1 Gaming My Community: Kids designing augmented reality games Abstract: What happens when teens design Augmented Reality (AR) games about their neighborhood? This paper presents a cross-case analysis of projects aimed at providing teenaged boys and girls the opportunity to learn about, design, and implement location-based games and tours of their neighborhood. The paper explores a series of AR game design experiments in diverse settings that guided teenage boys and girls using digital cameras, GPS-enabled handheld computers, and their own perspectives and stories, to document their community and create games. The paper describes the evolution of the design-process that guided the creation of the games, including the incorporation of the kids' own ideas; examines the logistical and technology-related challenges faced in each setting; and analyzes the evolution of the design approaches overall in order to come up with ideas on how to proceed in the future. Objectives... 1 Theoretical Framework... 3 Methods... 4 Data: Student AR game design experiments... 5 Design Experiment 1: Greenbush History... 5 Design Experiment 2: A Greenbush Story... 7 Design Experiment 3: Wild Moose and Mitchville... 8 Design Experiment 4: Nature Hill Design Experiment 5: Virtual Tree Tour Design Experiment 6: Game Unit Design Experiment 7: State Street Design Discussion Suggestions for Future Design Workshops Conclusion Bibliography Objectives The worlds we experience, whether real or virtual, are important to us because they hold meaningful stories, whether these worlds are real or virtual. In popular video games, players are active participants in making sense of and giving meaning to the virtual worlds they inhabit in order to navigate them. Consider a game where players physically move through real geocultural environments in order to solve problems and reach self-determined goals. One genre of games that has the potential to provide this type of experience is Augmented Reality (AR) simulation games. Augmented Reality (AR) simulation games are games played in the real world, in locations such as neighborhoods, historical sites, or watersheds, that use computer technologies to layer data over the real world in order to provide contexts or stories through which players' actions take on meaning. The layered data might include video, text, or images, which designers manipulate to create fictional characters, events, and indeed entire worlds. Designers of these Citation: Martin, J., Jan, M., Mathews, J. Holden, C. (2008). Gaming My Community: Kids Designing Local Video Games. Presented at American Educational Research Association Annual Meeting. New York, March 24-29, Available online
2 Gaming My Community - 2 games can tie specific information to time and space, so that when a player arrives at a particular location, like a statue, s/he can be presented information on the sculptor, the history of the statue, or even an historical picture of the landscape before the statue was constructed. Whereas some approaches to augmented reality use head-mounted displays to layer 3D images over the real world environment, the approach discussed in this paper uses handheld computers to provide relatively low-resolution information tied to specific places. Other than the outdoor, GPS-assisted gameplay noted in the introduction, some key features of these AR games have been: overarching narrative that gives meaning and direction to the players actions, virtual characters or Non-Playing Characters (NPCs) who express their perspectives on the place and its issues, and player identities or roles that give players differential access to content and carry intimations of the abilities of professions (e.g. doctors can take blood samples). In part, the narrative of the game is propelled by an introductory text (and possibly video). In addition, it unfolds through the actions of the player and the NPCs. The NPCs also function as delivery vehicles for other artifacts such as historical photographs, videos, and newspaper documents, and the way they interact with players is one of the chief means of developing player identity. In order to give an authentic voice to the NPCs and to endow them with at least information plausible to that place, we have found it necessary to do considerable research into the location, its use, and its history. This involves taking photos and video on location, observing people as they use the place, reading historical accounts, talking with current experts, and reflecting on common, current themes (e.g irrigation runoff, urban condo development) and their instantiation in local place. The game comes from a process of reflecting and refining this research into an experience in line with what the editing tool can offer. It authentically represents that place through several perspectives, and gives players a deep but manageable experience of the place. As a cultural work, an AR game reveals how its participants/designers incorporate existing community values while simultaneously reflecting cultural models of newer members. Within CHAT, Lave & Wenger (1991) look at Communities of Practice and assert that identity construction is closely tied to the activities a person is engaged in as bounded by the cultures of the groups that person is in. In other words, our existence is grounded in communities. What we like and dislike, our opinions and deepest heartfelt values, are based to a large extent on the activities and practices of those who came before us and those who currently live and interact with us. And we share these values, as Jerome Bruner (1996) notes, "All viable cultures make provisions for conserving and passing on their 'works,'" (p. 24). Using culturalism, Bruner describes the informal interchange between institutions and individuals in the transmission and modification of culture. Using these views of cultural transmission, this paper presents a cross-case analysis of design experiments (Brown, 1992) carried out by different members of the Local Games Lab in Madison, Wisconsin, that were aimed at providing boys and girls the opportunity to learn about, design, and implement location-based games and tours of their neighborhoods and communities. It describes the evolution of the design-process that guided the creation of the games, examines the logistical and technology-related challenges faced in each setting, and presents suggestions for further design experiments.
3 Gaming My Community - 3 Theoretical Framework We need to increase our connection to our physical places. Gruenewald (2003) calls for a reinvestment in place-placed pedagogies "so that the education of citizens might have some direct bearing on the wellbeing of the social and ecological places people actually inhabit" (p. 3). He is referring to embodied phenomena "that by their very nature occur in real time and real space" (Dourish 2001:101). Places are spaces that hold meaningful embodied stories that shape communities and identities, and so are important motivators. School and classroom design has evolved along with public education. Many in use today were built to be energy-efficient and flexible spaces. No longer do classrooms have huge banks of windows reminding students that the building exists in a larger place, but rather many are designed with little connection to the outside world, with one or two small windows that block out the distractions outside. Ellsworth (2005) examines museums, monuments, and other pedagogically charged environments outside of the classroom, maintaining that these open our aesthetic to teaching and learning in ways "largely unexplored by the official literature of educational research" (p. 9) She argues that these spaces that "speak to and about pedagogy indirectly through design -- a means that reaches beyond the limiting scope of language" (p. 10). The "processural paths" through mediated environments offer new pedagogies of sensation in our experiences -- not "as having bodies" but "as bodies whose movements and sensations are crucial to our understandings" (p. 27 emphasis original). A problem in creating powerful pedagogical places, Lemke (2005) notes, is the expense. He looks instead to video games and digital space. Jenkins & Squire (2002) argue that video game designers are doing exactly what Ellsworth is talking about, albeit with a different degree of physicality: Game worlds are totally constructed environments. Everything there was put on the screen for some purpose -- shaping the game play or contributing to the mood and atmosphere or encouraging performance, playfulness, competition, or collaboration. If games tell stories, they do so by organizing spatial features. If games stage combat, then players learn to scan their environments for competitive advantages. Game designers create immersive worlds with embedded rules and relationships among objects that enable dynamic experiences. (Jenkins and Squire, 2002: 65) There is a connection between place-based activity and sociocultural activity-based learning theories such as Activity Theory (Vygotsky 1978; Leon'tev 1979; Engeström 1990, Wertsch 1998), Situated Action (Suchman 1987, Lave 1988), Distributed Cognition (Flor and Hutchins 1991, Pea 1993), and Cultural-Historical-Activity-Theory (CHAT). Specifically, this connection links individual elements of thinking and activity or experience, with participation in a community (Dewey, 1910; Lave & Wenger, 1991; Leont'ev, 1978; Vygotsky, 1978; Wertsch, 1988). But communities -- even communities of practice -- often exist due to and take into serious account, the various elements of place. Our physical setting, whether designed or merely repurposed, also serves as a tool or mode of representation. In Hutchins' (1995) seminal study of the navigation of a ship, he demonstrated how socially situated cognition works in a system of distributed beings and tools; however, the role of space in that study is often overlooked: the system he examined was for navigating a ship through geographical space, and the physical places of the participants on the ship had much to do with options and possibilities available in those places -- what Norman (1993) refers to as affordances. Flor and Hutchins (1991) identify environment as a location of knowledge, and
4 Gaming My Community - 4 while Pea (1993) focuses on tools and modes of representation, physical setting plays a significant role in his research. Along these same lines, AR games put the focus directly on relationship between place and activity. Methods The paper draws together preliminary results from a series of independently undertaken research projects by members of the Local Games Lab that examine how kids design augmented reality games. The projects, the first of which started in 2005, engage students in designing historical, environmental, cultural, or other games and tours of the neighborhoods and communities where they learn. Many include a focus on developing students' technology, historical inquiry, and presentation skills. Moving from researcher-created games to studentcreated games, this design project stems from several years experience designing AR games around the context of local places. Since 2004, the authors have been working together as part of a U.S. Department of Education STAR Schools grant at the Local Games Lab designing and implementing these games and encompassing two-week curricula in middle school classrooms. Nearly a thousand students across six school districts have played through our designs. Throughout this time, we have also been individually and collectively looking for ways to involve teachers and students in the design of these games with the goal of advancing the technology and our ways of thinking about it as a design tool for individual participation and expression in communities, and as a tool for reflection to foster an increased sense of place. These are not, however, the stated goals of the research projects that we are funded to do. Consequently, much of the exploration we have done in this area has been on our own time as volunteers. A number of these experiments included the participation of more than one of us (Greenbush History, Nature Hill, and Game Unit and State Street), which resulted in a great deal of cross-pollination. As a result, the successes and failure of one project have informed the next. Up until this paper, however, we have not done a formal analysis of what our collective experiences have taught us. To write this paper, we have gathered together three years of research notes and memories of six to eight side projects (depending on how they are broken up) on designing AR games with students. We discussed and analyzed them separately and together, with close consideration of special constraints, challenges, and advantages faced by researchers, teachers and students looking to create AR games in each setting. We revisited what approaches worked and did not seem to work for each group and setting and developed strategies for streamlining the design process. This paper presents the stories of our experiences working with young people, as they tried to design games for the places in their lives. In telling these stories, we drew on video artifacts, students' production journals, individual and focus group interviews, observations, students oral and written analysis and critiques. Throughout the experiments, many of which were lead by one or two of the authors on their own, we compared and discussed experiences, sharing what worked and what did not, looking for similarities and differences in approaches, experiences, and results. We examined to what extent students develop new technology, research, and media production skills through the process of designing their games? To what extent does the goal of design enable students to research place? Are there changes in the way that the students view and/or talk about their neighborhood? What challenges or constraints appear as the project
5 Gaming My Community - 5 unfolds? How do these interfere with the design process? How are these challenges addressed? Data: Student AR game design experiments Design Experiment 1: Greenbush History The most engaged I've seen my students this year was when they were walking in small groups around the UW campus and Lake Mendota shore, holding handheld computers with GPS devices that provided data at various sites on our map, constantly challenging them to collaborate to solve a public health mystery. -- Fifth grade teacher. In the fall of 2005, eighteen fourth-graders from a public school played Mad City Mystery, an Augmented Reality Game on Handheld Computers (Jan, M. & Squire, K., 2004). Designed with High School Students in mind, Mad City Mystery was difficult even for the best readers in the class, hence we invited parents and graduate students to play with these students. After the game their teacher immediately asked if we could help him transform his class into an AR game lab, where students' would turn their research of local culture and place into Augmented Reality games. Creating a game for a classroom as we did in Mad City Mystery is not as daunting as turning the class into a game lab. We had to start from scratch with just about everything. What is a game design lab? What role would students and teachers play? How is a design lab different from a traditional classroom? What would students need to do in order to acquire the expertise of a game designer? Is being a game designer an important way of learning? Is it a good way of learning? What might students learn as designers? Through these fifth-graders we began our journey of deep reflection. Turning their research into games was intriguing to many kids, especially the boys, since many of them were avid gamers. However, an augmented reality game is not a game that can be designed the same way as many commercially available video games. We knew that we had to introduce the affordances (Norman, 1993) of Augmented Reality games to the classroom so that students would be more practical about their game designs. Additionally, we needed to help the students figure out a way to map the research they did in the Greenbush neighborhood onto an augmented reality game. There were some constraints these students had to live with. For example, there must be roles for the game. Students not only needed to think about what roles they wanted for their players, but also why these roles were appropriate. Without a functional classroom computers to work on, we focused on strategies and activities that would work -- discussing possible scenarios of the game, taking field trips and interviewing people to collect data. The students did a thorough job of research and mapping, having had collected dozens of stories and characters, and were facing the difficult task of sifting and winnowing their deep understanding of the neighborhood into an AR game narrative that could be ported into the current game editor and played in less than three hours. Meetings began to get heated as students, the teacher, and the researchers negotiated how the game should look. In an attept to move the project forward the researchers turned their focus away from which content to include, and instead tried to guide discussion and energy on shaping any potential game content into pieces that would actually fit in the game. In essence, we wanted to put off until later, discussions on the game narrative, which we figured would be settled by those
6 Gaming My Community - 6 students who contributed most energetically to the process. In order to maintain a continued connection between the neighborhood where the stories occurred and the game, we created game design scaffold sheets (figure 1.) with past and current maps represented as they might appear on the handheld computer. Students were asked to write from the perspective of one of the three game roles they had selected earlier (Ethnographer, Urban Planner, or Historian), as well as one of the themes they had been exploring in the neighborhood. Ideally, the stories were to include a game quest designed to propel game play. In one design session, students were asked to write word stories they had had heard and felt must be included in the game. Here are some examples of the narrative events they felt should be built into the game: CDA apartments: (media: photo) At one of the apartments, there is a man called Rob. He s a deadhead, he lives to question authority and causes mischief. No one ever would guess that he was legally blind just by looking at him, and there s why he s part of the 90% of all CDA residents who are disabled. Eugene Parks House at 09 Regent: (media: photo of him and house) Eugene Parks is a black man who ran for mayor, but lost. He was active in local politics, and did a lot for the community. Fraboni s on Regent: (media: go inside instead) Not only does Fraboni s sell Italian groceries, but they also have a deli and sandwich bar, as well as things to assist you in cooking - sets of cutlery, cooking tools, cookbooks. A homey smell of delicious food hits you as soon as you walk in the door; homemade pizza crust, falafel, the deli, and much more. Gravasa Family: (media: photo of Gravasas house) The Gravasas were bootleggers people who sold or made alcohol during Prohibition. One day two policemen came in and asked the son where the secret room was where he stashed the alcohol. The kid wouldn't tell. But then the police offered him 50 cents, so the kid turned in his father for 50 cents. Figure 1. The game design scaffold sheets with past and current maps represented as they might appear on the handheld computer.
7 Gaming My Community - 7 At the last game design session of the school year, there was a strong expression of discontent by the fifth graders that too much time had been spent negotiating what the game should look like. This was a classic 'too many cooks' dilemma. Individual kids promoted their own vision of the game, the teacher felt he had promised a certain vision of the game to the community members that he had to deliver on, and the researchers' -- knowing what was and was not practical with the game editing software -- tried to steer discussion towards what was practical. The end result was that by summer vacation we still had about thirty "must-have" stories and people -- about ten more than was reasonable for a game. Design Experiment 2: A Greenbush Story At the end of the spring 2006 experience with the fifth grade class, possessing deep knowledge of the Greenbush neighborhood, and a solid understanding of the concepts of a game design, three students indicated that they wanted to continue work on the game to see it to completion. In the fall we invited them to design the game as they envisioned it in a weekly after school game design workshop where they interacted directly with the game editing software. Of the three, one had schedule conflicts, and another dropped out after three weeks for unspecified reasons, leaving the third, Max, to finish his own version of the game. To begin, the researcher put a blank map on the PDA, and he and Max took a walk around the neighborhood to re-familiarize with it and calibrate the map. Back at the computer, Max got a short tutorial on using the game editor, and began exploring it, asking questions as issues appeared. The researcher's position was to let Max direct the game and decide what was and wasn't possible based on his own exploration of the game software, and to offer as much support as possible, answering questions and acting as a sounding board for ideas. In order to let Max spend as much time as possible working directly on the game creation, the researcher prepared (e.g., scanned, resized, uploaded) photos, maps, and other graphics that Max selected. Max's experiences with the game editor were mixed. He was often initially frustrated when he encountered bugs and interface inconsistencies but usually learned tricks to get around the problems, or avoided altogether the problematic parts of the editor. By February he had a game that was situated in 1958, in a pre-urban renewal Greenbush neighborhood, where the player was a 12-year old Jewish boy (Jews, African-Americans, and Italians being the predominant groups living there) who was collecting signatures for a petition to stop the city from bull-dozing the neighborhood while running his Sunday errands. One of the unique features of the game is that the center triangle of the neighborhood was bulldozed to the extent that streets that had existed in 1958 no longer exist, except in the memory of displaced residents and on old maps. Old plat maps (figure 2.) are used in the game along with photographs and information about buildings that no longer exist. The game narrative he created directs players to these places, providing them a taste of the Greenbush that once existed.
8 Gaming My Community - 8 Figure 2. Compilation map before urban renewal (used in game) and map of current area. A number of the current sidewalks (not shown) follow original streets. As an experiment on student design, this was moderately successful. We had hoped for more students, but feel that the summer break interrupted their enthusiasm for the project. On the other hand, the experiment demonstrated the potential of AR games as a medium for sharing stories about places. Max drew on his Jewish identity and modeled the role and quests based on interviews with a Jewish resident of the neighborhood who was his age in By keeping the game narrative simple (no branching narratives, etc.) the opportunities for complications were kept to a minimum. In the end the game was tested by the oldest players to play AR games -- residents of the Greenbush who were displaced during the 1960's urban renewal program. Design Experiment 3: Wild Moose and Mitchville From 2005 to 2007, as a doctoral research project to tie the computer as a design tool to adventure learning, an experiment in AR game design and play was implemented at a deep woods camp for boys. The vision was to have camper-created games structure a four-day hiking trip in a sixteen square mile Maine wilderness preserve. The challenges were significant -- no AR games had yet been designed or tested for such remote conditions, or for games lasting more than a few hours. Participants over the three summers were comprised of twenty year-old boys and five year-old men, as five distinct groups ranging from three to six members, each including a counselor. Most were middle- to upper-middle class Caucasians, whose parents paid ~$2500- $5500 for their son's 2-7 week deep woods camp experience, although a few attended at steep discount through a camp scholarship fund. The game software developers had promised a stable version by March 2005, but the software was not delivered until May. Initial testing by the group in Madison revealed that it was extremely difficult to create a game using the game editor, and many bugs emerged that
9 Gaming My Community - 9 frequently crashed the game engine, making extended play nearly impossible. A July update to the software fixed many of the problems, but it became clear early on that the technology was not yet ripe for use at a camp where the nearest electricity and Internet access was miles away. Instead, the researcher focused on creating a camper-designed place-based story that could be ported into a game when the technology matured. In 2005, the researcher approached a group of five experienced campers with this idea, handed them a GPS unit and notebook, and gave them the task of creating a game. They spent a four-day hiking trip mapping out potential game space, creating potential characters, and building a loose game narrative for the ARGH. Because they had not played an AR game, they did not fully understand what was possible or even what to do. Their notes were used to create Wild Moose (Martin, 2005). In 2006, a group started their trip by following the Wild Moose game narrative, and had a full day of experience with the AR game in heavy rain before the Bluetooth GPS unit shorted out in the unprotected pocket of a soggy eleven year old. The remainder of the four-day game had to be played in manual mode (without GPS triggering of events). Fortunately, the trip was lead by an enthusiastic Australian counselor, and with his lead, and a good understanding of the constraints and affordances of the game, the campers wrote a completely new game narrative called Mitchville: Where the War Began. Their design was based loosely on the ideas behind the 1984 movie Red Dawn, and John Marsden's (1995) Tomorrow, When The War Began. By the time the replacement GPS arrived, the researcher had ported the new narrative into a game and successfully tested the initial trigger point. The starting point was familiar to participants -- the base of a trail that they had all hiked. The game narrative unfolded message by message throughout the trip. At the first trigger spot, the handheld computer became a Star Trek style communicator and displayed an urgent "live message" from the assistant director of the camp (this paper's first author) announcing that camp had been taken over, and instructing them to get to the top of the mountain and await further contact. In order to get around the fact that the handheld computer was not a communication device, the narrative explained that the assistant director's communicator was damaged in the melee, thus making one-way communication the only option. As the trip progressed, they got more clues that helped them reconstruct the "real-time" story of what happened at camp -- and better understood how they could help. The narrative revealed that a rival camp for rich kids took over the camp to use the lake's "crystal clear water" for a Grey Poupon mustard factory so they could have tastier mustard on their sandwiches. The spring-fed lake is a big source of pride, so participants enjoyed and joked about the plot. "Saving" the camp and thwarting the invaders' plans involved climbing three mountains to triangulate and decode a radio signal from a transmitter that the invaders were using to communicate with their minions. (The logic of how triangulation of a radio signal helps in decoding it is unclear, and provided humorous conversational fodder among participants.) A second group in 2006 was able to play the game for two days, but a software crash on the morning of the third day disabled the computer. However, because the counselor had run through the game in manual mode, he was able to lead the group through the basic ideas of the game for the rest of the trip. The researcher was set to personally lead the third group in 2006, but was called away for a camp emergency, and missed the first six hours of the trip. In that time, the
10 Gaming My Community - 10 handheld computer was repacked too tightly into its protective case with the backlight button pressed, effectively draining most of the battery. Again, because the researcher knew the narrative well, they used the handheld computer sparingly, relying on the sturdier standard GPS. In 2007, a new model of handheld computer with integrated GPS helped solve the problem of protecting the hardware from the elements, and initial tests of the battery life promised good performance. However, heavy usage of the handheld computer (backlight on) by the counselor on the night before the first trip, and by the campers the first night again depleted the battery half way through the trip. On the second trip, the counselor was given three fully charged handheld computers with the game loaded, and was trained beforehand on how to swap them out in case they died. The first lasted two and a half days before the battery died. The second lasted six hours before the game engine crashed and corrupted itself. And the third was able to guide them through to the end of the trip. One of the biggest challenges in the research was working with the game hardware and software. In the original vision of the research, the game was to be revised somewhat each trip, so an evolving and constantly 'fresh' game could be played. The harsh realities of the research conditions continued to fight against that idea. From issues of buggy software to the difficulty of getting to town to download new versions, drivers, and work-around ideas, to simply having time to update, troubleshoot, and solve problems -- it was clear early on that goals would need to be more modest. Additionally, other than the first revision, it was difficult to get the players to come up with, and write down, new ideas for the game. Counselors and campers both reported that they talked about ideas as they played the game, but what they recorded and reported back were most often ideas on the user interface and content format (e.g. more video, make it 3-D, etc.) rather than suggestions to drive the narrative forward. This could be because the first group did such a good job, as the following comments suggest. Camper 1: I liked the espionage. Actually that was really fun.... we have no evidence to actually disprove the game, so because of that, it was kind of fun. Camper 2: you actually have imagination because it says things that are going on that aren't really going on, so it's just kind of neat like that. So you can imagine what's happening in the game instead of just hiking. Camper 3: On a regular trip you just want to get to the next campsite, but on this you have to get to this mountain to stop the radio signals then you have to go to this one and that one.... so you could finish the game and see what happens next. Alternatively it could be because the campers did not want to challenge the authors of the first revision. Counselor 1: [The first 2006] group came up with that whole story and that was really good. That was pretty impressive, I don't know if I could top that. It could also very easily be because writing is a school activity, and they were at camp -- not school. Regardless of the reasons for not coming up with as many new ideas as we had expected, the technological realities of the camp environment also did not allow a weekly or even biweekly game revision. However, as an interesting side note, multiple members of every trip reported diligently adding waypoints to the GPS device in order to further map out the land for future trips. Perhaps if the games could be revised and edited on the trip like a mobile Wikipedia page, the engagement in redesign would be enhanced. The next version of the game software (v.5.1) will
11 Gaming My Community - 11 begin to explore this possibility. Design Experiment 4: Nature Hill In spring of 2007, while running the STAR Schools AR game curriculum Sick at South Shore Beach (Mathews et al., 2006) with a middle school in a Wisconsin city of ~14,000 people, teachers there expressed interest in creating an AR game for "Nature Hill" -- an outdoor classroom area where a new middle school was being built. This design experiment, which was integrated into a technology elective course, took the form of three 50-minute workshops and a two-hour site visit that followed the playing of Sick at South Shore Beach. Before the kids began designing their AR games, the researchers came to the classroom and discussed the basic affordances and constraints of the AR technologies. In this first design session, with the teacher leading, they talked about what sorts of game narratives could be created and used in the space -- a wooded hill and outdoor science classroom/city preserve separated from a field by a regionally prime trout stream. Most of the students in the class had visited the game space on one or more occasions for recreational activities, and were thus familiar with the area. In the course of the meeting, students identified several conflicts of interest impacting the area, and recognized it as a contested space. To prepare for the second meeting, we created a sample game outline worksheet (figure 3.) and a blank version for the students to fill in with their own game narrative ideas. These sheets helped students connect their text to specific locations on the game map (locations not indicated on sample). In the second design session, students worked individually and in groups to brainstorm game narratives. While there were almost twenty students in the class, there were only a handful of different ideas written down; however, the teacher reported that for this "challenging" group of students, she was impressed with the level of their engagement. Figure 3. On the left, a filled out sample of the AR game design scaffold sheet; the map was based on the teacher's chalkboard drawing of the area in the first design session. On the right, structuring the game problem in terms of quests and tasks needed to fulfill the quests.
12 Gaming My Community - 12 The following week the class played Sick at South Shore Beach in Milwaukee for two hours, and on the bus-ride back they stopped for another two hours to explore the Nature Hill game-site and record their ideas and concerns about the property. Digital cameras were distributed, and students took roughly a hundred photographs of the area to potentially use in their game. The photos were printed out and posted for use in the third design session, where students arranged them on the map and wrote descriptions that helped them construct their narrative. Because of time constraints and the complexity of understanding the steps needed to design an AR game, very few students got to the point of structuring their games in terms of quests and tasks. Additionally, this was the last class period allotted for the process, so any additional game design needed to be done outside of class. The teacher reported that one student did finish a game design over her summer vacation, and indicated that it was good, but we have seen it yet. Design Experiment 5: Virtual Tree Tour In the spring and fall of 2007, Peter, an eighth-grader at a private school, whose teacher was working with us in designing an AR game to teach science, indicated interest in designing an independent project. Peter also received a grant to buy the equipment needed to play the game. The grant required that he design the Virtual Tree Tour and demonstrate it in public. As researchers, we were interested in (1) knowing the degree to which a middle school student could handle the technology, (2) piloting a workshop that engages middle school students as designers as a way of learning a particular subject, and (3) understanding what kinds of learning take place in the design process. Our goal was to design a workshop that encouraged Peter to generate knowledge by asking important questions and conducting his own research. In this case, Peter was no longer a student who learned about trees from textbooks or the Internet. He was a designer who would need to understand: 1) what is important in a tree tour? 2) who is the audience, and what kind of a tree tour is appropriate for them? and, 3) how can I best engage my audience? In order to facilitate this, the design workshop included three major themes: 1. Tools and learning: Designing an augmented reality tree tour requires understanding the game engine and game editor and how they relate to each other. As a result, we spent a big chunk of time messing around with the game engine and editor. The technological affordances were the focus of discussion when the context was appropriate. 2. Envisioning learners/users: Since the game requires producing specific knowledge about trees for specific learners, we spent a lot of time taking about what learners (in this case, Peter s middle school friends) need they need to know, and why. 3. Envisioning learners in games: Designing a tree tour also requires mapping between the functions provided by the game system and the intended learning goals conceptualized by the designer. This component was significantly harder than the other components. The workshop ran from April through November on a weekly basis, except for the two months when Peter attended pre-arranged summer activities. In each meeting, we revisited the above themes. By September we finished a tree tour game and was working on a more sophisticated revision when he decided in November that there was too much work in school to continue working on this project. In the early stage of this project a tree expert gave Peter an extensive tree tour, highlighting
13 Gaming My Community - 13 specific trees, such as burr oaks, which had been here for hundreds of years. The tour not only included characteristics of these trees, but also histories of the local place, positioning the trees as time channels to the past. Peter was enchanted by the tour, and envisioned a virtual tour that brought his friends to the same trees with stories similar to the ones the expert unveiled. We revisited these trees many times to discuss different ways of engaging players in the tour. Peter was able to identify the major characteristics of the trees he had visited and their Latin names. However, representing this information in ways that could engage his players seemed to be a hard task for Peter. As time went on, the researcher noticed that he lost his original vision for the game, and seemed to present trees in a straightforward and linear fashion, as many textbooks depict a subject. With researcher intervention, he modified the textbook approach by asking players to collect fallen tree leaves or identify bugs or birds that relied on specific trees. In the end game, players visited eight different trees and talked to two virtual people. The tree tour highlighted some of the more important features of these trees, and took the form of a miniversion of the tree expert's tour. Peter presented the Virtual Tree Tour at a school fair where most visitors were impressed by his design. Visitors were intrigued by his introduction of the technology and his ability to use the technology to design a virtual tree tour. This project sheds some light on the three issues we fore grounded in the inquiry. Peter was able to understand the technology much better than we had expected. He demonstrated great understanding of the game editor and was able to think about these editing features in terms of how they would affect game play. For example, in order to avoid confusion in game, he employed the anti-trigger feature to make the visited trees invisible so that players would look for trees not visited yet. He was also able to relate trees in personally meaningful ways. He spoke often about the relationship between trees and humans. On the other hand, the length of the design cycle made it difficult for Peter stay motivated. In addition, Peter was satisfied with his first version of the game and did not challenge himself to improve the game based on user feedback. Design Experiment 6: Game Unit In fall of 2007, similar to the Greenbush History example above, a design workshop emerged from students' interest after playing an AR game as part of a school curricular unit. In this case, ten students from a small alternative high school participated in a six-hour design workshop after playing Dow Day (Mathews, 2005), a historical simulation aimed at developing students' historical thinking skills. While the design workshop focused on teaching students how to use the tools needed to create AR games, many individual and small group discussions emerged around topics such as interactive storytelling (e.g., how can you use branching narrative to propel a story?), game mechanics (e.g., what is a game? how do you maintain game flow? etc.), and the redesign of the editor and game engine. The workshop started with a brief discussion related to the students' experiences playing Dow Day. The researchers asked students to share what they liked and disliked about the game and helped them brainstorm a list of changes they would like to see. Students were encouraged to reflect on this discussion, along with their experiences playing Dow Day, to inform their own design choices during the workshop.
14 Gaming My Community - 14 Next, students played a short AR game designed specifically for the workshop. The game was intended to: 1) demonstrate some of the features available in the game engine, especially those that did not appear in Dow Day; 2) provide students a second chance to see how the engine worked and allow them to experience playing an AR game in the area around their own school; 3) serve as an additional context for framing the groups' discussions surrounding the limitations and affordances of the software (e.g., how accurate is the GPS? what additional features would you like to see? etc.). After playing the game, the students were given the task of designing their own AR games. They self-divided into four groups, consisting of one to three students each. Each group then brainstormed game ideas and created a map-based storyboard similar to the ones described above. Three of the four groups decided to create fictitious games that borrowed heavily from their prior knowledge of video game genres (e.g., RPGs, adventure) and popular culture. The fourth created a game that investigated water run-off around the school. The games produced by the first three groups included a mystery game based around a theft, a fantasy-adventure game, and a vampire game. The latter two were both driven by quest structures where NPCs supplied players with tasks or quests to complete. The fantasy-adventure game, for example, required players to gather various items in order to obtain special powers. The students who designed this game reported they were inspired by their interest in Anime and role-playing games such as Final Fantasy and Kingdom Hearts. In contrast, the three students who made the game about run-off reported that they did not play many video games. The extent to which this impacted their choice of topic is unknown, but an interesting question for future research might investigate how students' background experiences with video games informs the way they think about and/or design AR games. Those who made the run-off game were unfamiliar with run-off as an environmental issue and did not know where water from their school or nearby parking lot ended up. In order to design their game, they took photos and made observations around the school and throughout the neighborhood and conducted some brief Internet research related to run-off as an environmental issue. In this way, the design process served as both an inquiry into using the AR design tools and as a method of inquiry into a topic the students were unfamiliar with. In the end, this game was the one most closely tied to place in that it guided students to specific locations and asked them to make observations. For example, players were guided to a local stream and asked to locate the water outfall (the pipe that directs street level water to the stream). In this way, the physical location became a critical part of the game. In contrast, the other games were driven by external narratives and were less place-specific. In essence, they were games that could be mapped onto any location. One challenge for AR game designers is to connect these two areas. That is, to create games that leverage what we know about games and gameplay and what we know about local places. Design Experiment 7: State Street Design In spring of 2008, we held a 15-hour workshop that focused on developing students' understanding of place. The key elements of place that we focused on were: places as contested spaces and the cultural influences on placemaking and the design of community spaces. Two different tools digital photography and augmented reality game design were integrated to
15 Gaming My Community - 15 facilitate the interrogation of a local place. Photography was conceptualized as a tool for note taking, gathering evidence and communicating key ideas about a place. Designing augmented reality games, conceptualized as an alternative to traditional writing, was used to help kids come up with narratives that communicate the ideas they wished to convey about the place. We hoped that the workshop would: increase student engagement, provide a shared experience that contextualized some of their earlier classroom discussions about place, and develop students' ability to use AR tools to construct stories and communicate ideas. Ten senior high school students from a local alternative high school participated in this project as part of their social studies and media production class. In pre-interviews, most of the students reported that they enjoyed hands-on activities, but felt that they were missing from many traditional classrooms. They also reported that they were unsure about their ability to understand the AR tools. We interviewed students individually with regards to their technology use and found that their experiences varied widely. Some of the students used technology (personal computers and other relevant forms like digital cameras) for only typical tasks such as writing papers or using Facebook or MySpace. These students remarked that they were worried about the level of technical expertise required of them for this project. Others made significant use of technology in their lives, both in and out of school, from video editing to maintaining a computer for the sole purpose of understanding virus behavior. The first task given to the students as designers was to take two pictures at school and provide descriptions of those pictures with the intent of describing to the viewer something important about the place (school) through their eyes. The aim of this activity was to get the students to start thinking of photography as a way to convey ideas and tell stories. The following week the students took a field trip to State Street, a popular street in the center of the city, to gather data for the purpose of researching and producing their game. To structure this activity, we held a brainstorming session just before heading outside. During this time we queried the students as a group about what they already knew about the place, what they wanted to find out that day, what kinds of evidence they could expect to find, and how they planned to gather that evidence for later use in their games. This discussion was further organized around themes that the students had previously covered in class, such as differential use of space, architecture, economic prosperity, as well as those that emerged (figure 4.) during the discussion, such as local versus corporate owned businesses. During this discussion, students and researchers shared prior experiences of the place. The purpose of this activity was to scaffold the data collecting experience by creating an organizational framework. In addition, based on our prior design experiences, we believed that sharing stories would exemplify the kinds of information students might be able to gather from conducting interviews. Our stories were also used to demonstrate the link between the place and the personalities or perspectives of the people who use it, which is a critical component of creating storylines and NPCs for place-based games.
16 Gaming My Community - 16 Figure 4. On the left: Questions about place that were brainstormed by the students before the field trip to State Street. On the right: Student sheets to map out notes and observations. After the brainstorming session, each student team explored the place, took pictures, and conducted interviews. They had a note sheet as well (figure 4.), but due to the bitter cold (-5 F), few notes were taken on it. We spent about four hours total on the street gathering data, with a lunch/warming break in the middle. Students took about a hundred pictures (figure 5.) and ten video clips, and conducted several successful interviews of people on the street and in stores. In particular, with respect to the theme of local versus corporate owned businesses, they found that no employees of corporate owned store would grant interviews or allow pictures to be taken. In each corporate stores they visited, they got the canned response, You ll have to ask corporate. Students noted that the local shopkeepers were more willing to give substantive interviews. This surprised the students and in fact allowed them to better identify which stores are corporate owned.
17 Gaming My Community - 17 Figure 5. Pictures taken in the field. After the field trip, we printed the photos and presented them on tables so that students had access to all of the photos taken during the trip. This encouraged students to reflect not only on the photos they took, but also on the photos taken by other students. Students exchanged conversation about what happened, heard stories from others and made sense of the locations they had not visited. Each student was required to make an augmented reality game with five different locations and photos. We were aware that a 15-hour workshop was too short to make good games, and focused on how the tools, digital photography and augmented reality game design, facilitated students inquiry of place and what kinds of learning occurred in the process. In terms of learning, we found that kids picked up embedded tech skills (e.g., using a card reader, using a PDA, video compression, html docs, etc.). The design process gave students opportunity/context to learn these skills from more capable peers or teachers. To our surprise, students used tech skills and resources from outside of school to organize and produce content. After the field trip, a student immediately uploaded photos to his Facebook account so that he could share that experience with friends online. When there were technical problems, they looked for solutions from each other. Another student, a quiet girl, surprised us by suggesting that students use PhotoBucket to resize photos for the game editor, while a third student found a new website that allowed people to resize images. In the State Street experiment we asked students to gather assets within the vague frameworks/themes we gave them. Once they had all of these we asked them to create a game / tour. In the Game Unit experiment, we noticed that the games were more game-like in some sense because students started with game concepts first and then filled in the content later. For example, the problem of figuring out how to emulate a game like Kingdom Hearts in an AR game format was much easier for students because they were able borrow from their prior knowledge of games (content, storyline, mechanics). It appeared to be easier for some students to whip up a game where one has to collect all of the pieces of a necklace in order to gain special power than it was to design a game to teach players about the history of State Street. This variety of tasks led students to discuss and debate some interesting questions, such as what counts as a game? It led to them envisioning who their players were and tweaking their description for end users. In this process, the place they represented through their AR designs was not just a place, but a place in which meanings were situated by usage. Through the interrogation and design, a place was re-experienced, relocated and redefined.
18 Gaming My Community - 18 Discussion As is evident in the student AR game design activities described above, we have not yet discovered a 'silver bullet' to structure the activity of AR game design. Place-based educators tend to agree that physically embodied experiences often give richly unique understandings of places. But that does not automatically translate into good AR design. Despite the problems, the authors have all noted robust successes in AR game design research. For some participants, the draw of creating these games is so strong that frustrations only motivate them to more thoroughly attack the problems. For others, the frustrations and obliquity of the task is overwhelming, requiring further structuring of the problems into smaller, and easier-to-succeedat chunks. In the Greenbush History the teacher and researchers tried, perhaps too hard, to control both the content and the structure of the game design, while inadequately providing the students a rich knowledge of the constraints and affordances of the game editor. Consequently the students -- some of whom had played the Mad City Mystery AR game, and some of whom had not -- became frustrated because they felt they had been asked to democratically create the game they wanted, only to be constantly steered toward the game their teacher wanted, or that the researchers said was possible. In Wild Moose and Mitchville, the research began with a group of kids' exploration for a game that, on its own did not result in a finished narrative. The researcher, however, turned the unfinished narrative into an AR game that a subsequent group was able to play. This experience with an AR game informed their understanding as they designed the Mitchville game. In Max's Greenbush Story the researcher gave wide latitude to the student designers, allowing them to create whatever type of game they could design on the editor. Max experienced moments of frustration when faced with the amorphous task of creating a game without clearly defined bounds or direction, and it is our feeling that the student who dropped out did so because of the enormity of that task. This suggested to us that more structure was needed, and led to the creation of the AR game design scaffold sheets used in the Nature Hill experiment. In Nature Hill, the researchers began with sheets designed to scaffold the activities, and these seemed to be somewhat effective in guiding the thought processes of the students. Unfortunately, the amount of time allotted to the curriculum was not enough to create a game. This experience really confirmed the difficulty of game design. As researchers we really struggled to break things down into small, easy to understand, components. It became clear that there are still needed pieces to further scaffold the task of game design. In the Virtual Tree Tour, the researcher tried to keep the end-task in mind for the student. The student was designing alongside other designers with a specific well-defined game idea. As a design activity, there was little evidence that Peter used the higher-level design thinking skills that require understanding the game as a dynamic system between the game and the players. In an interview after playing an augmented reality game designed by the Star School team, Peter acknowledged that he did not focus on the players' perspective, or on how a game designer might engage players. In Game Unit and State Street, because this was part of a social studies curriculum, we put
19 Gaming My Community - 19 more focus on teaching and learning aspects of space and place -- but framed it through the lens of being an AR game designer. The low levels of attendance and continuity of students in the classroom, coupled with the extreme cold of the February field trip, made progress on the game design slow, and sights were set with these things in mind. Thus, a deeper understanding of place through the task of creating a game was prioritized over finishing 'complete' games. Suggestions for Future Design Workshops From the sum of these experiences, we have come up with a number of guidelines that we will bring into our future AR design activities, and no doubt continue to refine. As was true with these design experiments, future work will be shaped not only by the experiences we have and share as a group, but by ideas, innovations, affordances and constraints of the situations and individual personalities that contextualize them. Despite the specific visions and interests fore grounded by the researchers (outdoor educator/designer, photographer/poet, teacher/musician, mathematician/gamer), certain approaches to AR game design have emerged. First, we agree that the most success in designing a game was with kids who had played AR games. This is a formative experience that offers a glimpse of what works and does not work, and does so in a way that is experientially-based rather than theoretically-based. For example, on a 200x200 pixel screen twenty locations may seem practical in a game, but it is not until one experiences on foot the scale of the actual space represented, that the differences between a oneblock game on that screen and a 16 square mile game on that screen become significantly evident. Furthermore, it provides an embodied experience for understanding the design affordances of the particular AR tools. Second, it is clearly important to scout the location of an AR game with an eye on design within that place. Playing an AR game can greatly inform the survey of a potential location regarding questions of scale and safety. When scouting, take notes and photographs to mark spots that stuck out as potentially significant. Imagine the space relative to your experiences playing AR games. Is it easy to walk through? Does it require crossing busy streets? Is it too loud for video or audio? Are there enough engaging points of interest to carry a game, and are they close enough together to be practical? It s important to be familiar with the area so you can accurately envision it when you re working on the editor. This minimizes the surprises you ll have to deal with after you port the game to the handheld. Third, we suggest using paper-based aids such as the ones pictured in this paper to assist in the 'mapping out' of the place and game. When the game space begins to become mapped out with notes and photographs, patterns may start to emerge that may guide the 'curriculum' of the game. For example, it would be difficult to create a game on stream life if the game area does not include a stream. Online tools such as Panoramio, Google Maps and Google Earth (and others that are beginning to appear) allow photographs, text, and even HTML representations layered onto maps and satellite images. We have made use of these tools in several of the experiments, and they are continuing to improve. Fourth, we have found that it helps to make a very simple prototype and play it. Create a simple tour of the game space that brings you to the most interesting areas of the space. This will 'make it real' and offer a small success. If it works in manual mode, bring it out to the game area. Essentially, this counts as a finished game, and all one needs to do is refine it. It also refreshes a
20 Gaming My Community - 20 connection with the game space and proffers a realistic perspective of the scale that one is working with. Fifth, write a simple game narrative that connects the pictures. You may need to add/remove/ reorder pictures in order to make the narrative coherent. Plug these elements into the game editor and go play the game with someone else. Is it fun? Does it promise to be fun? If so, great. While you re playing, have a camera and camcorder available in case there are pictures or videos that would better support the narrative. You may want to do some rough acting at this point to fill in and add realism to the narrative. Note how long it takes to play the game, so you have a good understanding for what you can add, or what you need to simplify in the next version. Sixth, reflect on what you have and revisit what you want. Start to finesse and polish in the game editor. Add a branching narrative or two if you want, but be sure that you have a solid and fun linear narrative down first. Play it again, this time with an uncooperative non-tech-savvy person. Do they like it? Does it engage them? Adjust as necessary. Repeat until you get the game you want. Conclusion Space and place shape learning. As Clifford Geertz prompts, "no one lives in the world in general" (1996: 262). Place-based technologies like GPS, Google Earth, Google Maps, and Google Local feed the need to better understand our relations to space and place, and usher in a world of technologies that further situate us in our places and present places of others. Playing AR games offer opportunities to interact and connect with place. But learning requires more than content; it requires active participation in the practice of content. In "Pedagogical Praxis," Shaffer (2004) suggests new technologies offer opportunities of Deweyian-type laboratories where "the focus is on learning and the conditions and processes that facilitate learning in technology-rich contexts writ large" (p. 1402). When one considers video games as learning environments (Gee, 2003; Kafai; 1995; Shaffer, Squire, Halverson, Gee, 2005; Turkle, 1995), it becomes clear that designing games, especially AR games, can be an exceptionally powerful learning experience. Designing an AR game not only situates a person more deeply in a place through the embodied understanding that is needed to design a good game, but also offers the agency that Maxine Greene (1978) argues is necessary in learning -- the ability to "see, shape and transform" one's world at a tangible level (p. 193). Just as the entry point for participation in online worlds has become increasingly attainable -- evolving from the early days of command-line interfaces and online bulletin boards to the point where hundreds of millions have personal, customized online identities on MySpace, FaceBook, and elsewhere -- so too will participation in mobile place-based gaming, especially as the necessary tools become more reliable and increasingly ubiquitous. Research experiences like this one that examine the design of AR games (Squire et al, 2007) have and will continue to propel the technology forward. The research described here was conducted adjunctive to more formal research that had to deal more directly with the logistics of partnerships between schools and other community organizations. These largely independent projects were aimed at teaching students and teachers