in South
LA was under-told
178
, partly because they were informal
179
. The East Side Riders – a core group
for the ParTour map – feels more like a club than a structured organization.
In motion with their custom chrome bicycles and uniform swath of red shirts, leisurely
taking over a full lane of traffic, the club created a noticeable spectacle. Part of what they
perform is their group identity, and their dedication to serving their own members. For example,
a leader with the East Side Riders insisted in interviews that his group is about “more than
bicycles – we are also about families, ex-gang members, black-and-brown, health… the [Watts]
community.” The Riders group cuts across social issues more than most nonprofits can afford
given the nonprofit need for mission statements and focused funding strategy.
There are many such bicycles clubs in the broader Los Angeles area, as we quickly
discovered. After the first map was printed, ParTour began producing less formal maps for
smaller events, and explicitly targeting similar bicycle clubs in the greater Los Angeles area (see
Figure 39). With relatively little effort, one exploratory ride recruited 120 riders. What is

178
Their story was under-told in part because they did not seek to integrate the CIT elements – they did not seek
media connections, and their interpersonal networks were largely separate from connections to local media and more
established community organizations. For example, the groups made heavy use of Facebook and text messaging to
organize rides – but these messages did not leave their private “friend” groups, and they did not reach out to invite
local journalists. Computer literacy in general appeared to be low.
179
Part of what makes the clubs hidden is their informal (tax) status. With nonprofit status, even tiny organizations
elsewhere in South LA tend to have relatively more structure, including the ability to seek grants, pay staff and file
taxes. In CIT terms, the clubs are more of an interpersonal network than an organization – but they still play an
important storytelling role in the ecology.
190

unusual is that the group included 12 clubs representing 43 different zip codes. There is a spirit
of reciprocity between clubs, hastened by those who ride and conduct community service in
another area. We have referred to this as “bicycle diplomacy,” echoing the efforts by US
diplomats to use bicycles in China in the 1970s.

Figure 39: Flyer describing a ParTour event with multiple bicycle clubs
from across Los Angeles

Source: RideSouthLA.com website

The clubs also demonstrate the importance of looking beyond formal nonprofit
organizations when looking for local partners. Clubs have their own impact strategies.
Remaining informal with a near-zero budget and an attitude of do-it-ourselves can actually be a
strategic move. One club leader explained this strategy in contrast to more direct efforts with the
city: “we tried to sit down with city planners about the need for bike lanes in this
neighborhood… but the county was telling us that it first takes a study of the street, which might
cost as much as $45,000 to evaluate.” The organizer viewed the light-touch approach of
ParTour as a real alternative, describing the approach as “let’s just do it – get out there and go!”
His hope is that the city will see this grassroots success, and be forced to admit, “if they can
[bring the people together] with limited-to-no funds, and well, we [city officials] have money set
191

aside for roads and things, then let’s give it a try … and put more bike lanes in this area.” In CIT
terms, bargaining power comes from community strength in proportion to integrated storytelling.
Could Re:Activism be optimized for local identity? The game was already staging a
performance in public space. At first glance, it might seem simple for the game to incorporate
some content around the identity of Atlanta as a community (akin to Macon Money’s effort to
brand the College Hill Corridor). However, seeking to build a real-world identity would have
been highly problematic. The problem hinges on having a high percentage of participants who
are not from the neighborhood. When outsiders are participants, the activity would have deep
problems of legitimacy if it attempted to assert a real-world community identity.
Cultural and network alignment may also be necessary for credibility in historically-
marginalized neighborhoods. Research outside of games has established the problem with youth.
For example, youth journalism programs with a mix of black and white students in urban areas
often have suppressed participation from black youth because the civic activity is perceived as “a
white thing” (Marchi, 2011). One cause is that low-income minority youth feel acute
“disjunctures” between the rosy ideals of civic life in textbooks versus the reality of their lived
experience (Rubin, Hayes, & Benson, 2009). As a preventative measure, aligning with
neighborhood organizations and a locally credible narrative may be vital for legitimacy,
including for recruiting and retaining participants. The designers of Re:Activism did seek to
involve local museums in their work – including an emerging museum of civil and human rights
– but more for content expertise than for their role in convening and facilitating resident
networks.
192

In sum, ethical fit to local identity may require explicit work to involve local groups
running the game, largely for themselves; then the identity narrative can emerge more naturally.
Again, Re:Activism was never intended to empower local groups (and therefore it is not a
Situated Game), but it may need to involve local groups just to legitimately address how
collective history has current meaning. The value of ‘local fit’ is partly in making clear that
addressing the group identity requires first having goals of collective empowerment.
Additionally, the uphill battle described by Re:Activism designers to find local partners is an
important reminder that partnership is one of the hardest components of localized design.
Missing: local media (for fit)
The difference in using ParTour as a tool is further underscored by considering the
second dimension of local fit. In brief, Re:Activism had a much narrower approach to the local
ecology of communication, focusing primarily on communication at the interpersonal level, with
few ties to local media or channels with local organizations.
Certainly, Re:Activism does fit reasonably to interpersonal channels of communication,
especially for public space. Communication in the game took place in prominent local spaces for
residents to talk, including a barber shop, parks, town squares and inside one restaurant.
Methodologically, the designers identified key nodes for interpersonal communication
with a shortcut that deserves mention. Rather than mapping communication assets, they
researched and identified locations in Atlanta where major civil rights events took place
180
; many

180
In fact, the process itself was “almost as much a research project as a game design exercise,” according to one of
the lead designers.
193

of those locations continue to be vibrant gathering spaces today. While likely incomplete, this
historical approach has merit, and future game designers simply seeking to identify
communication hotspots could do worse than to begin by reading the protest history of the
neighborhood.
On the digital side, Re:Activism benefited from using the phones that were already used
by participants. As a result, it was easy for participants to spread images in the game to their
personal networks. Adding the tools of ParTour rounded out the digital side of interpersonal
communication by providing a one-to-many channel for addressing fellow players.
Yet the interpersonal channel is only one of three levels in the ecology. ParTour built
ties to the second channel identified by CIT: communication via local media. Local media is
important because it tends be more focused on the stories that matter for group cohesion and
collective identity.
181
Why bother? Even when participation rates are high, most activities only
involve a small fraction of the community. Without local media, the storytelling is often
invisible. The need for scale is a similarly strong justification for social movements that stage
spectacle for the media.
Inadvertently, the Re:Activism channel could be picked up by local media, although there
was no evidence of syndication or syndication attempts. The educational approach of
Re:Activism was more to train participants about media spectacle, rather than to foster a media

181
Again, CIT emphasizes the distinct importance of community stories: those that concern the place-base
community, including problems facing the whole group, how the group has overcome challenges, and where the
group might go in the future (Kim & Ball-Rokeach, 2006a). According to CIT, the local media is crucially important
to deepening neighborhood identity, belonging and efficacy beliefs. (The terminology used by CIT is
“neighborhood stories,” for many good reasons; the slightly more generic “community stories” is used here for
accessibility with designers considering place-based communities, such as that of Reality Ends Here.)
194

spectacle that had real distribution. They used ParTour to push for critical thinking through
media making tools, not to directly promote stories from the game. In practice, no media targets
were identified by Re:Activism for use.
By contrast, in South LA the activity was designed to foster media ties with specific
channels for bridging. Media bridging considered mass and geo-ethnic media. Geo-ethnic
media
182
is often overlooked by media campaigns that focus only on reach, since geo-ethnic
outlets can be smaller; yet for the right kind of local stories, mass outlets like the Los Angeles
Times can be less reliable in frequency and getting the right frame. Research has shown that
community newspapers have distinct benefits over metro papers, including for community
integration (Finnegan & Viswanath, 1988). Of course, when mass media gets the story right,
there is local value too including external visibility and a sense of validation.
In South LA, ParTour organizers recruited local reporters, and placed the bicycle clubs at
the center of a narrative about community mapping. For the mainstream journalists, a Saturday
ride with ParTour to “map the neighborhood” was a community event worth covering even
before the map was published. The online stream of stories gave instant credibility to the story
of bottom-up mapping as a noteworthy group process. Meanwhile, hyper-local journalists found
the act of mapping to be a chance to renew conversations that went deeper, finding out what they
wanted to photograph and share as part of ongoing coverage. Importantly, neither media source
used many of the actual ParTour pictures, insisting on their own photography as part of
independent journalism.

182
Geo-ethnic media is defined as media that targets a specific geographical area (such as the blog Intersections
South LA), especially with emphasis on a particular ethnicity in the area (Kim, Jung, & Ball-Rokeach, 2006).
195

The presence of media stories connected back to the organizational level of the ecology.
One impact was seen with the East Side Riders, who used the press coverage to legitimize their
informal bike club with a larger nonprofit. How? One club leader explained that they were
doing the “same things we were already doing, but now we’re getting exposure for it – now that
we’re bringing more exposure to this place, they [the nonprofit] see us as real partners that can
help expose this place.” The activity strengthened the links between residents, informal clubs
and more formal neighborhood organizations.
For Re:Activism, targeting local media would have required additional investment – but
perhaps less than might be thought. The experience of collaborative design in South LA
revealed that neighborhood organizations often have valuable media connections already.
Involving local organizations was thus a way to connect with media, while also building media
ties for grassroots partners. For large institutions with media outreach staff (like at the local
University), it was more natural to target media at the city and national news level, which with
care could be linked to the authentic local stories emerging from residents, local organization and
hyper-local media.
Looking back at Re:Activism Atlanta – especially in terms of group identity and local
media strategy – stark differences have emerged. Compared to ParTour in South LA, the
Atlanta game stayed on the surface of community empowerment. Yet the possibility for
improvement goes in both directions. For the remaining dimension of fit – the mechanics of
greeting and meeting – it is worth returning to see how the original ParTour activity might be
improved.
196

Optimizing non-games via mechanics
Can bottom-up activities like ParTour benefit more directly from game principles? Once
the ParTour team started optimizing the activity as a challenge, new strategies started to emerge.
In particular, a novel strategy for distributing the paper map emerged from optimizing ParTour’s
mechanics. Part of the strategy was immediately actionable, and the rest points to how a full
game in South LA might deepen local networks further.
The analysis begins by considering ParTour’s implicit mechanics. Each phase of the
activity can be described in terms of a primary mechanic. In the mapping phase, the mechanic
centers on repeatedly taking pictures together that identify community assets, and captioning the
pictures with implicit public greeting.
183

The second phase of ParTour concerns the paper map distribution. Distribution took
place in many settings, including by bicycle. A few of the tactics will be briefly described,
beginning with the bicycle clubs as a key test group.
One afternoon, the East Side Riders joined with volunteers from the university and
headed to local businesses that were featured on the map. One was a local bicycle shop (see
Figure 40), a promising hub for distribution. Once in the store the clubs had to establish their

183
The first mechanic can be more precisely summarized as “photographing and streaming community assets as a
group.” The basic action is repeated for every picture that is submitted to the mapping system. The fit to existing
modes of greeting and meeting seems reasonable, since local groups already made widespread use of cellphone
photography – albeit for social uses more than deliberately co-creating a narrative of the community. (Further
description of this first mechanic is omitted, given the prior description of the mapping phase.)
197

legitimacy by chatting for a bit, since business owners often refuse the requests of well-
intentioned outsiders
184
.
For any paper-based distribution, the recycle bin is the enemy. Once permission was
granted to distribute the maps, the challenge was how to make them visible. Fliers in the bicycle
shop proliferated about concerts, advertisements, etc. The distribution team realized that a pile
of maps would drift to the back, and disappear within a week or two. (Weeks later, follow-up
rides confirmed that the maps had retreated from the front of counters and windows, pushed back
by an ever-flowing stream of fliers.) Unexpectedly, the most enthusiastic feedback appeared
online, where a photo of the bike store trip spurred one resident to rush back to the store for their
own copy.

184
In this case, underscoring the importance of local partners, we discovered that even though the map was free,
distribution was initially refused. One particularly articulate store manager explained that outside groups had often
come to the neighborhood in the past, raising expectations that were not always met, especially after students
graduated or funding dried up. They would rather not risk the disappointment. Bicycle club leaders were able to
talk about where they grew up, find people they knew in common with the owner, and provide enough cover to
leave behind the map. In other words, existing social ties were a prerequisite to distribution.
198

Figure 40: To distribute the map, bicycle club members visit a local bike shop;
afterward, a picture in front of the shop sparks
conversation online that connects physical with digital

Source: screenshot of Facebook exchange

Games invert the traditional celebration of efficiency. Traditional distribution strategies
try to maximize outreach while minimizing effort – like printing thousands of copies of a map at
199

scale. By contrast, game-based strategies celebrate the engagement that comes from a good
challenge and meaningful choices. Rather than mass distribution, the challenge of face-to-face
contact can lead to satisfying feedback and interpersonal connection. Our team began to wonder:
would distribution improve if balanced akin to games, where the challenge was to distribute the
map through conversations?
Our pilot with the bike shop pointed to a sustained strategy, including with individual
residents. Instead of attempting to hand out as many maps as possible, we could seek to deepen
the social ties behind each exchange, and maximize the connections and impact. To demonstrate
a genuine openness in the conversation, we started emphasizing how the map was unfinished.
Distribution could thus begin by asking, “What is missing?”
To answer, newcomers had to actually look at the map’s specifics. As a result, one-on-
one conversations became more substantive. Responses often included a discussion of
neighborhood priorities, what we should tell outsiders, and how the community might grow. In
other words, residents were having conversations about neighborhood issues.
The map was graphically designed to support conversation too, given the white space
showing openness for growth. Local bike clubs committed to “riding the map” again to reflect
on what should be added, and began posting ride details to the map’s website and social media
pages. By embracing an open story, ownership could be more easily shared and interest
sustained.
The strategy also worked when pitching the map to a captive audience. We experimented
at community events hosted by local organizations (e.g., a cooking demonstration at a
community garden). After handing out the maps, questions would be posed to the group: “What
200

is missing? Are there any surprises? How does this help our community?” Hands would go up
quickly, and ideas would flow. Only once the basis for conversation was established did we turn
to distribution, asking: “Where in your week do you encounter social groups that might
appreciate the map?” Individual answers often inspired nodding heads and ideas from other
group participants.
After group rides, the strategy could again be used to connect the digital to physical
experience. Real-time pictures (again, see Figure 40) can make the digital conversation feel
authentic and immediate. To prepare in person, participants were encouraged to post to their
personal networks with a link to the map, and again the question: “What is missing?” Social
networking websites can thus be something of an in-between: a place for distribution “by hand,”
when residents each post the map to their own feed and urge neighbors to try it and pass it along.
As with the collective mapping process, distribution can become a group effort. Like
community organizing, the goal is always to build the network. The distribution mechanic can
then be articulated as: “challenging peers to fill in the gaps in the community story,” and
optimized to fit with existing modes of greeting and meeting.
A thought experiment: distribution as a game
How large is the gap between ParTour and a more complete game? The approach above
only implemented a few game principles, mainly to shift away from efficiency and challenge
participants as a means to open and deepen participation. A thought experiment can help to
probe further, and illustrate why an explicit game might be useful – and yet carries its own costs.
Let us return to the bicycle shop, and consider how the distribution of maps could be
made into a game for the bicycle clubs. Competition is an easy way to make the stakes matter.
201

(For example, the club that wins the game could be given leadership privileges for a large ride in
South LA and receive considerable press attention.)
Rather than distributing the maps one time only, the game would involve returning
regularly to each store, and checking in with bike shop staff. Each refill would be deliberately
small, and perform a theatrics of care (not excess). More importantly, each refill visit is an
opportunity for neighborhood conversation – a chance for residents to gather input, to ask a shop
owner what they’ve been hearing about the map from customers, to ask what should be added.
However, challenging clubs to constantly refill maps at all bicycle shops in South LA
sounds like tiresome work. The irony is that a full game might require limiting how bicycle
clubs can distribute the map. Deliberate scarcity encourages meaningful decisions – like forcing
clubs to choose which shop should receive maps, or even limiting the number of maps they can
distribute. Scarcity can help make the stakes matter.
The winner is the club that signs up the most newcomers. Each newcomer must sign up
by texting a unique code on their map. Uncertainty is cultivated with a careful catch: each week,
clubs only get credit if they have correctly predicted the number of signups for their top three
stores. Suddenly the emphasis is on communication – unless the clubs can check in with the
stores, they will never know how many signups have taken place. Good feedback would also be
needed, so that teams are notified and can celebrate with every signup that occurs, as well as
weekly score updates to see if they matched their bet.
Would it work? Perhaps, but that is not the focus of this thought experiment (and
regardless, the evolution of a design is very hard to predict until it is concretely attempted).
Rather, the thought experiment helps to show the size of the gap between ParTour and a more
202

complete game. The gap is substantial between applying some elements of games to activities,
and going all the way. Crossing the gap requires more than a philosophical shift; in particular,
the feedback loops often require additional monitoring, and point to the need for either additional
staffing or new technology. Additional costs are more subtle: the game design and supporting
technology are often more hierarchical, serving as centralized systems that pull back from the
participatory design approaches used previously in ParTour.
A paradox of using games to structure participation is that they can shift power away
from informal community organizing and into a more rigid system of rules and feedback. The
financial costs can be substantial too. And there are cultural costs of shifting to more formalized
modes of participation, along with some ethical risks. Participatory design can be used to share
power in the process to some extent, but more research is needed to understand how such a
process could interface with the relatively top-down traditions of game design.
As mobile media grows, a new space for design is emerging between games and tools to
tell local stories. While the prior chapters pointed to how games can be more effective with
‘local fit’, the case of ParTour shows how participatory design can start with existing
storytelling activities, and find new opportunities with the methods of game design. Once again
the framework for ‘local fit’ can guide optimization to deepen the networking between residents,
local organizations and the ecology of local communication.

203

CHAPTER FIVE: CONCLUSIONS
Overview
Current theory on games is perhaps overly focused on digital spaces and the agency of
individuals. Local communities need more. For games to be used in campaigns for regional
growth and social justice, game studies must speak to building cohesion in local networks and
the community as a whole. In response, this dissertation proposes a framework to direct the
design of place-based games, especially for collective empowerment.
Most powerfully, the framework provides a conceptual tripod to break through the
complexity of embedding games in real-world communities. Specifically, the framework
reduces the challenge of fitting into a local community using three pragmatic dimensions:
network mechanics, group identity, and the ecology of local communication. With ‘local fit’,
game designers and evaluators can balance three cognitive perspectives and come much closer to
integrating the “how of games” with the “how of communities.”
This study is one of the first to specifically theorize games in the ecological terms of local
communities, and to propose methods for appropriate localization. ‘Local fit’ was specifically
investigated as an explanatory framework that targets what makes games distinct while
minimizing risks for local communities. The ecological approach helped to integrate analysis
across diverse levels, including individual behavior, group formation, organizational facilitation,
and the broader community identity.
To begin, this dissertation introduced “Locally Situated Games” as a new category of
game to orient analysis and improve design. In contrast to prior classifications of games in terms
of technology or play genre (such as “locative” or “role-play” games), the new category proposes
204

to group together games that seek to empower existing place-based communities. Joint language
is needed to define the category in a way that supports game design as a technique of building
community. Therefore the basic definition serves as a cognitive schema, linking games studies
to the ecological approach of community empowerment.
185
More formally:
Situated Games are defined as: game-based structures for real-world participation that
seek to strengthen a place-based community. This definition is introduced in the first chapter,
and subsequent chapters deepen the perspective using concrete examples. As a contribution to
the field, the definition provides a conceptual basis for comparison across case studies, and for
generalizing findings (i.e., to the now-specified category of games). Findings include design
guidelines – see the “design checklist” appendix.
Three primary cases were explored: Macon Money
186
, Reality Ends Here
187
, and
ParTour
188
. The goal was to triangulate across cases of Situated Games. Each case has
distinctive modes, strengths and weaknesses – as with any intervention into local communities

185
Exact language was validated as accessible through informal conversations with game designers and community
organizers; the theoretical language first emerged from literature reviews in game studies (especially, Fullerton et al.,
2004; Salen & Zimmerman, 2004) and the sociology of neighborhood communication (especially, Kim & Ball-
Rokeach, 2006b), alongside the sustained observations of games in this study.
186
Recall: Macon Money (2010) is an alternative currency game to encourage crossing lines of socio-economic
segregation. Low-tech paper ‘bonds’ served as physical game pieces, with online tracking. Funded prominently by
the Knight Foundation for nearly half a million dollars, the game points to how a game can serve as social policy to
connect strangers – to one another, and to local businesses. An evaluation found that residents met new people and
returned at high rates to local businesses.
187
Recall: Reality Ends Here (2011) is a game for students from a leading film school to form teams and make
multimedia, including 30-second short videos, visual effects, or longhand notes and movie scripts. Points went to
more complex projects, utilizing a collectible card system. For the film school that funded it, the game served as a
way to structure informal participation, and serve as a kind of institutional policy to foster small group formation
and build skills in making media in groups.
188
Recall: ParTour (2012) is participatory mapping activity in South Los Angeles that draws on game principles to
sustain engagement. Funded by tiny investments from participating nonprofit organizations, the activity shows how
lessons from games can provide a coalition strategy to build and rebrand their community as a neighborhood
that is safe, healthy and innovative.
205

with all the inherent complexities. Variance across cases is deliberate and valuable (e.g., not all
are strict games).
To optimize design for an established community, game designers need theory and
research-based frameworks to optimize their designs. As this study has shown, the language of
behavioral ‘incentives’ has only limited power to analyze or optimize real-world games for goals
of group empowerment. An incentive approach also carries overtones of manipulation that
pragmatically makes games unpalatable to many community partners.
The central premise of this study is that the success of a Situated Game may depend on
its “fit” to the local community. Conversely, poor fit may undermine the game – including
limiting participation, support from local organizations, and media coverage. Research was
guided by the umbrella question of how does “fit” matter for success?
A conceptual framework for ‘local fit’ was proposed in three dimensions, based on a
review of the literature and early analysis of the case studies. Pragmatically, the framework aims
to help future game designers to optimize their games for community empowerment, and to help
evaluators to investigate games in terms of local compatibility (i.e., good vs. poor fit).
Theoretically, the framework of ‘local fit’ seeks to embody the cognitive diversity that is
necessary to create innovative games (Fullerton et al., 2006), and to address the hard problems of
real-world communities with entrenched social issues (Page, 2008). Each dimensions represents
a different cognitive perspective on Situated Games, including the mechanics dimension (via
206

network formation and game studies)
189
, the group identity dimension (via symbolic interaction
and social movement studies
190
, and the communication ecology dimension (via the sociology of
neighborhood communication). To summarize from the first chapter, the three dimensions are:
1. Mechanics: Fit of the game mechanics to local community modes of meeting and
greeting, especially the actions that residents often take to find one another and to
grow the place-based community.
2. Group identity: Fit of the game’s performed narratives to the local narratives of
community identity and neighborhood brand.
3. Communication ecology: Fit of the communication in the game to the local ecology
of communication, especially through channels to local organizations and media.
Ensuring fit requires research as part of the game design process. Such research is often
time-intensive and may be costly too. In return, optimizing for fit can build partnerships and
increase the odds of success, but ensuring fit is only expected to be worth the trouble if
community empowerment is a high priority (a priority that is a prerequisite for Situated Games).
Data collected
Data on each case study in this dissertation was collected and analyzed with multiple
methods. Real-world games have effects that are complex, ecological and multi-level.

189
The “mechanics” dimension emphasizes interpersonal interaction, drawing on game mechanics in game studies.
The mechanics are oriented toward how social networks are primarily initiated, which is to say through acts of
greeting and basic meeting. At the least, deepening ties requires repeatedly meeting.
190
For collective identity, two perspectives are drawn on: for collective empowerment, the study of social
movements has usefully described the value of a group identity; at a more micro level, identity can be performed by
individuals on the stage of their everyday life (Goffman, 1990), as described in sociological perspective of symbolic
interactionism, which also draws heavily on social psychology.
207

Especially if tracked longitudinally, deep evaluation could easily cost several times the game
budget for development and operation. Therefore, each case emphasizes a particular set of
methods to probe one or two dimensions of ‘local fit’. Some of the data I collected directly for
this study; I also draw extensively on the work of collaborators, often re-analyzing others’
original data with their support. Much of the deep analysis was collective, and credit is due to
many collaborators – see again the Acknowledgements forward to this dissertation. Below is a
summary of the data sources and my role in data collection (additional details are in appendices
for each case):

(1) Macon Money: I advised the primary evaluation by Taylor and Whatley (2012a) in my role
on the Games Advisory Group to the Knight Foundation. For this study, I re-analyzed much
of their raw data, which included pre- and post-game surveys of players (n > 250), interviews
with 26 players, an online survey of the 40 participating businesses, and a database tracking
the 5021 bonds. I additionally conducted participant observation in Macon, Georgia at
community planning meetings for the College Hill Corridor approximately two years after
the game finished, and conducted follow-up interviews with one of the primary game
designers, the local Knight Foundation representative, and local staff for the game who
continue to work on community development. Primary sources additionally included the
“how to” kit for replicating the game that was published by the designers, social media
produced by players through participating in the game, website content from partners in the
game, and press stories on the game.

208

(2) Reality Ends Here and Xposure: My informal interviews with the designers began as the
game was first conceptualized and tested in the spring of 2011, and continued through
reflections on the third year in early 2014. In the first year of the game, a total of 119
projects were created by 103 individuals. To analyze the network relations of the players, I
scraped the game website using custom scripts, and used UCINet for statistics and
visualization tools. A similar analysis was completed for the second year, which included 80
students. To investigate the attempt to scale the game, a series of semi-structured interviews
were conducted at the second school during and after the launch. To compare the two
schools in terms of the communication preferences of their incoming students, an online
survey was distributed to incoming students at both schools (n>110 for each school).

(3) ParTour and Re:Activism: I conducted participant observation with ParTour for two years
beginning in 2011. The storytelling technology was first developed in the participatory
design process of Mobile Voices, and several members of the ParTour team had participated
in that process including me. Informal interviews were regularly conducted by various
members of the research team; often the interviews were videotaped and some were posted
online, consistent with the philosophies of action research that seek to foster co-generative
learning (Elden & Levin, 1991). A series of more formal semi-structured interviews were
conducted in 2012 with five organizational leaders, bicycle club leaders, nonprofit staff and
university researchers. For the study of Re:Activism, the lead designer was interviewed in
advance, and participant observation was conducted in two trials of the game in Atlanta in
209

the summer of 2013, each with more than 20 players; semi-structured interviews with the
game designers were conducted afterward.
Models for impact beyond the players
The deeper goal of Situated Games is to affect the broader community, beyond the
immediate players. Often the strict number of participants can seem small – a few thousand in
Macon Money, less than half the incoming students in Reality Ends Here, and dozens of
bicyclists in the mapping parade of ParTour. Impact is often better modeled in terms of
catalyzing networks and redirecting the community trajectory, rather than attempting to recruit
all residents as deep players. How can this be done? This study identifies several models to link
the core game with broader community impact; each points to a potential area for further
research:
(a) Game as group rallying call: the game is a public performance – like the media
spectacle of a parade – that rallies the community around a particular identity,
shared vision and common concerns. Success depends on a broader campaign,
where the game is just one of several aligned interventions – and the sum is
greater than the parts. Here the core players are like actors, with the stage facing
a larger public.
(b) Game as movement affiliation for organizations: like a buy local movement (or
even a revitalization coalition), local organizations join the game to show they are
a part of “something bigger.” The game can catalyze longer-term affiliations,
e.g., between shops and their customers as in buy local patterns that persist after
the game is over.
210

(c) Game as network catalyst for key individuals: the game provides a filter for future
leaders, identifying which people are willing to take risks by experimenting in
public and empowering them by legitimizing their role as bridge builders between
key constituencies. For example, multiple local leaders in ParTour have
subsequently received jobs in local government and community coalitions, and
several cite their ParTour involvement as giving them credibility and the network
visibility to be hired. Empowering these key network leaders may be worthwhile
for certain kinds of network change.
(d) Game as socio-economic policy: for example, see the “expansionary monetary
policy” of Macon Money (Sethi, 2013). Beyond core players, the effect was to
add liquidity to the local economy at large and direct financial flows toward
independent local business. Akin to tax breaks, the dollars transferred per
business can be modest, yet still boost important capacity or forestall a looming
bankruptcy. The game resists targeting all businesses, instead focusing
participation on specific sectors (e.g., independent retail outlets), or even on
specific stores that embody the community history and serve as anchors (e.g., for
restoration and anti-gentrification campaigns).
Essential variance in Situated Games
Variance across the case studies was substantial in order to robustly describe Situated
Games. For example, must the activity always be a ‘game’? How much of the action should be
real-world? Variance was particularly emphasized in terms of the basic criteria that define
Situated Games: real-world actions, network-building goals, and game-like activities.
211

Real-world actions as part of the activity. All case study activities met this criterion, in
the sense that making progress in the game was congruent with affecting the broader
world. In general, these actions must be observable. The variance is substantial:
o gain new customers and revenue for local independent business (Macon Money),
o introduce strangers across lines of socio-economic segregation, including town-
gown divides (Macon Money, ParTour)
o launch videos for global distribution that tell community narratives (Reality Ends
Here)
o distribute images that counter stereotypes, in local & mainstream media (ParTour)
o form interdisciplinary teams (Reality Ends Here)
o raise awareness about community history (ParTour, Re:Activism Atlanta)
o experience group protest (Re:Activism Atlanta)

Network-building goals tied to place-based communities and local cohesion. The big
picture vision for each game varied. Compared to the criterion of hard real-world action,
this network criterion is about intentions, especially the priorities of the designers and
operators. Distinguishing between game actions and goals proved especially valuable to
frame the actions modestly
191
while still providing leadership for larger and longer-term
visions of group empowerment. Again the variance is noteworthy:

191
Recall the Macon stakeholder who “had to keep saying over and over and over again, Macon Money is about
introductions… that’s all – that’s all it’s about. I think these games have way too many expectations on them.”
212

o community connectedness, more social capital (all)
o a deepener sense of place and belonging (all)
o momentum to “buy local” and sustain local business (Macon Money, ParTour)
o greater equity across race and class (Macon Money, ParTour)
o trust across town and gown, group relations (Macon Money, ParTour)
o deeper associational life (Reality Ends Here, ParTour)
o confidence in organizing small groups (Reality Ends Here)
o rebranded neighborhood / shifted collective identity (ParTour)
o collective narratives aligned across communication levels (Macon Money,
ParTour)
o collective efficacy (ParTour)
o stronger coalitions (Macon Money, ParTour)

Adapting a game can be harder (and sometimes impossible) if the networking goals are
low priorities. For example, when Reality Ends Here was introduced at the second
school, the goals seemed to narrow to emphasize the cultivation of individuals’
multimedia skills, with less prioritizing of the group (alumni network) identity and
network cohesion. One explanation for the failure at the second school is that the game
was not localized enough, and that a higher prioritization of network building goals could
have helped the designers to insist on more radical changes to adapt for local fit (e.g.,
around school-wide sporting events and specific social media). As a second example, a
lack of network-building goals can result in a shallow adaptation; for example, the
213

integration of ParTour for Re:Activism only happened in terms of the activist tool for
camera phones, rather than tapping ParTour’s deeper philosophies of group storytelling
that engages local organizations.

Game-based design that structures participation. Variance in gameness was a key
theoretical investigation for this study. As the opening chapter declared, game-based
may be a more appropriate than strict ‘games’ for real-world activities.
The need for flexibility can emerge within the game. For example, in Reality
Ends Here some players promised “no pointsing.” On their own, the players were
vowing not to sacrifice culture and artistic quality for the raw accumulation of points.
Simultaneously, other players clearly strove to maximize points; neither is inherently
right, and the flexibility to support both was valuable. Debates over the norms of
participation are vital in democratic life and can be healthy for real-world games.
192

Flexibility on “game-based” criteria, even to the extent of embracing the negotiation of
rules by the community, can leave important room for appropriation of the game by the
community.
“Game-based” can still be rigorous. One theoretical move is to articulate the
game using cultural and group-centric language, while still insisting on core game
principles. Consider the challenge of winners and losers. Strictly losing at pure games

192
Even when players do not explicitly debate nature of the game, some ambiguity is expected. Similar gray area
persists in the field of game studies itself, from categorical definitions to cultural debates over what counts as a
‘game’ (Salen & Zimmerman, 2004). In Macon Money, for example, some players saw it as a game; however, other
participants saw the rewards as raw incentives (not playful) and the outcomes as predictable (not uncertain).
214

can be acceptable (e.g., in poker), but problematic when tied to economic exchange and
racial divides in historically marginalized neighborhoods. Macon Money made it easy for
many people to succeed, shifting from declaring individual winners to framing victory as
a success for the neighborhood. Designers can thus consider how to declare the game as
a collective victory. Bonds in Macon Money were distributed unevenly by mail,
sacrificing fair play in the technical game but increasing fairness at the socio-economic
level by improving access for under-represented populations. This study recommends
against simply relaxing the notion of the ‘game’ to be a matter of convenience, instead
recommending a shift in game-based criteria to emphasize cultural constraints and
recognize opportunities for collective victory.
In sum, the criteria for “game-based” activities was validated and refined through
the consideration of all case studies. As the opening chapter defines in more detail,
game-based activities are those that structure participation through: (1) playful
challenges, (2) feedback loops and (3) uncertain outcomes.
For civic designers working on ordinary activities – not games – the value of the
game-based criteria is to provide a conceptual target for how they can improve. Many
civic designers are new to game theory and need a basic conceptual frame for
incorporating philosophies and practices of game design into their work – without
abandoning the established activity. Currently there is a dangerous profusion of slippery
215

terminology and concepts, a side-effect of the growing interest of cross-applying games
to other fields.
193

The utility of a basic framework for “game-based activities” can be seen in this
study most clearly in the analysis of ParTour, where the analysis first revealed how
existing activities may already be succeeding for reasons that are best articulated in game
language. Then the analysis showed how the principles could also be used to help
optimize iterative activities like distributing paper materials, in order to deepen the focus
on building network ties as part of the activity. Sustaining the activity often means
leaving the materials and participation loops open-ended; with the map, participants were
challenged to distribute the map by starting conversations that invite participation,
especially “What is missing from the map?”
Findings and discussion of fit
How can the framework of ‘local fit’ guide Situated Games? Each dimension of ‘local
fit’ provides a coherent and semi-independent perspective on how the activity functions. (The
dimensions are: network formation, collective identity, and local communication flows.)
Scaling to new sites: replication versus adaptation
The value of ‘local fit’ was perhaps most evident in differentiating success from failed
adaptation. In particular, the attempt to scale Reality Ends Here to a second school provided a

193
The slippage is especially noticeable in cross-applications of game theory, including in movements like
“gamification” that often imply “game elements” may be sufficient in isolation (e.g., points and badges); as of this
writing, for example, the highly visible Wikipedia page on gamification (https://en.wikipedia.org/wiki/Gamification)
is full of such ambiguity. See also the literature review in the first chapter.
216

rare point of direct comparison. The very practices that made Reality Ends Here an initial
success did not work with Xposure. Was more localization needed? Ironically the problem may
be an overly faithful replication of the game.
Funders and policy makers often demand that impact is scaled. While this dissertation
understands the importance of pushing for scale in many games, it can be dangerous with
Situated Games. Replication models may undermine the growth of Situated Games. Enshrining
a successful game as a “delicate” artistic triumph is dangerous; in the worst form, such elitism
leads to a preference for creating new games from scratch rather than adapting prior models.
Therefore, one contribution of this dissertation is to critique current scaling models for
place-based games, and to propose an alternative model based in ‘local fit’. The critique is
deliberately narrow, since traditional scaling makes sense for many current games (e.g.,
education games about the federal budget), but not all. The danger of traditional scaling models
is partly that their limitations are seen as weaknesses in the basic design, rather than
shortcomings of the adaptation approach.
The category of Situated Games provides a clear counter-example to replication. In
particular, place-based community building is a movement about local distinction, and thus
games that empower all communities in the same way would actually be flattening the sense of
place and undermining local attachment. Situated Games and their outcomes must vary by place.
At the least, each community can have a different narrative and identity; more deeply, regional
economies may specialize in different sectors (Blakely & Leigh, 2010), and require different
development strategies. To understand how localism works, the “defense of place” approach is
217

more than a political rallying point – it is also a frontier to build important theory (Escobar,
2001), including new models for scaling games through adaptation.
Two kinds of scaling emerged with Situated Games. The first kind addresses how impact
can be grown for a single game, especially to reach beyond players to the broader community;
this was discussed above, e.g., how a game can serve as a “rallying call” to the broader
community. The second kind of scaling concerns the diffusion of the model, where impact is
scaled to additional communities as they adopt a prior game for their own use.
Prior research on the scaling process has been studied for new technologies in terms of
the diffusion of innovation
194
; similarly, the diffusion of policy has been studied across cities
195
.
Neither is sufficient for Situated Games, because they assume that diffusion happens after
“innovation” is complete.
196

The critique implicit in Situated Games is that the innovation is ongoing, not finished
with the original game. The value of ‘local fit’ is to provide a concrete framework for
reasserting fit for each community adaptation, decentralizing the process of innovation. Of
course, research on appropriation has shown how local cultures have long adapted technologies
for their own use and empowerment (e.g., Bar et al., 2007). Yet for Situated Games, the
appropriation must go beyond crowds of individuals, to empower groups (and organizations) that
seek to import and deliberately redesign existing games.

194
The approach typically looks at the rate of adoption of new technologies (like cellphones) by users, derived from
Rogers’ 1962 description of the “Diffusion of innovations”; see (Rogers, 1995).
195
See, for example, the diffusion of antismoking policy in 675 of the largest U.S. cities (Shipan & Volden, 2008).
196
For a good discussion on the dangers of decoupling diffusion from innovation, see Dearing and Meyer (2006, p.
31).
218

Currently, adapting a game is often guided by a “toolkit.” After Macon Money the
designers released an “In Your Town” guide; after Reality Ends Here came “Reality: The
Manual.” Toolkits are one way to appease funders, and are often necessary to document and
understand the original game. Toolkits are certainly an important moment for reflection across
stakeholder groups.
This dissertation proposes that Situated Games are particularly in need of distinguishing
between toolkits for replication and adaptation. Both are valuable, but ‘local fit’ can only be
possible when the approach emphasizes adaptation. (The term ‘adoption’ is not used because
Situated Games are not simple or fixed technologies for adoption; even the replication of
Situated Games requires substantial content creation.)
A design commitment to sharing means optimizing the model for more than replication.
Understandably, most toolkits currently emphasize replication – since describing the current
game is a more straightforward act of interpolation, rather than the extrapolation needed to guess
where the game will need to be flexible. The adaptation approach proposes that ‘local fit’ must
be re-established for any new community, as shown in Figure 41. ‘Localization’ is adaptation
focused on a particular locality, and is especially important for Situated Games.

219

Figure 41: Adaptation model for scaling using dimensions of ‘local fit’
to (1) optimize localization/fit and (2) compare contexts

‘Local fit’ seeks to limit the burden of investigating localization for every game, and to
help avoid blaming local staff implicitly when localization fails. In a failure to scale, the
breakdown often happens in multiple points, across levels of analysis and sites of observation.
Too often, comparisons between sites are made purely in terms of demographics or leadership.
The investigation structured by ‘local fit’ helps to compare local contexts. Specifically,
each dimension points to a way to measure context shifts: Does the new context have (a)
different local modes of greeting and meeting? (b) a different kind of local identity or group
story? (c) different local modes and channels of communication? Such questions can improve
localization by helping understand the adaptation process, prioritizing how Situated Games work
as depicted in Figure 41.
Solving problems of fit is non-trivial, akin to a puzzle with complex edges. Yet the
framework will only be successful if it can concisely assess situational differences. In
Game
Community
Fit
Context comparison:
(a) Social modes for network growth?
(b) Community identity and narrative?
(c) Ecology for multi-level communication?
Game2
Community #2
Adaptation
Re-Fit
(3 dimensions)
220

diagnosing the failure to replicate Reality Ends Here, all three dimensions of ‘local fit’ pointed to
differences between the two communities. Summarizing the differences was straightforward
with ‘local fit’, as seen in Chapter Two.
197

Context questions are emphasized for Situated Games, not universally. In Re:Activism,
by contrast, the localization was only at the level of content because the goals were not focused
on collective empowerment. There is nothing wrong with games like Re:Activism that
emphasize individual transformation and education; in fact, staying focused on such goals can
limit the risk of failure. Thus the fact that Re:Activism is excluded from the category of Situated
Games (on the basis of its lack of network-building goals) shows the importance of Situated
Games as a limiting filter.
The scaling of ‘local fit’ goes beyond the “participatory scaling” that has been used with
many technology projects. Perhaps most importantly, ‘local fit’ adds the layer of community
organizations and the media to prior research on “involving stakeholders” in the design process.
For example, ‘local fit’ serves as a friendly amendment to prior research on “participatory
scaling” in educational games (Martin, Dikkers, Squire, & Gagnon, 2014) that encourages local
educators to adapt place-based games for their own uses.
Equity is a final reason to emphasize adaptation. A core goal of the ‘local fit’ framework
was to accommodate marginalized and underserved communities. Requiring elite designers is a

197
For details, see the “Summary of local fit opportunities” in Chapter Three. To show how each lens is distinct for
Xposure: (1) the lens of mechanics pointed to social greeting in social media tools like Instagram – which were
much more heavily used at the second school; (2) the lens of group identity suggested a shift from entirely
underground to visibility in academics (pre-professional) at socially (like the school-branded athletic events and
Greek life parties); (3) finally, the ecological lens of communication channels pointed to the lack of shared housing
as a missing catalyst, and the potential to compensate by making the game more visible in institutional channels and
gathering spaces.
221

particular burden in traditionally marginalized communities. While fit is often necessary for the
original design, subsequent scaling projects have often reduced budgets, and rarely are there as
many (if any) elite game designers. Therefore, this study investigated non-game activities (i.e.,
ParTour), seeking overlap toward more accessible language and bottom-up tools to diagnose fit.
The bottom-up approach can be understood, in the extreme, as a kind of fitting where the
designers act more as observers and mirrors, and the interventions are minimal. Predictably, few
insights emerge in terms of mechanics, which are better left to games and techniques like
usability testing
198
. However, insights did come in the dimensions of group identity and the
ecology of communication, including the benefits of asset mapping.
Analytic power of ‘local fit’
A big picture of ‘local fit’ emerges by looking across the case studies, as shown in Table
4. The story of Situated Games is thus indexed, particularly using the “strong” examples of fit as
a roadmap. The “strongly” rating was used to indicate where the empirical evidence appears
good, and more importantly, where the analysis of this study dove deeply to highlight a specific
kind of fit. To be clear, there is significant gray area in making such summative evaluations,
which are moving targets as the games themselves continue to evolve. At the same time, the
three-pronged approach to ‘local fit’ balances the evaluation, with its roots in three separate
disciplinary perspectives.

198
See, for example, the vibrant professional field of Games User Research that has rapidly emerged in the games
industry in the past five years. Major commercial game studios now have in-house usability labs and teams of
assessment professionals to optimize their games. Drawing on methods of human-computer interaction, social
psychology and the stealth analysis of aggregate use data, the field is often ahead of academic methods for
diagnosing games and providing actionable feedback to inform design.
222

Table 4: Summary of game cases in terms of ‘local fit’
Qualifies as a
“Situated Game”?
(1) Mechanics fit to
the existing social
modes
(2) Fits to local
identity
narratives that
stage participation
(3) Communication
channels of the
game fit to local
ecology
Macon Money Yes, although the
game has only thin
uncertainty
Strongly Yes Somewhat
Reality Ends Here Yes Yes Strongly Strongly
…comparison:
Xposure (a second school
applies Reality Ends Here)
Yes, although
network building
goals are weaker
Somewhat No Somewhat
ParTour in South LA No, the activity is
missing key game
elements
Yes, for meeting
within group rides
and for social
media picture
sharing
Strongly Strongly
…comparison:
Re:Activism Atlanta
(using ParTour/Vojo tools)
No, the activity is
missing network
building goals
No No Somewhat; only on
an interpersonal
level

The mapping in ParTour served a dual purpose for this dissertation. First, asset mapping
was introduced as a methodology to identify some of the communication ecology; second,
mapping was simultaneously a community activity. Rarely are activities for engagement
simultaneously considered for data collection. As an activity, ParTour provided engagement
around group reflection and network diagnosis. Implicitly, the mapping activity asked the group:
“What are local strengths for us, and what do they mean to us?”
ParTour may seem extreme. Certainly ParTour goes beyond what most games will do to
scaffold reflexive storytelling in their community. However, if the goal is to empower the
community then the community may need to diagnose its own limitations and assets, and build
223

support for responsive action. As a side benefit, game-based principles can also be extended to
improve ParTour, as was shown in the shift to turn map distribution into an open challenge and
group conversation.
A contrasting methodology is network analysis, which was proposed as being especially
relevant for Situated Games, and was demonstrated in the case study of Reality Ends Here.
Although network methods are rarely presented as bottom-up techniques, this study points to
how network visualizations may provide a vital interface for revealing the network to itself. For
designers, the Reality Ends Here case demonstrated the formative value of network analysis in
optimizing games. For staff on the ground, network analysis was shown to provide insights that
could inform community organizing in real-time.
Findings also emerged for each dimension of ‘local fit’, as summarized below.
Mechanics that fit to local social modes (fit #1)
The first dimension of ‘local fit’ evaluates game mechanics, which structure the repeated
actions of participants in the game. Mechanics are at the heart of how games work, and how
game designers optimize the feedback loops involved. In order to build networks, the game
mechanics must support social meeting; ‘local fit’ proposes aligning with established social
modes. (For theoretical detail, see the more detailed discussion in Chapter Two.)
Concretely, games can strengthen networks in many ways. As illustrated across the case
studies, the possibilities include: (1) new ties; (2) deepening ties; (3) bridging groups; (4)
positioning key people, like influencers or bridgers; and (5) fostering diversity within teams.
Knowing which kind of tie to foster is important, and should be an explicit part of planning a
224

Situated Game. Additional research is needed to move past fitting the existing community to
consider prioritizing ties that might change the network for strategic community goals.
One discovery is that the traditional phrasing of professional game designers may need to
be broadened slightly. To better address real-world games, designers may need to articulate
mechanics more culturally than in raw formalist rules and immediate procedures. In games, play
emerges in the space between formal rules and culture (Salen & Zimmerman, 2004, p. 104); for
real-world games, the emphasis must shift toward the cultural end of the spectrum. The cases in
this study showed that even top-tier game designers often revert to technical language of literal
rules that obscures the social action (e.g., “match icons” in Macon Money, or “procedural
prompting” in Reality Ends Here). Such language is accurate, but should conceptually defer to a
more cultural frame for the primary mechanics – especially when the goal is building a shared
vocabulary with traditional community builders.
To articulate the mechanic with a more cultural lens, this dissertation introduces a simple
two-step process (see Table 5). For game designers, this approach is akin to articulating
mechanics using language of the “experience goals” (Fullerton et al., 2004) with a focus on
social interaction. The goal of re-articulating the mechanic is initially for analysis
199
, yet the
articulation will also help with design. A final articulation is not easy, but this method helps
produce joint language for “how the game works” that is half game design and half the strategic
planning language used by civic organizations.

199
For example, iteration emphasized by mechanics helped to set aside annual events like the Soap Box Derby that
were good for the community identity but less optimized to meet and mix with new people.
225

Table 5: Process for articulating mechanics with a socio-cultural lens
(1) First, describe the game in a sentence that explains what players primarily do in
the activity, using simple verbs. (Each primary verb must be at the core of how
players make progress in the game, especially through repeated action.)
(2) Second, rephrase to emphasize the verbs that will build the network, especially
the social actions of greeting and meeting and how the experience will feel to
players. The articulation should be consistent with the designers’ primary theory
of change for the game, i.e., their belief about how social actions will in fact lead
to network ties and group cohesion. The verbs will often match the game’s real-
world actions, with the more visionary “network-building goals” staying in the
background.

How did designs obtain good ‘local fit’ in terms of mechanics? Some of the most
promising mechanics showed mutual reinforcement with existing social modes, beyond
compatibility. The games did not “spark” passion as much as tap into it. Passions included
creating media in Reality Ends Here, bike riding in ParTour, and meeting friends at concerts in
Macon Money. Often, the digital structures of the game helped make existing passions more
visible and provided continuity after the physical events were over. The magic circle of the
game gave participants an excuse to overcome the social awkwardness of introductions; still, the
game depended on pre-existing social interest.
The downside of fit is that the game can also inherit negative social forces. Tightly
fitting to currency-based greeting in Macon Money was linked to seeing the game as a hoax,
especially by African American participants, possibly leading to a suppression of participation.
226

Although game structures can compensate (e.g., by over-sampling to under-represented groups),
without care the mechanics can aggravate existing stereotypes and divides. Real-world games
may need distinct tools to address inequality, and this study proposes that reflecting on
mechanics and inherited social modes is an important first step.
Nearly every game had more than one primary mechanic. One reason is that the official
game often hid a secondary mode of play. For example, in Macon Money the game played by
businesses had different tactics (like hosting events) than the game played by citizens. The
analytic act of articulating mechanics often forced analysis to acknowledge how the game was
working on multiple levels, and for multiple audiences. Juggling such disparate core
descriptions of the activity can initially feel awkward, but ultimately produces a more ecological
picture of the activity.
An ironic consequence of good fit is that the activity spills over, with more positive
action than the formal system can easily track. For example, with Macon Money the example of
Justine showed how a single “match” seen by the game actually involved five instances of social
pairing and greeting different groups of people. Similarly, the pictures of community assets and
the group ride in South LA were often reposted beyond ParTour, including to friend groups in
social media that were not tracked by the activity. The focus of mechanics is primarily on the
choices of individual players, which underscores the need for additional dimensions of fit to
capture the group-level dynamics.
Collective identity and neighborhood brand (fit #2)
Games in public space enact a kind of visible performance of community that highlights
what the space means, and what the participants are about. The risk is that organizations and
227

residents are both likely to reject activities that invoke the community identity without matching
“who we are.” This second dimension considered how the game’s performed narratives fit to the
local community identity and neighborhood brand.
Certain narratives are more important than others. One contribution of this research to
game studies is to underscore the importance of narratives that are specifically tuned to the
place-based community, echoing the attention on “neighborhood stories” in the sociology of
neighborhood communication (Kim & Ball-Rokeach, 2006b). Stories of collective efficacy and
cross-group belonging are likely to be particularly valuable in building civic engagement (Kim &
Ball-Rokeach, 2006a). Aggregating community stories to speak of collective identity is a
perspective from movement studies which is applied to the community-level of empowerment
for collective action
200
.
Quantitative measures of community were not explicitly investigated in this study, but
could easily complement it. For example, there are multiple scales of the “sense of community,”
which often correlate with participation in associational life, trust in government, psychological
empowerment and even life satisfaction.
201
Technology interventions, for example, have been
tracked in terms of whether they help build a sense of community (e.g., Du, Rosson, Carroll, &
Ganoe, 2009), and the same could be done with games. However, the measures are limited, and

200
Not all Locally Situated Games need to seek collective action. More fundamentally, the capacity for collective
action is an indicator of collective strength – which is a broad goal. The premise is that strong communities can
self-organize to address issues of public concern.
201
One scale for the “sense of community” (Prezza, Pacilli, Barbaranelli, & Zampatti, 2009) has dimensions of
“Membership, Shared Influence, Social Climate and Bonds, Help in Case of Need, and Needs Fulfillment”; another
(Peterson, Speer, & McMillan, 2008) has dimensions of “needs fulfillment, group membership, influence, and
shared emotional connection.”
228

often quite particular.
202
In order to lay the conceptual groundwork, this study prioritized a
qualitative approach to capture a multi-faceted sense of community identity in lived experience,
such as in connection to landmarks and associations.
Landmarks have particular power in place-based identity. Activities with good fit often
allowed players to perform their connection to landmarks, from the “Bullpen” of Reality Ends
Here to the Watts Towers in ParTour. Such performances echo Goffman’s (1990) performance
of everyday life, a design philosophy used in many locative games (Flanagan, 2009, Chapter 6);
Watson calls it the “dramaturgy of the local” (2012, p. 54).
Yet the greater goal was to transcend individual landmarks, and perform the place-based
identity at a larger scale. Perhaps the best example comes from Macon Money, and the identity
of being a “corridor business”:

“If you ask any of our local businesses, Macon Money was a big part of being
recognized as a ‘corridor business’ – not just being a storefront, but something that was
a bigger part of the community. You became a local business that was part of
neighborhood revitalization.”

Fitting to the collective identity also raises the stakes. A safe path was chosen by
Re:Activism Atlanta, focusing on protest in terms of human rights more than collective identity,

202
For example, “sense of community” is typically oriented toward residential notions of community at the block-
level, such as asking residents about their attitudes toward neighbors. Moreover, the constructs work on the premise
that “community” is a general attitude, and ignore the possibility of empowering sub-communities, or of having
residents feel a sense of community with “the East Side” but not my neighbors.
229

and avoiding any strong statements about “who we are.” However without goals of collective
empowerment, the activity does not actually qualify as a Situated Game. Thus the need for
fitting to the collective identity only emerges as the game takes on network-building goals, again
validating the framework.
Local ecology of communication (fit #3)
Communication is often the primary means of forming network ties, and communication
flow is essential for weaving together the stories describe above. Yet very few games have
explicit strategies for aligning with the local ecology.
203
Prior work on geo-ethnic media has
described how interventions must often fit to both geography and ethnicity (Kim et al., 2006);
this study goes further by considering the special needs when the fit concerns games.
To begin, this study points to a premature focus on communication technology as a
common and insufficient approach to the ecology. Technology assessments often ask residents
about their phones and plans, attempting to understand latent constructs like technology
‘readiness’ (e.g., Parasuraman, 2000). While technology is part of the communication
environment, an exclusive focus risks missing the most important social modes of
communication.
Language to describe ecologies is often inconsistent or incomplete. Frameworks like the
“environmental design” used by Reality Ends Here provide a starting point (Watson, 2012), but

203
Research on neighborhood communication has revealed hidden constraints, especially when looking beyond the
individual level (Ball-Rokeach et al., 2012). For example, some communities rely heavily on organizations like
churches, while others rely on Chinese-language newspapers, and others use online social media. The costs of low
alignment are subtle, more fizz than pop – such as low participation. In general, the risks seen in this study concern
undermining the basic goal of increasing cohesion and group actualization.
230

serve more as philosophical alignment than actionable strategy. Designers and organizers seem
to particularly struggle to recognize the multiple levels of the ecology. (One Macon Money
staffer seemed to have a good grasp of the need but was quite self-conscious, describing the
game as “a multi-tiered connection device… it’s weird to say it… but it was!”)
This dissertation proposes language to help make the ecological perspective
straightforward and actionable, drawing on communication infrastructure theory (Kim & Ball-
Rokeach, 2006b) to differentiate three levels for optimizing channels:
(d) Channels between participants, where talk is direct (including physical and digital
spaces), emphasizing actual use patterns, not simply access to technology;
(e) Channels to community organizations, where participants hear from local groups
that are shaping the collective conversation, such as local churches, libraries,
health centers, schools, sports leagues, and business clubs;
(f) Channels facilitated by local media, especially where there is some broadcast
power and legitimacy, as with local TV news, group pages on social media (e.g.,
Facebook), and geo-ethnic media. Local media occasionally bridges to national
broadcast media.

The ecological perspective also pointed to dependencies outside the formal game. For
example, local leadership was repeatedly shown to play a critical role. In Macon Money, the
evaluation found that success depended on trusted human leadership, and differentiated the game
from a less successful effort by the same designers (M. Taylor & Whatley, 2012a). In Reality
231

Ends Here, interviews showed that game facilitators were essential – and that their role was
unique at the institution, a cross between a teacher and a community organizer.
What methods can help optimize fit with the communication ecology? Introducing the
game slowly and closely monitoring it across levels was important, in recognition that it is
“difficult to playtest this type and scale of project outside of the community it was designed for,”
according to one of the lead designers. Additionally, the perspective of communication asset
mapping offers one concrete strategy, as described above and in Chapter Four.
Risks, knowledge gaps, and open questions
Ethics and empowerment
Questions of manipulation go hand-in-hand with empowerment. Both focus on agency,
and the authenticity of choices that are structured. The tension is implicit with real-world games,
which are powerful precisely because they structure playful participation.
In fact, games often cultivate an exaggerated sense of agency.
204
Games are designed to
be experienced, with heightened feedback loops around choices that will make a difference in the
game. Perceptions of agency can be empowering (e.g., efficacy beliefs are linked to many
positive social outcomes), unless that agency is misleading.
Manipulation is probably the most stringent rallying call for an ethics review, with all the
associations of hidden agendas, deception, and preying on weaknesses in human psychology.

204
Agency has long been at the center of game studies; more recently, media-focused analysis of convergence
culture (H. Jenkins, 2006) have pointed to how digital and immersive games may be particularly good at generating
a new sense of social agency in players (Mcgonigal, 2003).
232

For commercial uses of real-world games (e.g., gamification) there are particularly substantial
concerns, and formal requirements might be the most important first step. Yet for civic games,
manipulation is in many ways too low a bar, all-too-easy to dismiss by foundations and
nonprofits who genuinely believe they are contributing to the greater good.
Blunt ethical frameworks would have done little good in analyzing Macon Money. The
obvious manipulation concerns were fairly easy to dismiss. While there was some alignment
with the token economies used to condition mental patients in the 1960s (Glynn, 1990), in
Macon Money the goal of supporting local businesses was transparent, widely supported by
players, and relatively easy to track. Additionally, the legal positioning by Knight and others in
response to state gambling laws shows that existing regulatory frameworks are having some
effect, especially with real-world currency.
Of course, open play always risks transgression and cannot be fully predicted or
contained. Research into the ethics of real-world games with digital ties is just beginning in
earnest (Flanagan, Belman, Nissenbaum, & Diamond, 2007; Montola et al., 2009; Schrier &
Gibson, 2010, Chapter 10). Beyond manipulation, concerns arise for the inadvertent
participation of non-players, the obfuscation of consequences, and how the “magic circle” of
games can embolden players to ignore real-world laws and violate social norms. Additional
research will continue to inform Locally Situated Games, however their spirit of community
good will and empowerment inoculates them against many ethical risks. In the meantime, what
contributions can this study make?
To inoculate against manipulation in more thorough ways, the more ambitious goal of
empowerment is useful. Empowerment can be seen as a higher bar, beyond than threshold
233

manipulation criteria and more proactive. Yet empowerment is ambiguous; while nearly
universally embraced as a good thing, empowerment (and disempowerment) can be approached
with multiple theories and frameworks, from personal efficacy to race relations and legal
structures like human rights.
One approach is to embrace a multi-level view of empowerment. As a guide and point of
departure, consider the broad work on “empowerment types” by Ibrahim and Alkire (2007),
which emerged from a review of work on international human development. Four types of
empowerment are proposed, drawing on the notion of empowerment as an expansion of
agency
205
; below, I briefly link the three most relevant
206
to Situated Games:
• Empowerment as personal control (pre-requisite). If others are making your
daily decisions, then you lack basic control. At times, games have been
celebrated for giving individuals something they do have control over. Yet for
real-world games, this level is most useful to underscore a limitation of Situated
Games: even signing up to play requires a certain basic level of agency and
empowerment. In Macon Money, for example, deeply isolated and marginalized
people lack the power to even sign up for the game. Such basic access questions
also align with technology participation divides. Game designers need to do more
to acknowledge who is not empowered to play.

205
The authors cite Amartya Sen and Rowlands’ typology as particularly influential in forming their four
empowerment types.
206
Four possible “exercises of agency” are proposed by the authors. Empowerment comes from increases in each:
choice, control, change and communal belonging (Ibrahim & Alkire, 2007, p. 388). Of these, I have excluded
personal change, which concerns whether individuals believe they can change personally, since it is most outside
Locally Situated Games. More broadly, the framework of Ibrahim and Alkire is applied in my analysis only loosely,
since my goal is not to use their specific indicators so much as their underlying conceptual framework to help
organize a reflection on empowerment with Locally Situated Games.
234

• Empowerment as choice within a group: especially, “who decides?” In each
social context, from a residential household to a real-world game, power relations
structure who makes which decisions. Games are fundamentally about
structuring decisions, yet few frameworks address how games inherit external
power relations. For example, games can benefit from incorporating technology
design principles of social translucence, which can increase equity by simply
giving feedback on who is participating (Erickson & Kellogg, 2000). Similarly,
this study points to the methodology of network analysis and visualization as an
important opportunity for formative feedback. As shown in Chapter Two, game
designers can address homophily by gender, discipline and ethnic group by
structuring the game to address group formation.
• Empowerment of joining: the communal power of being with a group. When
the sum is greater than the parts, joining a group is a kind of empowerment. In
Reality Ends Here, this kind of empowerment was seen in the formation of media
making teams, and feedback loops that demonstrate to the groups what they
accomplish collectively. As an important caveat, if the group has unequal power
relations, then joining can actually be disempowering for some people. This
study proposes that game designers can benefit from considering bottom-up
approaches from community organizing, such as the balancing that happened in
the ParTour coalition between informal bike clubs, official nonprofits, and
university groups.

235

Of course, the empowerment list above is largely focused on the individual and micro level of
analysis, to match concerns over manipulation. The community-level is often better aligned with
Situated Games, and is explored more below in terms of collective empowerment.
One implication of ‘local fit’ is that each dimension can be optimized for ethics, often
with game-specific strategies. Several contributions to the literature are summarized below:
1. Meaningful choices on social impact (for fit#1)
This study points to a shift in the approach to meaningful choice
207
for games, showing
how ethical transparency can be part of good design. For a real-world game, this study proposes
to analyze the choices with a cultural and civic approach to the mechanics. For example, if the
mechanic in Macon Money were defined simply as “matching symbols,” then the challenge lacks
cultural depth, and the choices lose much of their meaning.
One finding is that real-world games may need to incorporate more feedback loops on
social impact. As described in Chapter Two, good games empower players to know “if an action
had an outcome.” Players of real-world games are often motivated by real-world outcomes, such
as participating in the “buy local” campaign of Macon Money. Yet game designers rarely give
feedback to participants on longer-term cause and effect. Doing so is understandably hard,
requiring tracking impact over longer timescales. But there is can be low-hanging fruit that is
currently missed; for example, the funded evaluation of Macon Money was not emailed to

207
Meaningful choices are at the core of good games. The principle of “meaningful choice” in game design
emphasizes designing for choices that are interesting to players, not simply offering more choices (as described with
game mechanics in the first dimension of ‘local fit’). Meaningful choices are satisfying because they have an impact
on the game’s challenges. Cultivating a sense of fairness is implicit; players must believe that the right choices will
pay off. Such principles go far back in the history of game studies (e.g., Huizinga, 1938).
236

players or posted on the game website. From a cultural perspective, failing to report on impact
to players undermines the game design, beyond being insufficient transparency for civic
activities.
2. Ethical spectacle (for fit#2)
Performing group identity in the theater of public space is part of what distinguishes
Situated Games. Cultivating spectacle is a core strategy for the bike parades of ParTour in
South LA, and in the beautiful bills of Macon Money exchanged at concerts, and in the staged
clandestine clues of Reality Ends Here. Spectacle creation often involves media making, which
was one of the inspirations for the protest theater of Re:Activism to adopt activist media tools. In
real-world games, spectacle often invokes media ritual
208
, for better and worse.
Spectacle has often been critiqued in Marxist and enlightenment traditions as dangerous.
Mass media in particular has controversially been accused of “manufacturing consent” (Herman
& Chomsky, 2008). Even earlier accusations called the labor process itself a “game” that
generates an “acquiescence in the social relations of control inscribed in the labor process”
(Burawoy, 1979, p. 82). Yet increasingly, media scholars have recognized that spectacle can be
used to confront established power, including to summon moral ideas of the ‘social good’
(Cottle, 2006).
One of the most powerful provocations of “ethical spectacle” comes from Stephen
Duncombe (2007). Duncombe acknowledges that most political and media spectacle may be

208
Media ritual has long been described as having the potential to draw people together in commonality by the
“creation, representation, and celebration of shared even if illusory beliefs” (Carey, 2008, p. 33).
237

hiding the underlying power structures. Yet he powerfully argues that the antidote must involve
some spectacle, since a rational thought model of democracy is inherently only half the picture.
To articulate vision and build political will, Duncombe argues that a healthy democracy actually
needs spectacle, but only if it is ethical. The “ethical” spectacle is described as being
participatory, inviting ordinary people to contribute and change the event, and never trying to
hide the truth.
209
Such principles were incorporated into Reality Ends Here, which Watson
(2012) describes explicitly as “participatory spectacle.”
The importance of ethical spectacle to Situated Games is underscored by ‘local fit’.
Specifically, ethics are implicit in fitting the game’s performed narratives to the local community
identity and neighborhood brand (i.e., the second dimension). For example, in Macon Money
the game aligned with the identity of the College Hill Corridor, which emerged through a
significant process of gathering community input, and which often stayed open and participatory
by asking residents publicly to state what the Corridor means to them. However, when
neighborhood brands become pure marketing that does not reflect public opinion (let alone a
plurality of viewpoints), then real-world games risk reinforcing an unethical spectacle.
Therefore, ‘local fit’ is not sufficient to ensure ethical spectacle; designers must go further,
reflecting on preliminary principles like those of Duncombe, and seeking to ensure their designs
allow ordinary people to critique and modify the spectacle as they perform it.

209
More specifically, Duncombe (2007, p. 170) offers a five-part criteria for ethical spectacle: (1) when participants
contribute actively of their own will, rather than as stage props; (2) when the spectacle is open to a plurality of
perspectives, rather than insisting on a single perspective; (3) when the spectacle encourages transparency; (4) when
the spectacle seeks to get at some real empirical truth, not simply tell a story; and (5) when the story is meant to be
read as a vision, not a literal statement of current tradition.
238

3. Ecological accountability (for fit#3)
Striving for ‘local fit’ may inherently minimize some ethics risks, and align with greater
empowerment. In particular, fitting to the local ecology means inviting the game into an existing
network of checks and balances, especially in terms of neighborhood organizations, local leaders
and community media. The challenge is that few local organizations have experience in looking
beyond rhetoric to the deeper mechanics of games.
Rhetoric is the bare minimum. In the case of ParTour, the tradition of negative
storytelling in South LA was a premise for the activity and broader community organizing. Yet
for games like Macon Money, the game needed more than linguistic compatibility with the
College Hill Alliance. The most challenging ethics and empowerment questions with Macon
Money were subtle, like how the game sought race-neutral public spaces, or taped into specific
church audiences, or even how the game was considered a hoax disproportionately by certain
groups. Power and oppression are often especially inscrutable and intractable in lived activity.
Deeper alignment requires transparency beyond rhetoric. Yet transparency in game terms
is difficult. Ideally, community partners will grasp the underlying forces of the game as an
activity system, understanding the mechanics, etc. When that is not possible, game designers can
at least describe the meaningful tensions behind choices in the game. As Squire (2002) has
observed, “characterizing the tensions of an activity system can help participants understand and
react.” One goal of ‘local fit’ is to provide a shared language for community partners to talk
about games with designers, and have mutual conversations that can consider power dynamics
and empowerment opportunities in the structures of a game.
239

One limitation of accountability via ‘local fit’ is that it presumes the basic existence of a
local civil society and group identity. In other words, fit can only happen if the community side
has some basic coherence, independence and freedom from corruption. Ironically, the very
communities with weak civil society may simultaneously be those who most need games to build
cohesion, even as their low cohesion makes fit difficult and impact less likely. (This is a poor-
get-poorer bias for place-based interventions in general.) Additional research is needed to
understand the minimum conditions for depending on the local ecology for oversight.
Aggravating the participation gap?
Even as real-world games address some divides, they may amplify others. Anticipating
some trade-offs can help design interventions with more wisdom. In particular, what about non-
participants? Those who chose not to participate (or simply cannot) may be excluded from the
network benefits, and may even be alienated from the network. When marginalized groups stay
away from Macon Money, the cash infusion will likely benefit already-privileged groups. Will
the alumni network be less accessible and valuable for students who choose not to join Reality
Ends Here, despite paying the same tuition?
Participation gaps (Henry Jenkins et al., 2007) can align with structural inequalities of
class and ethnicity. Even the possibility of worsening inequality underscores the importance of
careful facilitation in real-world games. Facilitation was emphasized in Reality Ends Here and
Macon Money, from monitoring the emerging game, to special outreach for under-represented
groups. Human facilitation seems essential in every example considered in this study.
Further research is needed to investigate how network-building games in particular align
with known participation gaps. Prior studies, for example, have shown that role-play games are
240

associated with people already more motivated to be civically engaged (Dalisay, Kushin,
Yamamoto, Liu, & Skalski, 2014). In Macon Money there was a “hardcore” group of
exceptionally active players; interviews indicated they all had a common thread: passionate
players were already community-minded and tied to the area. Socially isolated individuals seem
much less likely to receive the benefits of real-world games; understanding this gap, and whether
it is worse than for non-game activities, deserves attention.
Participatory design and bottom-up games
Perhaps the most aggressive design strategy for sharing power is participatory design. In
this study, the attempt to share power in the design process came in the Vojo tools that were
inherited by ParTour. However, much as a good movie can be ruined by too many producers, it
is not obvious how “participatory game design” could actually work. Current models for
participatory game design are limited to involving a few players as co-designers, or else
considering how players adapt large digital games by modifying or “modding” the basic game to
accommodate their own goals; this study proposes several ways we might go further.
In participatory design, basic consultation is the low bar, and is nearly universally
endorsed. In Macon Money, for example, the professional design firm proudly declared that
“because we involved our stakeholders from the research phase through implementation it was a
healthy process.” But involving stakeholders is not the same as allowing them to do the actual
design, or even giving them veto power over features.
Some research has explored delegating some design to players. The research has focused
primarily on large games with fan communities (e.g., T. L. Taylor, 2006), including the practice
of “modding” digital games to customize them. For place-based games, research on
241

participatory design has focused on empowering individual learners (Martin et al., 2014). One
contribution of this study is to amend such research to look beyond individuals, especially to
recognize local organizations as key stakeholders who have the staff and aligned resources to
reflect quickly and often deeply on the design and sustain the changes it seeks. If an exception is
required to allow for participatory design with games, then that exception is perhaps most
justified – and perhaps demanded – for Situated Games.
In the short term, scaling is a kind of design that may be more open to participatory
approaches. Prior research on technology appropriation hints at some of the possibilities (e.g.,
Schiavo, Rodríguez, & Vera, 2013). Of course, games have particular ways of working; the
frameworks of this study can help guide the appropriation process to be able to address game-
specific principles like mechanics and meaningful choice. Additional work is needed to help
leading designers go beyond offering toolkits and recipe books that primarily support replication,
to help them embrace scale as a kind of design, and participatory methods as a way to design for
better fit.
Moment in time: on the cusp of real-world games
Social play, from hopscotch to bowling, has always been important in local communities.
Now a new opportunity is emerging for real-world games to weave physical space with digital
flows. After decades of narrow digital innovation
210
, media-based games are broadening to

210
For 35 years, the digital game industry has served an entrenched status quo and a minority of the population, in
what some scholars have called a “hegemony of play” (Fron et al., 2007). Early digital games were claimed
temporarily by boys, but the pendulum has started to swing back toward women and families. A telling example is
The Sims, a casual life simulation that has often been compared to a dollhouse game. More than 150 million copies
have been sold worldwide, and women have played a prominent role in the game’s success. There is not gender
242

living rooms and local streets. This time
211
, games might actually deepen the sense of place. For
games to empower local communities, several barriers deserve mention.
European policymakers have begun looking at digital games to address “social exclusion,
inclusion and empowerment,” with the European Union releasing a first significant report in
2013 (Stewart et al., 2013). Other countries will likely soon follow. Such policy reports begin to
address the intersection of market forces, game-making tools, and government policy. Yet the
focus is resolutely on targeting individuals, whether for education, behavior change or civic
engagement. Notions of collective empowerment and networks are almost entirely absent.
Why the oversight? The problem may be in the dominant frame of “games as products.”
By contrast, the games in this study are not available for download or purchase. In fact, they
typically span several forms of media – including paper tokens, social media accounts, and
highly produced videos. Nor can they be reduced to being a “media strategy,” since they more
fundamentally structure participation than deliver messages. Perhaps the most powerful
argument for games as participation structures comes in how economists could debate Macon
Money as “expansionary monetary policy.” Transitioning to games as informal policy (or
perhaps as services) may help transcend seeing games as products.

parity yet in videogames, but the gap has been rapidly closing on many fronts. Yet issues of gender oppression and
empowerment in games remain vital areas of research and activism (Gee & Hayes, 2010; Ivory, 2006; Prescott &
Bogg, 2013; Williams, Martins, Consalvo, & Ivory, 2009; Wirman, 2014).
211
Prior work with digital games may have done the opposite. It is telling that “localization” in the commercial
games industry typically means adapting for a different language, at best a national culture. By contrast,
“localization” for social movements is a bottom-up organizing strategy to connect with place with a network
sensibility (Escobar, 2001). The console game industry in particular is a hybrid of two dominant cultures via
American and Japanese companies (Consalvo, 2006), and so an implicit flattening seems likely as other countries
adopt American and Japanese games into their own cultures. Non-digital games have been exported from a more
diverse set of countries, e.g., the Settlers of Catan (1995) from Germany.
243

The problem is partly a lack of unifying categories like Situated Games. Even success
stories like Macon Money, with its significant visibility in the press and published evaluation, are
described as singular experiments. Lacking a category, the game is not an “example of” – it is
simply an outlier. Outliers can be tracked for outcomes, but there is no sense of how they should
be linked. Thus Macon Money was declared “too expensive” for building a sense of community,
without comparing the impact to similarly sized tax abatements, let alone measuring the
economic and civic effects of building human capital.
The impact of Situated Games only emerges when they are analyzed across multiple
conceptual dimensions. This dissertation asserts that theory will be under-developed until
Macon Money can be functionally compared with hyper-local games like Reality Ends Here, and
with bottom-up activities like ParTour. To do so, this dissertation has proposed and
demonstrated the framework of ‘local fit’.
Another significant barrier is the presumed source of further innovation. Commercial
games are often assumed to have a natural lead in innovation. However, civil society may
actually have intrinsic advantage at the local level, where trust and face-to-face networks are
vital. The rise of mobile media is shifting how local networks form and communicate, even as
the basic principles of cohesion and accountability remain. Ironically, civic groups working on
games still defer to industry for guidance, and have not yet recognized how and where they may
already have leading tools and networks.
In the end, a paradigm shift may be needed in how civic groups approach design. For
real-world games, a kind of paradox needs to be embraced at the boundary of play and
sustainable change. On one hand, games succeed when they are lighthearted – playful and even
244

transgressive – shaking up norms of the everyday street with collective action. Certainly in this
study, play and the magic circle opened new space to meet neighbors, venture across lines of
socio-economic segregation and build teams to span disciplines; not to mention re-enacting
protest and fighting stereotypes.
Yet simultaneously, community change is long, hard work that demands sustainability
and organizational accountability. This study showed how local empowerment depends on a
multi-level approach that respects history and local identity, going beyond user-centered design
to consider the local ecology of organizations and media storytellers. Sustainability is vital, but
then again, so is play. The goal of this study, which provides shared language across theory and
empirical findings, is to sustain the healthy paradox toward a productive space for design.
More real-world games are on the way, built atop the feedback loops of local
communities, physical and digital. Fitting to the local ecology involves substantial extra work,
and raises the ethical stakes. But if the goals are to empower the local community, there may be
no better way than to pursue ‘local fit’.

245

BIBLIOGRAPHY

Ahmad, M. A., Borbora, Z., Shen, C., Srivastava, J., & Williams, D. (2011). Guild play in
MMOGs: rethinking common group dynamics models. In Social Informatics (pp. 145–
152). Springer.
Allen, D. S. (2006). Talking to strangers: Anxieties of citizenship since Brown v. Board of
Education. University of Chicago Press.
Anderson, B. (1983). Imagined communities: Reflections on the origin and spread of
nationalism. London and New York: Verso.
Anderson, J., & Rainie, L. (2012). The Future of Gamification. Pew Research Center. Retrieved
from http://www.pewinternet.org/2012/05/18/the-future-of-gamification/
Bagley, E., & Shaffer, D. W. (2010). The epistemography of urban and regional planning 912:
appropriation in the face of resistance. In Proceedings of the 9th International
Conference of the Learning Sciences - Volume 1 (pp. 81–88). International Society of the
Learning Sciences.
Ball-Rokeach, S. J. (1985). The origins of individual media-system dependency: Different
stories, questions, and ways of thinking. Communication Research, 12, 485–510.
Ball-Rokeach, S. J., Gonzalez, C., Son, M., & Kligler-Velenchik, N. (2012). Understanding
individuals in the context of their environment: Communication ecology as a concept and
method. In International Communication Association Annual Conference. Phoenix, AZ.
Ball-Rokeach, S. J., Kim, Y. C., & Matei, S. (2001). Storytelling neighborhood paths to
belonging in diverse urban environments. Communication Research, 28(4), 392–428.
Bandura, A. (2006). Guide for constructing self-efficacy scales. In Self-efficacy beliefs of
adolescents (pp. 307–337). Information Age Publishing.
Bar, F., Pisani, F., & Weber, M. (2007). Mobile technology appropriation in a distant mirror:
baroque infiltration, creolization and cannibalism. In Seminario sobre Desarrollo
Económico, Desarrollo Social y Comunicaciones Móviles en América Latina (pp. 1–41).
Buenos Aires: Abaporu Project. Retrieved from
http://abaporu.wordpress.com/2007/05/14/technology-appropriation/
Bennett, W. L., Wells, C., & Rank, A. (2009). Young citizens and civic learning: two paradigms
of citizenship in the digital age. Citizenship Studies, 13(2), 105–120.
Birkland, T. A. (2001). Introduction to the Policy Process. M.E. Sharpe.
246

Björk, S., Holopainen, J., Ljungstrand, P., & Mandryk, R. (2002). Special issue on ubiquitous
games. Personal and Ubiquitous Computing, 6(5-6), 358–361.
Björk, S., & Peitz, J. (2007). Understanding pervasive games through gameplay design patterns.
In Situated Play: Proceedings of the 2007 Digital Games Research Association
Conference (pp. 440–448). Retrieved from
http://lmc.gatech.edu/~cpearce3/DiGRA07/Proceedings/058.pdf
Blakely, E. J., & Leigh, N. G. (2010). Planning local economic development: theory and
practice (4th ed.). Sage Publications, Inc.
Boal, A. (1979). Theater of the Oppressed. London: Pluto Press.
Bogost, I. (2011, August 9). “Gamification Is Bullshit.” The Atlantic. Retrieved March 12, 2014,
from http://www.theatlantic.com/technology/archive/2011/08/gamification-is-
bullshit/243338/
Brady, H. E., Verba, S., & Schlozman, K. L. (1995). Beyond SES: A Resource Model of
Political Participation. The American Political Science Review, 89(2), 271–294.
doi:10.2307/2082425
Broad, G. M., Ball-Rokeach, S. J., Ognyanova, K., Picasso, T., Stokes, B., & Villanueva, G.
(2013). Understanding Communication Ecologies to Bridge Communication Research
and Community Action. Journal of Applied Communication Research, 41(4), 325–345.
doi:10.1080/00909882.2013.844848
Brough, M., Lapsansky, C., Gonzalez, C., Stokes, B., & Bar, F. (forthcoming). Participatory
Design of a Mobile Platform for Social Justice? Learning from the development of
VozMob. International Journal of Learning and Media.
Brunori, D. (2005). State Tax Policy: A Political Perspective. The Urban Institute.
Burawoy, M. (1979). Manufacturing Consent: Changes in the Labor Process Under Monopoly
Capitalism. University of Chicago Press.
Calori, C., & Chermayeff, I. (2007). Signage and wayfinding design: a complete guide to
creating environmental graphic design systems. Wiley. Retrieved from
http://www.bookforfree.us/PDF/Signage-and-Wayfinding-Design:-A-Complete-Guide-
to-Creating-Environmental-Graphic-Design-Systems-BY-Chris-Calori-ID11052.pdf
Carey, J. W. (2008). Communication as Culture, Revised Edition: Essays on Media and Society.
Routledge.
Castells, M. (1979). The urban question: A Marxist approach.
247

Castells, M., Fernandez-Ardevol, M., Qiu, J. L., & Sey, A. (2006). Mobile Communication and
Society: A Global Perspective. MIT Press.
Castronova, E., Williams, D., Shen, C., Ratan, R., Xiong, L., Huang, Y., & Keegan, B. (2009).
As real as real? Macroeconomic behavior in a large-scale virtual world. New Media &
Society, 11(5), 685.
Chen, N.-T. N., Dong, F., Ball-Rokeach, S. J., Parks, M., & Huang, J. (2012). Building a new
media platform for local storytelling and civic engagement in ethnically diverse
neighborhoods. New Media & Society, 14(6), 931–950.
Chen, N.-T. N., Ognyanova, K., Zhao, N., Liu, W., Gerson, D., Ball-Rokeach, S., & Parks, M.
(2013). Communication and socio-demographic forces shaping civic engagement patterns
in a multiethnic city. In Communication and community (pp. 207–232).
Cohen, C. J. (2010). Democracy Remixed: Black Youth and the Future of American Politics.
Oxford Univ Pr.
College Hill Alliance Receives Excellence in Economic Development Award from IEDC. (2013,
October 10). Mercer University News. Retrieved from
https://news.mercer.edu/articles/2013/college-hill-alliance-receives-excellence-in-
economic-development-award.cfm
Consalvo, M. (2006). Console video games and global corporations: Creating a hybrid culture.
New Media & Society, 8(1), 117–137. doi:10.1177/1461444806059921
Cottle, S. (2006). Mediatized rituals: beyond manufacturing consent. Media, Culture & Society,
28(3), 411–432. doi:10.1177/0163443706062910
Coulson, A. (2014, February 10). Gamification in Community Engagement Part 1: Offline -
commsgodigital. CommsGoDigital. Retrieved from
http://www.commsgodigital.com.au/2014/01/gamification-in-community-engagement-
offline/
Currid-Halkett, E., & Stolarick, K. (2010). Cultural capital and metropolitan distinction: Views
of Los Angeles and New York. City, Culture and Society, 1(4), 217–223.
doi:10.1016/j.ccs.2011.01.003
Currid-Halkett, E., & Stolarick, K. (2011). The great divide: economic development theory
versus practice-a survey of the current landscape. Economic Development Quarterly,
25(2), 143–157.
Dalisay, F., Kushin, M. J., Yamamoto, M., Liu, Y.-I., & Skalski, P. (2014). Motivations for game
play and the social capital and civic potential of video games. New Media & Society.
doi:10.1177/1461444814525753
248

De Sousa Santos, B. (1998). Participatory budgeting in Porto Alegre: toward a redistributive
democracy. Politics and Society, 26, 461–510.
Dearing, J. W., & Meyer, G. (2006). Revisiting Diffusion Theory. In Communication of
innovations: a journey with Ev Rogers. Sage.
Deterding, S., Khaled, R., Nacke, L., & Dixon, D. (2011). Gamification: Toward a Definition.
Presented at the CHI 2011, Vancouver, BC. Retrieved from http://gamification-
research.org/wp-content/uploads/2011/04/02-Deterding-Khaled-Nacke-Dixon.pdf
DiSalvo, C., Clement, A., & Pipek, V. (2013). Communities: Participatory Design for, with, and
by communities. In Routledge International Handbook of Participatory Design (pp. 182–
209). New York, NY: Routledge.
Du, H., Rosson, M. B., Carroll, J. M., & Ganoe, C. (2009). I Felt Like a Contributing Member of
the Class: Increasing Class Participation with Classcommons. In Proceedings of the ACM
2009 International Conference on Supporting Group Work (pp. 233–242). New York,
NY, USA: ACM. doi:10.1145/1531674.1531709
Duke, R. D. (1974). Gaming: the future’s language. New York, NY: Sage Publications.
Duncombe, S. (2007). Dream: re-imagining progressive politics in an age of fantasy. New York:
New Press; Distributed by W.W. Norton.
Earl, J., & Kimport, K. (2009). Movement Societies and Digital Protest: Fan Activism and Other
Nonpolitical Protest Online. Sociological Theory, 27(3), 220–243.
Elden, M., & Levin, M. (1991). Cogenerative learning: Bringing participation into action
research. Participatory Action Research, 127–142.
Erickson, T., & Kellogg, W. A. (2000). Social translucence: an approach to designing systems
that support social processes. ACM Transactions on Computer-Human Interaction
(TOCHI), 7(1), 59–83.
Escobar, A. (2001). Culture sits in places: reflections on globalism and subaltern strategies of
localization. Political Geography, 20(2), 139–174. doi:10.1016/S0962-6298(00)00064-0
Finnegan, J. R., & Viswanath, K. (1988). Community ties and use of cable TV and newspapers
in a Midwest suburb. Journalism & Mass Communication Quarterly, 65(2), 456–463.
Fjeld, M., Lauche, K., Bichsel, M., Voorhorst, F., Krueger, H., & Rauterberg, M. (2002).
Physical and virtual tools: Activity theory applied to the design of groupware. Computer
Supported Cooperative Work (CSCW), 11(1-2), 153–180.
Flanagan, M. (2009). Critical Play: Radical Game Design. The MIT Press.
249

Flanagan, M., Belman, J., Nissenbaum, H., & Diamond, J. (2007). A Method For Discovering
Values in Digital Games’. In Situated Play, Proceedings of DiGRA 2007 Conference.
Florida, R. (2003). Cities and the creative class. City & Community, 2(1), 3–19.
Florida, R. L. (2002). The rise of the creative class: and how it’s transforming work, leisure,
community and everyday life. Basic Civitas Books.
Foucault, M., Burchell, G., & Gordon, C. (1991). The Foucault effect: studies in
governmentality: with two lectures by and an interview with Michel Foucault. University
of Chicago Press.
Freire, P. (1968). Pedagogy of the Oppressed. (M. B. Ramos, Trans.) (30th Anniversary, 2000.).
New York: Continuum.
Fron, J., Fullerton, T., Morie, J. F., & Pearce, C. (2007). The hegemony of play. In Situated
Play: Proceedings of Digital Games Research Association 2007 Conference (pp. 1–10).
Tokyo, Japan.
Fullerton, T. (2014). What Games Do Well: Mastering Concepts in Play. In W. G. Tierney, Z.
Corwin, & G. Ragusa (Eds.), Postsecondary Play: The Role of Games and Social Media
in Higher Education (pp. 125–145). Baltimore, MD: John Hopkins University Press.
Fullerton, T., Chen, J., Santiago, K., Nelson, E., Diamante, V., Meyers, A., … DeWeese, J.
(2006). That cloud game: dreaming (and doing) innovative game design. In Proceedings
of the 2006 ACM SIGGRAPH symposium on Videogames (pp. 51–59). ACM.
Fullerton, T., Swain, C., & Hoffman, S. (2004). Game Design Workshop: Designing,
prototyping, and playtesting games. Focal Pr.
Galarneau, L. (2005). Spontaneous communities of learning: A social analysis of learning
ecosystems in massively multiplayer online gaming (MMOG) environments. In
Changing Views: Worlds in Play. Vancouver, British Columbia, Canada. Retrieved from
http://summit.sfu.ca/item/287
Gamberini, L., Martino, F., Scarpetta, F., Spoto, A., & Spagnolli, A. (2007). Unveiling the
structure: effects of social feedback on communication activity in online multiplayer
videogames. In Online Communities and Social Computing (pp. 334–341). Springer.
Retrieved from http://link.springer.com/chapter/10.1007/978-3-540-73257-0_37
Gamberini, L., Martino, F., Spagnolli, A., Baù, R., & Ferron, M. (2011). “Your Team Cohesion
is Low”: A Systematic Study of the Effects of Social Network Feedback on Mediated
Activity. In Online Communities and Social Computing (pp. 172–181). Springer.
Retrieved from http://link.springer.com/chapter/10.1007/978-3-642-21796-8_18
Garver, E. (1994). Aristotle’s Rhetoric: An Art of Character. University of Chicago Press.
250

Gee, J. P. (2005). Learning by design: Good video games as learning machines. E-Learning and
Digital Media, 2(1), 5–16.
Gee, J. P., & Hayes, E. R. (2010). Women and gaming: The Sims and 21st century learning. New
York, NY: Palgrave Macmillan.
Geertz, C. (1973). The interpretation of cultures. New York: Basic Books.
Glynn, S. M. (1990). Token Economy Approaches for Psychiatric Patients Progress and Pitfalls
Over 25 Years. Behavior Modification, 14(4), 383–407.
doi:10.1177/01454455900144002
Goffman, E. (1990). The Presentation of Self in Everyday Life (15th Edicao/Edition.). Penguin
Books, Limited (UK).
Gordon, E., & de Souza e Silva, A. (2011). Net Locality: Why Location Matters in a Networked
World. Wiley-Blackwell.
Gordon, E., Schirra, S., & Hollander, J. (2011). Immersive planning: a conceptual model for
designing public participation with new technologies. Environment and Planning-Part B,
38(3), 505.
Granovetter, M. S. (1973). The strength of weak ties. American Journal of Sociology, 1360–
1380.
Grzeskowiak, J. (2011). Governments engage residents with games. The American City &
County, 126(7). Retrieved from
http://search.proquest.com.libproxy.usc.edu/docview/875657686
Hargittai, E., & Walejko, G. (2008). The participation divide: Content creation and sharing in the
digital age. Information Communication and Society, 11(2), 239.
Hart, D., Donnelly, T. M., Youniss, J., & Atkins, R. (2007). High School Community Service as
a Predictor of Adult Voting and Volunteering. American Educational Research Journal,
44(1), 197 –219. doi:10.3102/0002831206298173
Hemment, D. (2006). Locative arts. Leonardo, 39(4), 348–355.
Herman, E. S., & Chomsky, N. (2008). Manufacturing consent: The political economy of the
mass media. Random House.
Huizinga, J. (1938). Homo ludens: A study of the play-element in culture. Taylor & Francis.
Hunt, D., & Ramon, A.-C. (Eds.). (2010). Black Los Angeles: American Dreams and Racial
Realities. NYU Press.
251

Ibrahim, S., & Alkire, S. (2007). Agency and empowerment: A proposal for internationally
comparable indicators. Oxford Development Studies, 35(4), 379–403.
Ito, M. (2005). Personal, Portable, Pedestrian: Mobile Phones in Japanese Life. MIT Press.
Ito, M., Baumer, S., Bittanti, M., boyd, danah, Cody, R. G., Herr-Stephenson, B., … Tripp, L.
(2009). Hanging Out, Messing Around, and Geeking Out: Kids Living and Learning with
New Media. The MIT Press.
Ivory, J. D. (2006). Still a Man’s Game: Gender Representation in Online Reviews of Video
Games. Mass Communication and Society, 9(1), 103–114.
doi:10.1207/s15327825mcs0901_6
Jacobs, J. (1961). The Death and Life of Great American Cities. New York: Vintage.
Jacobs, J. (1970). The economy of cities. New York: Vintage.
Järvinen, A. (2008, March 8). Games without frontiers: Theories and methods for game studies
and design (Doctoral dissertation). University of Tampere, Finland. Retrieved from
https://uta17-kk.lib.helsinki.fi/handle/10024/67820
Jenkins, H. (2006). Convergence culture: Where old and new media collide. NYU Press.
Jenkins, H. (2007, March). Transmedia storytelling 101. Confessions of an Aca/Fan. Retrieved
from http://henryjenkins.org/2007/03/transmedia_storytelling_101.html
Jenkins, H., Clinton, K., Purushotma, R., Robison, A., & Weigel, M. (2007). Confronting the
challenges of participatory culture: Media education for the 21st century. MIT Press.
Jenkins, H., Ford, S., & Green, J. (2012). Spreadable media: Creating value and meaning in a
networked culture. New York, NY: New York University Press.
Juul, J. (2013). The art of failure: an essay on the pain of playing video games. Cambridge,
Mass.: MIT Press.
Kahne, J., & Middaugh, E. (2008). Democracy for some: The civic opportunity gap in high
school. Policies for Youth Civic Engagement, 29–58.
Kahne, J., Middaugh, E., & Allen, D. (2014). Youth, New Media, and the Rise of Participatory
Politics (No. 1). Youth and Participatory Politics Research Network. Retrieved from
http://ypp.dmlcentral.net/sites/default/files/publications/YPP_WorkinPapers_Paper01.pdf
Kahne, J., Middaugh, E., & Evans, C. (2009). The civic potential of video games. MIT Press.
Retrieved from http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=11851
252

Khatib, F., DiMaio, F., Cooper, S., Kazmierczyk, M., Gilski, M., Krzywda, S., … Popović, Z.
(2011). Crystal structure of a monomeric retroviral protease solved by protein folding
game players. Nature Structural & Molecular Biology, 18(10), 1175–1177.
Kim, Y. C., & Ball-Rokeach, S. J. (2006a). Civic engagement from a communication
infrastructure perspective. Communication Theory, 16(2), 173–197.
Kim, Y. C., & Ball-Rokeach, S. J. (2006b). Community storytelling network, neighborhood
context, and civic engagement: A multilevel approach. Human Communication Research,
32(4), 411–439.
Kim, Y. C., Jung, J. Y., & Ball-Rokeach, S. J. (2006). “Geo-Ethnicity” and Neighborhood
Engagement: A Communication Infrastructure Perspective. Political Communication,
23(4), 421–441.
Kligler-Vilenchik, N., & Shresthova, S. (2012). Learning Through Practice: Participatory
Culture Practices. Los Angeles, CA: Civic Paths Research Group at the USC Annenberg
School for Communication and Journalism. Retrieved from
http://ypp.dmlcentral.net/sites/all/files/publications/Learning%20Through%20Practice_K
ligler-Shresthova_Oct-2-2012.pdf
Klopfer, E., & Squire, K. (2008). Environmental Detectives--the development of an augmented
reality platform for environmental simulations. Educational Technology Research and
Development, 56(2), 203–228.
Kretzmann, J. P., & McKnight, J. L. (1993). Building Communities From The Inside Out: A Path
Toward Finding And Mobilizing A Community’s Assets. ACTA Publications. Retrieved
from http://www.openisbn.org/download/087946108X.pdf
Laurel, B. (2001). Utopian entrepreneur. Cambridge, Mass.: MIT Press.
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation.
Cambridge University Press.
Lee, J. J., & Hammer, J. (2011). Gamification in Education: What, How, Why Bother? Academic
Exchange Quarterly, 15(2), 2.
Lefebvre, H. (1974). The production of space. (D. Nicholson-Smith, Trans.). Oxford,
Cambridge: Blackwell.
Lerner, J. (2014). Making Democracy Fun: How Game Design Can Empower Citizens and
Transform Politics. Cambridge, Massachusetts: The MIT Press.
Levine, P. (2007). The future of democracy: Developing the next generation of American
citizens. Lebanon, New Hampshire: Tufts University Press.
253

Light, J. (2008). Taking Games Seriously. Technology and Culture, 49(2), 347–375.
Ling, R. S., & Campbell, S. W. (2009). The reconstruction of space and time: mobile
communication practices. New Brunswick: Transaction Publishers.
Macklin, C. (2010). Reacting to Re: Activism: A Case Study in the Ethics of Design. In Ethics
and Game Design: Teaching Values Through Play (p. 274). Information Science
Reference.
Macklin, C., & Thomson, G. (2011). Re:Activism: Serendipity in the Streets. In S. Dikkers, J.
Martin, & B. Coulter (Eds.), Mobile Media Learning: Amazing Uses of Mobile Devices
for Learning. ETC Press. Retrieved from
http://www.etc.cmu.edu/etcpress/content/mobile-media-learning
Marchi, R. (2011). From disillusion to engagement: Minority teen journalists and the news
media. Journalism. doi:10.1177/1464884911431379
Martin, J., Dikkers, S., Squire, K., & Gagnon, D. (2014). Participatory Scaling Through
Augmented Reality Learning Through Local Games. TechTrends, 58(1), 35–41.
Matei, S., Ball-Rokeach, S. J., & Qiu, J. L. (2001). Fear and misperception of Los Angeles urban
space. Communication Research, 28(4), 429.
Mathews, J., & Holden, J. (2011). Place-based Design for Civic Participation. In S. Dikkers, J.
Martin, & B. Coulter (Eds.), Mobile Media Learning: Amazing Uses of Mobile Devices
for Learning (pp. 131–148). ETC Press. Retrieved from
http://dl.acm.org/citation.cfm?id=2331562.2331570
McFarland, D. A., & Thomas, R. J. (2006). Bowling young: How youth voluntary associations
influence adult political participation. American Sociological Review, 71(3), 401–425.
Mcgonigal, J. (2003). This Is Not a Game: Immersive Aesthetics and Collective Play. In
Melbourne DAC 2003 Streamingworlds Conference Proceedings. RMIT University.
McGonigal, J. (2011a). Reality is Broken: Why Games Make Us Better and How They Can
Change the World. New York, NY: Penguin Press.
McGonigal, J. (2011b, March 1). We don’t need no stinkin’ badges: How to re-invent reality
without gamification. Presented at the Serious Games Summit at the Game Developers
Conference, San Francisco, CA. Retrieved from
http://www.slideshare.net/avantgame/we-dont-need-no-stinkin-badges-how-to-reinvent-
reality-without-gamification
Mobile Voices (Añorve R., F. Bar, M. Brough, A. Cisneros, S. Costanza-Chock, A. Lucía
Garcés, C. Gonzalez, M. de Lourdes González Reyes, C. Jimenez, C. Lapsansky, M.
Mancia, M. Rodriguez, and C. Wallis), (first). (2011). Mobile Voices: projecting the
254

voices of immigrant workers by appropriating mobile phones for popular communication.
In Communications research in action: scholar-activist collaborations for a democratic
public sphere. New York, NY: Fordham University Press.
Montola, M., Stenros, J., & Waern, A. (2009). Pervasive games: theory and design. Burlington,
MA: Morgan Kaufmann.
Mowbray, C. T., Holter, M. C., Teague, G. B., & Bybee, D. (2003). Fidelity criteria:
Development, measurement, and validation. American Journal of Evaluation, 24(3), 315–
340.
Nardi, B. A. (1996). Context and Consciousness: Activity Theory and Human-computer
Interaction. MIT Press.
National Institute on Media and the Family. (2008). Fact Sheet: Effects of Video Game Playing
on Children. Retrieved July 8, 2008, from http://www.parentfurther.com/technology-
media/facts/facts_effect.shtml
Nicholson, S. (2012). A user-centered theoretical framework for meaningful gamification. In
Games+ Learning+ Society (Vol. 8, pp. 223–230). ETC Press. Retrieved from
http://press.etc.cmu.edu/content/gls-80-conference-proceedings
Nicodemus, A. G. (2013). Fuzzy vibrancy: Creative placemaking as ascendant US cultural
policy. Cultural Trends, 22(3-4), 213–222.
Norris, P. (2001). Digital Divide: Civic Engagement, Information Poverty, and the Internet
Worldwide. Cambridge University Press.
North, P. (2007). Money and Liberation: The Micropolitics of Alternative Currency Movements.
U of Minnesota Press.
Oldenburg, R. (1989). The great good place: Cafés, coffee shops, community centers, beauty
parlors, general stores, bars, hangouts, and how they get you through the day. Paragon
House New York.
Ong, P., Firestine, T., Pfeiffer, D., Poon, O., & Tran, L. (2008). The State of South LA. Los
Angeles, CA: UCLA School of Public Affairs. Retrieved from
http://www.academia.edu/220952/The_State_of_South_LA
Opp, K. D. (2009). Theories of political protest and social movements: a multidisciplinary
introduction, critique, and synthesis. Taylor & Francis.
Page, S. E. (2008). The difference: How the power of diversity creates better groups, firms,
schools, and societies. Princeton, NJ: Princeton University Press.
255

Parasuraman, A. (2000). Technology Readiness Index (TRI) a multiple-item scale to measure
readiness to embrace new technologies. Journal of Service Research, 2(4), 307–320.
Pastor, M., Benner, C., & Matsuoka, M. (2009). This could be the start of something big: how
social movements for regional equity are reshaping metropolitan America. Cornell Univ
Pr.
Pelling, C. (2005). Tragedy, Rhetoric, and Performance Culture. A Companion to Greek
Tragedy, 83.
Peterson, N. A., Speer, P. W., & McMillan, D. W. (2008). Validation of A brief sense of
community scale: Confirmation of the principal theory of sense of community. Journal of
Community Psychology, 36(1), 61–73. doi:10.1002/jcop.20217
Plastrik, P., & Taylor, M. (2006). Net gains: A handbook for network builders seeking social
change (p. 117). Retrieved from
http://networkimpact.org/downloads/NetGainsHandbookVersion1.pdf
Poremba, C. (2007). Critical potential on the brink of the magic circle. In Proceedings of DiGRA
2007 situated play conference. September (pp. 24–28). Retrieved from
http://www.digra.org/wp-content/uploads/digital-library/07311.42117.pdf
Prescott, J., & Bogg, J. (2013). Gender Divide and the Computer Game Industry (1st ed.).
Hershey, PA: IGI Global.
Prezza, M., Pacilli, M. G., Barbaranelli, C., & Zampatti, E. (2009). The MTSOCS: A
multidimensional sense of community scale for local communities. Journal of
Community Psychology, 37(3), 305–326. doi:10.1002/jcop.20297
Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community. New
York, NY: Simon & Schuster.
Raczkowski, F. (2013). It’s all fun and games… A history of ideas concerning gamification.
Retrieved from http://www.digra.org/wp-content/uploads/digital-library/paper_344.pdf
Ramati, P. (2012, April 1). Soapbox derby features record number of entrants. The Telegraph.
Macon, GA. Retrieved from http://www.macon.com/2012/04/01/1971403/soapbox-
derby-features-record.html
Robertson, M. (2010, October 6). Can’t play, won’t play. Hide&Seek. Retrieved from
http://www.hideandseek.net/2010/10/06/cant-play-wont-play/
Rogers, E. M. (1995). Diffusion of innovations. Simon and Schuster.
256

Rubin, B. C., Hayes, B., & Benson, K. (2009). “It’s the Worst Place to Live”: Urban Youth and
the Challenge of School-Based Civic Learning. Theory Into Practice, 48(3), 213–221.
doi:10.1080/00405840902997436
Ruiz, S., Stokes, B., & Watson, J. (2012). Mobile and Locative Games in the “Civic Tripod:”
Activism, Art and Learning. International Journal of Learning and Media, 3(3).
doi:10.1162/IJLM_a_00078
Said, E. W. (2012). Culture and Imperialism. Random House LLC.
Salen, K. (Ed.). (2008). The ecology of games: Connecting youth, games, and learning.
Cambridge, MA: The MIT Press. Retrieved from
http://mitpress.mit.edu/catalog/item/default.asp?tid=11392&ttype=2
Salen, K., & Zimmerman, E. (2004). Rules of play: Game design fundamentals. The MIT Press.
Sampson, R. J. (2012). Great American city: Chicago and the enduring neighborhood effect.
Chicago, IL: University of Chicago Press.
Schell, J. (2008). The Art of Game Design: A book of lenses. CRC Press.
Schiavo, E., Rodríguez, S., & Vera, P. (2013). Appropriation of ICTs by informal communities
in metropolitan cities. The case of the “La Salada” market in the Latin American context.
The Journal of Community Informatics, 9(3). Retrieved from http://ci-
journal.net/index.php/ciej/article/view/715
Schirra, S. M. (2013). Playing for impact: the design of civic games for community engagement
and social action (Thesis). Massachusetts Institute of Technology, Cambridge, MA.
Retrieved from http://dspace.mit.edu/handle/1721.1/81134
Schlozman, K. L., Verba, S., & Brady, H. E. (2010). Weapon of the strong? Participatory
inequality and the Internet. Perspectives on Politics, 8(2), 487–509.
Scholz, T. (2013). Digital labor: the Internet as playground and factory. New York: Routledge.
Schön, D. A. (1983). The reflective practitioner: How professionals think in action. New York,
NY: Basic Books.
Schrier, K., & Gibson, D. (2010). Ethics and game design: Teaching values through play.
Information Science Reference.
Schulzke, M. (2011). How Games Support Associational Life: Using Tocqueville to Understand
the Connection. Games and Culture, 6(4), 354.
257

Schwartz, B., Ward, A., Monterosso, J., Lyubomirsky, S., White, K., & Lehman, D. R. (2002).
Maximizing versus satisficing: Happiness is a matter of choice. Journal of Personality
and Social Psychology, 83(5), 1178–1197.
Sethi, R. (2013, April 25). Macon Money. thoughts on economics, finance, crime and identity...
Retrieved from http://rajivsethi.blogspot.com/2013/04/macon-money.html
Shaffer, D. W., & Gee, J. P. (2012). The right kind of GATE: Computer games and the future of
assessment. In M. C. Mayrath, J. Clarke-Midura, D. H. Robinson, & G. Schraw (Eds.),
Technology-based assessments for 21st century skills: Theoretical and practical
implications from modern research. Charlotte, NC: Information Age.
Sherman, A. (2011, August 10). How Mobile Apps Are Helping Urban Explorers Discover Their
Cities. Mashable.com. Retrieved from http://mashable.com/2011/08/10/mobile-apps-
cities/
Shipan, C. R., & Volden, C. (2008). The Mechanisms of Policy Diffusion. American Journal of
Political Science, 52(4), 840–857. doi:10.1111/j.1540-5907.2008.00346.x
Sicart, M. (2008). Defining game mechanics. Game Studies, 8(2), 1–14.
Smith, C. (2011). GNOi iPhone App Unveiled at GNO, Inc. Annual Meeting (Press Release).
Greater New Orleans, Inc. Retrieved from http://gnoinc.org/press-releases/gnoi-iphone-
app-unveiled-at-gno-inc-annual-meeting
Soja, E. W. (2010). Seeking spatial justice. Minneapolis, London: University of Minnesota Press.
Squire, K. (2002). Cultural framing of computer/video games. International Journal of
Computer Game Research, 2(1). Retrieved from http://gamestudies.org/0102/squire/
Squire, K., Jan, M., Mathews, J., Wagler, M., Martin, J., DeVane, B., & Holden, C. (2007).
Wherever you go, there you are: Place-based augmented reality games for learning. The
Educational Design and Use of Simulation Computer Games, 265–296.
Statistics-At-A-Glance (2010-2011): USC School of Cinematic Arts. (n.d.). University of
Southern California. Retrieved April 28, 2012, from
http://cinema.usc.edu/admissions/statistics/index.cfm
Steinkuehler, C., & Williams, D. (2006). Where everybody knows your (screen) name: Online
games as “third places.” Journal of Computer-Mediated Communication, 11(4), 885–909.
Stewart, J., Bleumers, L., Van Looy, J., Mariën, I., All, A., Schurmans, D., … Misuraca, G.
(2013). The Potential of Digital Games for Empowerment and Social Inclusion of Groups
at Risk of Social and Economic Exclusion: Evidence and Opportunity for Policy (EUR -
Scientific and Technical Research Reports). Institute for Prospective and Technological
258

Studies, Joint Research Centre, European Union. Retrieved from
http://publications.jrc.ec.europa.eu/repository/handle/111111111/29415
Stokes, B. (2010). Peter Packet Challenge: Featuring a Videogame, Corporate Partnership Trade-
offs, and Pre-Teen Activists. In From Participatory Culture to Public Participation (1st
ed.). Los Angeles, CA: USC Civic Engagement Research Group. Retrieved from
http://sites.google.com/site/participatorydemocracyproject/case-studies/peter-packet
Stokes, B. (2012). Restructuring Civic Engagement: Meaningful Choice and Game Design
Thinking. In A. Delwiche & J. J. Henderson (Eds.), The Participatory Cultures
Handbook. New York, NY: Routledge.
Stokes, B., Watson, J., Fullerton, T., & Wiscombe, S. (2013). A Reality Game to Cross
Disciplines: Fostering Networks and Collaboration. In Defragging Game Studies:
Proceedings of DiGRA 2013 International Conference. Atlanta, GA.
Stokes, B., & Williams, D. (2012). Civic Beyond Play: Ties to Public Life for Small-Group
Gamers. In Proceedings 8.0 (pp. 283–290). Madison, WI: ETC Press. Retrieved from
http://www.etc.cmu.edu/etcpress/content/gls-80-conference-proceedings
Stokols, D. (1996). Translating social ecological theory into guidelines for community health
promotion. American Journal of Health Promotion, 10(4), 282–298.
Susi, T., Johannesson, M., & Backlund, P. (2007). Serious games: An overview. Retrieved from
http://www.diva-portal.org/smash/record.jsf?pid=diva2:2416
Taylor, M., & Whatley, A. (2012a). Macon Money Final Evaluation Report. Knight Foundation.
Retrieved from http://www.knightfoundation.org/publications/games-and-community-
building-final-report
Taylor, M., & Whatley, A. (2012b). Macon Money Social Impact Game: Summary of Evaluation
Results (p. 50). Network Impact and Cause Communications. Retrieved from
http://www.knightfoundation.org/publications/games-and-community-building-final-
report
Taylor, T. L. (2006). Beyond Management: Considering Participatory Design and Governance in
Player Culture. First Monday, (7). doi:10.5210/fm.v0i0.1611
Thomas, D., & Brown, J. S. (2009). Why virtual worlds can matter. International Journal of
Learning, 1(1), 37–49.
Tocqueville, A. de. (1848). Democracy in America. (H. Reeve, Trans., J. C. Spencer, Ed.) (8th
ed.). New York: Pratt, Woodford, & Co.
Valente, T. W. (2010). Social networks and health: Models, methods, and applications. New
York, NY: Oxford University Press.
259

Valente, T. W., Gallaher, P., & Mouttapa, M. (2004). Using social networks to understand and
prevent substance use: A transdisciplinary perspective. Substance Use & Misuse, 39(10-
12), 1685–1712.
Vanden Abeele, V. A., & Van Rompaey, V. (2006). Introducing Human-centered Research to
Game Design: Designing Game Concepts for and with Senior Citizens. In CHI ’06
Extended Abstracts on Human Factors in Computing Systems (pp. 1469–1474). New
York, NY, USA: ACM. doi:10.1145/1125451.1125721
Villanueva, G., & Broad, G. (2012). Communication Asset Mapping: An Ecological Application
Toward Building Healthy Communities in South Los Angeles. Presented at the
Conference of the National Communication Association, Phoenix, AZ.
Von Neumann, J., & Morgenstern, O. (2007). Theory of Games and Economic Behavior (60th
Anniversary Commemorative Edition). Princeton university press.
Vygotsky, L. S. (1978). Mind in Society: The Development of Higher Psychological Processes.
(M. Cole, V. John-Steiner, S. Scribner, & E. Souberman, Eds.) (14th ed.). Harvard
University Press.
Watson, J. (2012, August). Reality Ends Here: Environmental Game Design and Participatory
Spectacle. Dissertation, USC School of Cinematic Arts.
Watson, J., & Reality Design Team. (2013). Reality: The Manual (1st ed.). Los Angeles, CA:
The Reality Committee. Retrieved from http://realitycommittee.org/wp-
content/uploads/2013/06/Reality-Manual-Book-First-Edition-Web-rev1.pdf
Wilkin, H. A., Ball-Rokeach, S. J., Matsaganis, M. D., & Cheong, P. H. (2007). Comparing the
communication ecologies of geo-ethnic communities: How people stay on top of their
community. Journal of Electronic Communication, 17(2). Retrieved from
http://www.cios.org/EJCPUBLIC/017/1/01711.HTML
Williams, D., Martins, N., Consalvo, M., & Ivory, J. D. (2009). The virtual census:
representations of gender, race and age in video games. New Media & Society, 11(5),
815–834. doi:10.1177/1461444809105354
Wirman, H. (2014). Gender and Identity in Game-Modifying Communities. Simulation &
Gaming, 1046878113519572. doi:10.1177/1046878113519572
Yaroslavsky, Z. (2012, February 28). County gets a new bike plan at last. Retrieved April 23,
2012, from http://zev.lacounty.gov/news/county-gets-a-new-bike-plan-at-last
Youniss, J., & Yates, M. (1997). Community service and social responsibility in youth.
University of Chicago Press.
260

Zhong, Z.-J. (2011). The effects of collective MMORPG (Massively Multiplayer Online Role-
Playing Games) play on gamers’ online and offline social capital. Computers in Human
Behavior, 27(6), 2352–2363. doi:10.1016/j.chb.2011.07.014
Zhu, M., Huang, Y., & Contractor, N. S. (2013). Motivations for self-assembling into project
teams. Social Networks, (in press/avail online). doi:10.1016/j.socnet.2013.03.001
Zichermann, G. (2010, December 10). VCs level up with “gamification” investments.
VentureBeat. Retrieved from http://venturebeat.com/2010/12/10/vcs-level-up-with-
gamification-investments-2/
Zimmerman, E. (2013, September 9). Manifesto: The 21st Century Will Be Defined By Games.
Kotaku. Retrieved from http://kotaku.com/manifesto-the-21st-century-will-be-defined-
by-games-1275355204

261

APPENDICES

Appendix 1: Macon Money timeline and data sources
Key dates for the game and data collection are detailed below. See also the interactive
timeline
212
published by the Knight Foundation.
1) Fall 2010: The game launched (October 6
th
) and the first bonds are redeemed
2) Spring 2011: The final public events for the game are in early April, the last bonds are
redeemed May 31 2011, and the last day for using Macon Money currency in businesses was
June 15 2011. Online conversation in the message boards and Facebook group tapered off
shortly thereafter.

Sources from the initial evaluation of Macon Money led by Network Impact
1. Online survey of players: Includes some genuine pre-post pairs; however, the survey was
launched mid-game and so there are also many “mid-game” comparisons with “post
game.” As a result, the effects of the game are likely to be under-reported. Data captured
includes the registrant’s name, street address, zip, email, how they found out about the
game, number of times they received a bond, frequency of visiting specific areas,
anticipated benefit of playing, attitudes about Macon’s future, demographic
characteristics including age, race/ethnicity, tenure in Macon, and Mercer student status.

212
The interactive timeline, published April 2012 by the Knight Foundation as part of the evaluation report release,
is available at http://knightlive.org/games/timeline.html
262

2. Interviews: transcripts including 26 with players, some designers, and stakeholders at
organizations in and around Macon, Georgia.
3. Business Survey. Results of online survey to 40 businesses including why they decided
to participate, factors that impeded/facilitated participation, resulting changes in clientele,
promotion strategies they used, factors that improved outcomes for businesses, and
willingness to invest in the game if it is played again in Macon.
4. Paper Questionnaire of Participants: answers to 1-page paper survey
5. Bond tracking database: includes 5021 bonds (courtesy of Area/Code, as of April 2011)
Additional data sources:
• Participant observation of community planning meetings in the fall of 2013 in Macon
Georgia with attendees from the Knight Foundation and the College Alliance
• Follow-up interviews with key stakeholders in Macon, with the game designers, and with
Knight Foundation staff, including:
o Designer interview of Kati London by Benjamin Stokes (Ruiz et al., 2012 – see:
http://civictripod.com/interview-kati-london/)
• Press coverage of the Macon Money game during and after the run
• Academic discussion of the game in blogs and scholarly articles (including Sethi, 2013)

Key Limitations:
• No player tracking. The game did not assign a unique identifier to each player. Although
local project staff had the means to track bonds and bills, they could not track the players
263

except at the moment when they exchanged their bonds for full bills. The result,
according to the evaluators was that there is no way of determining which players spent
bills at what businesses and in what denominations/amounts. Inferences can be drawn
based on the subsequent survey data, which addressed some of these questions.
Appendix 2: Data for Reality Ends Here and Xposure
Timeline for data collection
1. Aug-Dec 2011 is Year 1 of play at SCA (The School of Cinematic Arts at the University
of Southern California) with primary facilitation by Jeff Watson and Simon Wiscombe
a. Interviews with the game’s designers and facilitators begin in March of 2011, and
are conducted periodically going forward
b. Network data is scraped from the website for first network analysis (119 projects
by 103 individuals)

2. Aug-Dec 2012 is Year 2 of play at SCA
a. Interviews with the game’s designers and facilitators continue
b. Network data is scraped from the website for second network analysis (80
individuals)
c. Exit survey conducted on attitudes & some network data collected, December
2012

264

3. Aug-Dec 2013 is Year 3 at SCA);
additionally, this is the pilot year for School Two (the School for Communication &
Journalism)
a. Incoming student surveys at both schools are conducted to compare media skills
and orientation, August and September 2013
b. Interviews are conducted with the game’s designers and facilitators at both
schools
c. Network data is scraped for both schools (not referenced by this analysis,
however)
d. Exit interviews conducted with School Two students, culminating February 2014
Network data summary
Network data was collected by analyzing the game website, which listed collaborators for
each completed project. To access the data, the game website was scraped and analyzed
independently with custom scripts, then analyzed with the UCI Net software.
Data was drawn from two years, each year involving more than 100 participants
213
and
producing hundreds of team projects. Relational pairs were inferred between each collaborator
on a group project, and thus were always mutual (bidirectional). Players who completed
multiple deals together were assumed to have proportionally stronger ties.

213
Participants in the study were the incoming undergraduate classes at the USC School of Cinematic Arts in the fall
of 2011 and 2012. According to the University, the overall undergraduate mix for undergraduates of all grades is
approximately 60% male, with approximately 10,000 living alumni and 850 enrolled students (“Statistics-At-A-
Glance (2010-2011): USC School of Cinematic Arts,” n.d.).
265

Survey of incoming students at both schools (conducted fall 2013)
The quantitative data comparing the incoming students of each school is from an online
survey that was distributed to all incoming students (freshmen and transfer students) in the
beginning of the 2013-14 school year. Each of the two schools did their own distribution,
spreading the link to the online survey via a mass email.
Completing the survey was optional; a lottery of small rewards were offered. The
distribution dates were a bit different, with some School Two students doing theirs a month later
and thus after classes had begun. (The School of Cinematic Arts students mostly finished their
surveys August 15-25
th
2013, while the School Two students had a group around September 3
rd

and another push around September 23
rd
.)
A reasonable sample (a total of 240 respondents) was obtained for each school. The
response rate among the relevant student populations was approximately 45% in each case, or
about 126 from the School of Cinematic Arts, and 114 from School Two.
Interview process overview
Interviews were at the heart of understanding how the game unfolded, to complement
direct observation in person and face-to-face. Early interviews were exclusively with the game’s
designers to understand their intent. (An incredible debt of gratitude is due to these formative
conversations for shaping the research design for the game and the broader framework for ‘local
fit’.) Later that first year, a series of interviews in the first year with administrators and
stakeholders at the School of Cinematic Arts were designed to capture some of the institutional
constraints, the barriers to adoption, and the internal perceptions of how the game was providing
different structures than what the classrooms and existing campus could offer.
266

As the game’s impact became more apparent, interviews shifted to focus on the
conditions for success and how the game was being optimized between years. For the
comparison of the School of Cinematic Arts and School Two, interviews were conducted at
several points as the game unfolded with staff and faculty, to complement the observation of the
game online and in casual conversations with students. To probe some of the students’ views on
the failure of game at School Two, longer interviews were conducted with staff at both schools,
brief interviews were conducted with several students, and in-depth interviews were conducted
with two students from School Two who had engaged the most with the game, in hopes of
understanding why their participation did not go further, and how the implementation at their
school was distinct.
Appendix 3: Data and timeline for the ParTour case study
The case study of ParTour is based on an approximately two year research program of
collaborative design, participant observation, and stakeholder interviews. It began in the fall of
2011.
ParTour began as an innovation in partnership, as much as in technology. The partners
deserve much of the credit for the work, as well as data collected through interviews and social
media posting over several years. Consider the complementary roles of the three primary
organizations: First, for the technological dimension, the Mobile Lab at the University of
Southern California (USC) Annenberg School for Communication and Journalism adapted a
platform for everyday cell phones that was initially developed by Mobile Voices. Second, the
community-organizing partner was T.R.U.S.T. South LA, which also set the advocacy agenda
for the first bicycling map. Third, for theory on how media affects civic outcomes, the USC
267

Metamorphosis Project was involved in the design and evaluation, and helped the project focus
on storytelling networks. These three partnerships represent a positive outcome for the project in
connecting research and practice across the community/university divide. These initial
partnerships evolved over time, as sustainability depended on organic alignment with community
partners, their funding, and the issues that seemed to most resonate with community members.
To give a little more detail, the project’s predecessor, Mobile Voices (also called VozMob
and Voces Moviles)
214
, is the original source for the technical functionality for cellphone-based
storytelling. More importantly, the methods of participatory design and philosophy of joint
storytelling emerged at Mobile Voices.
215
The analysis of ParTour as a tool draws on participant
observation and field notes taken between the time when I first joined the Mobile Voices team in
the fall of 2009 as a volunteer and design researcher, until late 2011 when my participation
shifted to ParTour.
The ParTour project began in the fall of 2011, testing the mapping functionality that had
been added to the Mobile Voices software, and beginning the collaboration with community
bicycling groups in South Los Angeles. In early 2012, the project launched a website focused on
bicycling in South LA (http://RideSouthLA.com), and released its first print map in April of that

214
Officially, Mobile Voices is “a platform for immigrant and/or low-wage workers in Los Angeles to create stories
about their lives and communities directly from cell phones. VozMob appropriates technology to create power in our
communities and achieve greater participation in the digital public sphere.” This description comes from the
group’s website; for more see http://vozmob.net/en/about
215
The process might best be described as community-based participatory design, in the model of DiSalvo et al.
(2013). Several articles have been written on the specific process for Mobile Voices (Brough, Lapsansky, Gonzalez,
Stokes, & Bar, forthcoming; Mobile Voices (Añorve R., F. Bar, M. Brough, A. Cisneros, S. Costanza-Chock, A.
Lucía Garcés, C. Gonzalez, M. de Lourdes González Reyes, C. Jimenez, C. Lapsansky, M. Mancia, M. Rodriguez,
and C. Wallis), 2011).
268

year
216
. A second major print map was launched in the spring of 2013, focusing on bicycling
and food toward a healthier South LA.
217
Group walks for mapping were used to create the
second map, as well as the group rides. Between 2011 and 2014, a half-dozen mapping events
were conducted with a variety of community events in South LA, most often in connection to
one of the print maps, but sometimes as independent events.
Interviews were periodically conducted by various members of the ParTour research
team. Often the interviews were videotaped and posted online, consistent with the philosophies
of action research that seek to foster cogenerative learning (Elden & Levin, 1991). I also kept
field notes during this time, with some of the best analysis on the activity’s core features
emerging in “debrief” sessions after events with community partners and university students. A
series of more focused and semi-structured interviews were conducted in the spring of 2012 with
each of the community partners, including several community-based organizations.
Game-like “quests” and “missions” were also tested with ParTour as part of mapping in a
variety of settings, including with an after-school group at a local High School, and in the
massive street festival called CicLAvia that regularly closes miles of Los Angeles streets to cars,
and instead welcomes 100,000 riders and walkers. Field notes detailed some of what worked,
such as the need to keep tagging limited to a narrow range of tags in order to stay focused on
missions. Yet most of these design experiments served as background context for the project.
Work with the game Re:Activism Atlanta provided the most substantial investigation of
the overlap between games and the storytelling methods. I had conducted interviews with the

216
Details on this first “Watts Ride” map, including a press release announcing the launch, are available at
http://ridesouthla.com/watts-ride-map/
217
Details on the “Healthy Food Map” are available at: http://ridesouthla.com/food-map/
269

lead designer of the game, including one that was published online as part of an investigation of
mobile games for civic engagement and learning (Ruiz et al., 2012, sec.
http://civictripod.com/colleen–macklin/). For the Atlanta localization, I participated in two runs
of the game in the summer of 2013. In one case I joined a team, documenting the player
experience. In the second case, I managed the integration of ParTour tools in the “game office”
during the live game. Interviews with players and designers after the game were semi-structured.
The case study is based on my own notes, the aforementioned interviews, and the analysis of
media produced by players in the game.

270

Appendix 4: Recipe for participatory mapping with ParTour
The informal guide below was created to emphasize the “social recipe” that accompanies the tool
in order to maximize the feedback loops and align levels of the communication ecology.

Written by the RideSouthLA Team, May 2013

WHY THIS GUIDE: We think it’s easy to do basic mapping. But it’s much harder to maximize your
impact. That’s why we are creating this “recipe” guide for community-based organizations.

==================
STRATEGY:
1. Augment an Event (don’t create new events). These do not need to exclusively be mapping
events. It is best to augment an existing event that your organization wanted to do anyway, such
as a group bicycle ride or community pride parade or a neighborhood walk.
2. Advertise your process
218
and pick your visual spectacle
219
in advance. How the group moves
through streets will perform a kind of theater that should raise curious questions from bystanders
and make for good pictures by participants and the press. Ethically, the procession should be
something everyone can contribute to and modify in their own way. (We call this “ethical
spectacle,” to borrow a term from Stephen Duncombe.)
3. Pick a partner to help bridge to a campaign. Success is about making this bigger than one
organization – all our maps are about being inclusive, and it really helps to muddy the waters by

218
For example, we often call our mapping rides “parades” to emphasize the slow pace and fun, to differentiate from
rides that emphasize speed and are less family friendly.
219
For example, when the mapping was looking for healthy food assets, we brought in a bicycle with a trailer filled
with healthy food – which took extra work, but appeared in press photos and more.
271

having more than one nonprofit partner. A good litmus test is whether the mapping falls under
the umbrella of a neighborhood or community campaign that is bigger than any one
organization, and is open to a coalition rather than owned by one group. This way the mapping
can be “about the neighborhood,” rather than just about one organization’s campaign.

PREPARATION:
1. Pick a stable URL in advance, like an event page on Facebook or a group website and keep it up-
to-date. This way your participants will know where to go when they get home to see pictures
and help distribute the group stories. This URL can be your Vojo page (see below), although you
will have more flexibility if you create your own home page that simply links to Vojo.
2. Create your group page on Vojo
220
. For example, http://www.vojo.co/en/groups/ridesouthla
3. Print the instruction hand-out. Practice it once the day before to make sure it’s working!
4. Embed a journalist or two to invite journalists. Creating a media advisory can also help bring in
journalists.
5. Tell participants to bring a picture of home or a home connection to the issue (like veggies in
their fridge for a food justice campaign). This way they’re thinking of it in advance.

DURING THE EVENT:
1. Show the picture stream early on iPads to show that pictures are real, and to give legitimacy to
the efforts of participants

220
Vojo is a community storytelling tool that uses ordinary cellphones to capture pictures and stories, and turn them
into a vibrant group blog online. For details, see http://vojo.co/en/content/about
272

2. Give one simple “mission” for what to map. The narrower the better. Telling them to “map our
walk” is too vague, as is the general issue. Include saying why “each of us must tell our
networks” as part of our civic duty.
3. Make the training mission “real” by mapping which zip codes people come from.
4. Ask participants what they have submitted so far. This builds a sense that it matters. Give people
a chance to share publicly what they’ve captured.
5. OPTIONAL: Host a slideshow of the pictures at a happy hour afterward. We did this at a local
bar and it was incredibly successful. To do this, save a bunch of the pictures from the website to
your computer, and start a slideshow (however your computer does this).

AFTERWARD:
1. Send link to pictures that same day. Remind people of the mission to spread the word to their
networks.

TOWARD A PRINT MAP
1. We recommend a three step process.
1. First identify likely areas for mapping.
2. Second, find the likely format (one page versus a three-foot tall banner are very
different).
3. Third, do your mapping in an iterative way, showing a sketch early on, but keeping it
open for input. If possible, always indicate that the map is an early version. Once a
graphic design is completed with pictures from early mapping, host an event with
residents to “validate the map” via a walk or ride.
273

Appendix 5: Design checklist for making Situated Games
How can designers and strategists implement the findings of this study? This appendix
seeks to articulate how ‘local fit’ could be optimized within the design cycle. The following
checklist is not a guarantee of success, but does illustrate where questions of ‘local fit’ often
arise, and how optimization might translate into design strategy. For the framework, the ability
to articulate a design checklist is a contribution that demonstrates some of the pragmatism of
‘local fit’ for applied design.

1. Identify the place-based community (e.g., neighborhood identity or an alumni network
tied to a particular place). Make sure the place can be described in more detail than its
geographic perimeter. Place-based communities have a shared group identity (often
fragmented), built on narratives of how that place is distinct from its neighbors –
including in terms of distinct history, challenges, opportunities, social modes, and
communication ecology. Consider summarizing historic sites of public protest or
contention, as was done with Re:Activism.
a. Who has the legitimacy to define that place? Defer to community-based
organizations, not just focus groups – and certainly not just the real-estate
marketing of the neighborhood brand. Principles of the communication ecology
have shown that the ability of a community to tell its own story depends on strong
communication flows to local organizations, so look to those flows when listening
for definitions of the place. Games that attempt to shape the place without
274

legitimate roots may inadvertently undermine the ability of the community to self-
organize.

2. Be honest whether your priorities are for a “Situated Game.” The framework of
‘local fit’ only applies for games with the right goals and basic criteria. (For specifics,
see the first chapter section on “Definition and criteria for inclusion.”) For example,
consider:
a. Will your game prioritize the empowerment of existing local communities, not
just individuals, and not just by “creating” community?
b. Will your game seek to affect the real-world, with actions in the game affecting
the broader community beyond the game?
c. Is the activity game-like, beyond having a few game features? For example, is
progress in the game uncertain and with genuine challenge?

3. Clarify who will design and when. The design process inherits existing power relations,
and establishes many others. Participatory design is currently a fringe practice for games,
but increasingly must be considered for games that seek to embed with local
communities. This is tricky – the design process can undermine local autonomy, or
advance community empowerment; of course, bad design helps no one, and design by
committee is rarely successful. To begin, there should be some articulation of how and
when power will be shared, and with whom (e.g., veto authority to specific community
leaders over the goals and game mechanics, with an approved process for input).
275

a. For future scaling, consider inviting a few “friendly spies” periodically into the
design process from another community. Their job is to write the “adaptation
manual” from an outsider’s perspective. Complementing the usual toolkit with
such an outsider view can better guard against over-localizing the design.

4. Study the community in a “design research” mode (optional but recommended). If
design is a form of problem solving, then ‘local fit’ can point to several key questions
that should be asked before diving into design:
a. What are important spaces (especially in public space)? See above for identifying
the place. In addition, the method of community asset mapping described with
ParTour can provide a powerful means to identify local resources from the
bottom-up, building legitimacy for the game with residents and organizations.
b. What organizations and media outlets tell the “group story” for the community
and of its groups? Especially: who articulates shared concerns, history, and a
vision forward? At the least, have a list of local geo-ethnic media, and anchor
organizations. Sometimes it is useful to draw a conceptual map of how such
organizations and outlets relate. For ongoing navigation and design, it is often
best to find a local leader who will not only back the project, but will serve as a
continuous networker across local organizations and government (this was vital
for Macon Money).
c. What are the existing social mechanics? How do newcomers find connections in
the existing community? What are the existing modes of greeting and meeting?
276

Where do they happen online? Where do they happen offline? Are the dominant
modes individual-to-individual, or are there times that individuals address a
group? What daily routines help to bring individuals into contact?

5. Design your game. This is the hard part. Games are incredibly difficult to design and
balance. There are some excellent foundational books available, two of which have been
emphasized in this dissertation (Fullerton et al., 2004; Salen & Zimmerman, 2004). Yet
these texts focus primarily on the deep methods of games; similar expertise is needed on
the design team in methods of community building and engagement, including
community organizing, parade/event planning, etc. Hiring a team that includes people
with a track record in making games and other people in community organizing may be
essential. Below are a few steps that are often useful, and how they might align with
‘local fit’.
a. Brainstorm several game ideas. If you only have one “killer idea” that stands
head-and-shoulders above the others, then your team is probably not thinking
broadly enough. Macon Money and Reality Ends Here only narrowed in after
investigating many possibilities. Having points of comparison, not just a current
favorite, often leads to better design.
b. Are there similar prior games? Identify several similar games or activities akin
to what you are attempting. Do not reinvent the wheel, or pretend that you are the
first. There are always models, and it is vital to be able to articulate what you are
277

building upon. Hopefully the umbrella of “Situated Games” can provide a filter
for appropriate antecedents.
c. Articulate your game’s theory of change (in other words, how the game will
ultimately affect the broader community). Game designers may find it easier to
consult with strategic planning consultants, who can use the language of local
nonprofit planning. This dissertation argues for treating the following steps as
separate and distinct:
i. How does the game strengthen networks? Is the emphasis on creating
new ties, deepening existing relationships, targeting key network nodes,
connecting isolated groups, bridging groups, etc.?
ii. How does the game strengthen the communication ecology? This may
include strengthening the collective identity. See the appendix on “Game
strategies to strengthen the communication ecology.”
iii. What is the broader change (beyond the game)? Simply articulating
your best guess can help optimize the game. Be brutally honest: most
people in the community will not play the game, and in fact the scale of
most Situated Games is far from the magnitude of established services
(e.g., K12 education). Earlier in this chapter, several ways were discussed
in more detail including: (1) games as a rallying call for broad concerns;
(2) games that spur local organizations to affiliate; (3) games as network
catalysts for key individuals; (4) games as socio-economic policy.
278

d. Optimize for ‘local fit’. The three dimensions of fit are the like the legs on a
chair, where each leg is needed to stabilize the game on local terrain.
i. Mechanics: Articulate the “game mechanics” in terms that emphasize
social interaction, especially the meeting and greeting that builds
networks. For a detailed description, see the two-step process in Table 5.
Make sure your game gives feedback on progress toward the social impact
goals that motivate many players; this may include collecting email
addresses and notifying all players of a “post-game impact” evaluation.
ii. Group identity: Describe how the game will perform and embody local
narratives of group identity. Seek to improve the fit between what the
game performs and the identity the community has – or seeks to have.
iii. Communication ecology: First consider the existing community modes on
three levels.
221
Describe how communication will happen in the game,
including between players, with non-players, between players and local
institutions, and between players and local media. Each of these
“channels” of communication can be vital. If the community has a strong
existing channel in place, the game should consider using it; over reliance
on weak channels is likely to undermine participation.

221
The three channels are: (a) between participants, where talk is direct (including physical and digital spaces),
emphasizing actual use patterns, not simply access; (b) to community organizations, where participants hear from
local groups that are shaping the collective conversation, such as local churches, libraries, health centers, schools,
sports leagues, and business clubs; (c) facilitated by local media, especially where there is some broadcast power
and legitimacy, as with local TV news, group pages on social media (e.g., Facebook), and geo-ethnic media. Local
media occasionally bridges to national broadcast media. This list is introduced in the analysis of Reality Ends Here
in the section on “Fit #3: to the local ecology of communication.”
279

e. Iterate and simulate. Given the complexity, no design will be complete on a
first pass. In addition to the usual techniques of game design (e.g., paper
prototyping), the design process should be considered a form of strategic planning
for the host organizations (Stokes, 2012), where each round of design helps them
to articulate their strengths and weaknesses, as well as the design’s specific
opportunities and threats.
i. Validate by pitching a different community. Rather than simply asking
if the game works for current users (which promotes replication),
designers might want to pitch community organizations that are likely to
host the game in the future (which shifts the emphasis to adaptation).
Organizations often have a hard time articulating what they want or need,
so be cautious; often the most valuable input is their warnings about the
limitations of their own community and how the game might possibly fail.

6. Operate your game. Situated Games require careful monitoring at the community level,
and much of the “design” is fine-tuned live. Formative assessment is essential.
a. Facilitate. The designers of Reality Ends Here emphasize the fundamental
importance of the game facilitator to set the tone for play and meaningful
engagement. Hiring the right facilitator may be as important as hiring the right
designer.
b. Optimize for ‘local fit’. Similarly, much of the optimizing for ‘local fit’ may
emerge during the operation of the game. See above for the dimensions.
280

c. Consider network analysis for live feedback. As demonstrated with Macon
Money, network analysis can provide guidance for organizers. For example,
analysis can identify key leaders who bridge between groups, and track how the
network grows denser over time.

7. If successful, go beyond replication toolkits. The premise of ‘local fit’ is that Situated
Games cannot be bluntly generalized; they must be reconsidered and adapted for each
localization attempt. Too often, toolkits assume a replication model of scale, and focus
too heavily on localization in terms of content. At the least, toolkits should begin to
speculate on the fidelity criteria that may help guide the adaptation process.
Additionally, toolkits can use the framework of ‘local fit’ to recommend what kinds of
communities are likely to be compatible.
222
More ambitiously, ‘local fit’ can be used to
help make the design more robust to adaptation by reflecting explicitly on variance and
dependencies in each of the three dimensions. Keep in mind that when a game is made
entirely predictable, it has likely lost much of the flexibility needed to fit deeply into local
communities, and thus may actually be flattening the distinctions of place.

222
Contexts can be compared in each of the three dimensions: (1) Are there different local modes of greeting and
meeting? (2) Is there a different local identity or community brand? (3) Are there different local modes of
communication – either between residents, with the local media, or with community-based organizations? These
questions are demonstrated in the analysis of Xposure.
281

Appendix 6: Game strategies to strengthen the communication ecology
One way that games can empower communities is to strengthen the communication
ecology. Taking a communication lens, this section considers how game design can build
capacity according to Communication Infrastructure Theory (Kim & Ball-Rokeach, 2006a). The
strategies below are not comprehensive, but help illustrate how the findings of this study can be
articulated in pragmatic terms.

1) Foster communicative relations (mechanics/fit#1): At the most basic level, networks
form when individuals meet and deepen their relations. Strategies discussed in this study
include: helping individuals to (a) meet peers and form social ties, such as between
neighbors who might begin to greet one another on the street; (b) to associate and form
small groups, or even deepen coalitions of organizations; (c) to form relationships with
journalists and media organizations for sharing “community stories” and building the
collective identity for a broader audience.

2) Shift or deepen the collective identity/story (fit#2): This study emphasizes the importance
of “community stories” that reveal the group’s collective strength, can broaden the sense
of community belonging, or can speak directly to the collective identity. The activities of
the game can help (a) identify existing stories for retelling and crystalizing; or (b)
perform a collective story that enacts the group identity on a public stage.

282

3) Build the infrastructure of community storytelling (fit #3): Social connection depends on
healthy flows of communication. Games can help by building new discursive spaces
223
:
• Creating discursive spaces for interpersonal communication, either in physical or
online spaces, including gathering points in public space or online neighborhood
message boards (what the Metamorphosis Project calls ‘hot spots’); additionally,
increasing access to such spaces
• Creating new local media channels, or turning existing channels to focus on local
concerns; additionally, increasing access to such channels
• Aligning and bridging media channels, especially across ethnic media, or those
targeting a narrow age or other demographic band
• Connecting channels across organizations and outlets, especially involving
community-based organizations to and local media outlets (geo-ethnic media)

223
This list comes from ongoing conversations with the Metamorphosis team led by Sandra Ball-Rokeach on how
the theory of Communication Infrastructure can contribute to community change.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The patterns, effects and evolution of player social networks in online gaming communities

PDF

Network influences in health initiatives: multimedia games for youth in Peru

PDF

Civic expression in Little Tokyo: how art and culture empowers communities and transforms public participation

PDF

Game recognize game: performative archives and alternate reality games

PDF

Virtual worlds as contact zones: development, localization, and intergroup communication in MMORPGs

PDF

Understanding alliance building and communication across multiple technology platforms among California's immigrant-serving NGOs

PDF

Adaptive agents on evolving network: an evolutionary game theory approach

PDF

Food is the medium: food movements, social justice and the communication ecology approach

PDF

Ecology and network evolution in online innovation contest crowdsourcing

PDF

The good, the bad, and the longitudinal: testing dynamic prosocial and toxic behaviors in online commercial games

PDF

The evolution of scientific collaboration networks

PDF

Remote exploration with robotic networks: queue-aware autonomy and collaborative localization

PDF

Upvoting the news: breaking news aggregation, crowd collaboration, and algorithm-driven attention on reddit.com

PDF

The evolution of multilevel organizational networks in an online gaming community

PDF

From worldbuilding to worldblending: transmedia rhetoric, identification, and immersion in storyworlds

PDF

Effects of economic incentives on creative project-based networks: communication, collaboration and change in the American film industry, 1998-2010

PDF

Avoiding middle-class planning 2.0: media arts and the future of urban planning

PDF

Infrastructures of the imagination: building new worlds in media, art, & design

PDF

Transformative learning in the public sphere: the educational dimensions of a broad-based community organizing movement

PDF

Learning distributed representations from network data and human navigation

Asset Metadata

Creator Stokes, Benjamin (author)

Core Title Civic games with 'local fit': embedding with real‐world neighborhoods and place‐based networks

School Annenberg School for Communication

Degree Doctor of Philosophy

Degree Program Communication

Publication Date 07/14/2014

Defense Date 06/02/2014

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag civic engagement,community,design,ecology,empowerment,games,games for change,gamification,identity,local,mechanics,network,OAI-PMH Harvest,place‐based,real‐world,scaling,serious games,social capital,social justice

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bar, Francois (committee chair), Jenkins, Henry (committee chair), Ball-Rokeach, Sandra J. (committee member), Fullerton, Tracy (committee member)

Creator Email bgs@benjaminstokes.net,bstokes@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-439302

Unique identifier UC11287592

Identifier etd-StokesBenj-2676.pdf (filename),usctheses-c3-439302 (legacy record id)

Legacy Identifier etd-StokesBenj-2676.pdf

Dmrecord 439302

Document Type Dissertation

Format application/pdf (imt)

Rights Stokes, Benjamin

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA