Fostering Ambiguity: Decontextualizing and Repurposing a Familiar Public Display Clinton Jorge1, Julian Hanna1, Valentina Nisi1, Nuno Nunes1, Miguel Caldeira1, Giovanni Innella2 1 2 Madeira-ITI, School of Design, University of Madeira. Northumbria University. Campus da Penteada Newcastle upon Tyne NE1 8ST 9020-105 Funchal United Kingdom Portugal {clinton.jorge, julian.hanna, miguel.caldeira}@m-iti.org, {valentina, njn}@uma.pt, giovanni.innella@northumbria.ac.uk ABSTRACT Innovations in HCI tend to rely on exploring new technologies and novel forms of interaction. For decades artists such as Jenny Holzer have sought to provoke the public with art installations by repurposing public displays and exploring ambiguous messaging. Gaver argues that ambiguity can be intriguing, mysterious, and delightful, something that engages users and allows them to explore, discover, and interpret situations for themselves. In this paper we describe MStoryG, a public digital art installation that employs a decontextualized and repurposed airport split-flap display to support collaborative storytelling. In order to explore whether ambiguity attracts the glances of passersby and through curiosity invites interaction we devised a high fidelity software prototype that facilitated rapid deployment of experiments at two different locations. In addition to evaluating user engagement with the installation we define guidelines for others seeking to repurpose familiar objects in order to attract and engage passersby. Author Keywords Digital Art; Public Displays; Exquisite Corpse; Split-Flap Display; Interactive Storytelling; Repurposing Displays. ACM Classification Keywords H.5.m. Information interfaces and presentation (e.g., HCI): Miscellaneous. General Terms Evaluation; Design; Interfaces. INTRODUCTION Human beings are curious creatures, and as such we are Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. CHItaly. September 16 – 19, 2013, Trento, Italy. Copyright 2013 ACM 978-1-XXXX-XXXX-X/XX/XX...$10.00. attracted to objects which evoke some degree of mystery and uncertainty. We require a feeling of wonder in our lives [21], often seeking to escape the limits of mundane reality by immersing ourselves in the imaginary worlds of fictional stories [20]. Storytelling is an intrinsic part of peoples’ lives and a fundamental component of the human experience [26][11]. Through stories people explain, problem solve, teach, strategize or summarize [11]. Grandparents tell stories to grandchildren about their experiences, conveying important messages, and describing memorable and valuable events [26][11] often using objects to reinforce meaning or provide additional context. As newspaper columnist Ina Hughs stated: “Objects are stories solidified”1. Objects tell stories, not in the verbal (human) sense of the meaning, but in their history, aspect, or state of belonging, for example through relationships that people have with objects or memories the object withholds. People interpret, construct mental models around, and otherwise give meaning to objects that usually, besides triggering emotions and nostalgia, have a purpose, e.g. to support a task within a context. Over time an object may accumulate multiple interpretations and layers of ambiguity, or have a single interpretation that evolves or changes [24]. Art and design may be as much about re-appropriating, repurposing, decontextualizing and de-familiarizing objects as creating something completely new. Artists are masters at designing experiences that lead to multiple interpretations and ambiguity. They look at objects differently than other people and are able to interpret and alter the interpretation of objects by design, in order to through their creativity, express themselves, transmit a message, or influence the viewer’s experience. Similarly, digital art is often created with the intent to provoke feelings, associations, and reactions, where in many cases the focus is an object that may be associated with a particular location or context, and digital technology. 1 http://www.knoxnews.com/news/2013/apr/27/ina-hughsobjects-are-stories-solidified/ Interactive digital art installations are playing an increasingly predominant role in everyday life; interactive technologies enhance our experiences in urban public spaces. In accordance with Simmel [25], “interactive play in public places can help to build feelings of connectedness; it can draw heterogeneous people together”. Brignull and Rogers [1] found that people often resist interacting with public installations because of feelings of social embarrassment. They furthermore argue that special care should be taken when designing interactive installations for public use. These installations should invite interaction and entice users to cross the activity thresholds of peripheral awareness—people are aware of the installation but do not know much about it—and focal awareness—people pay more attention to the display, learning more about it—into the direct interaction activity space, and back again. The authors designed the Opinionizer, a tool to create and display content for public viewing, and found that remote interaction reduced users’ feelings of social awkwardness at the cost of reducing the honey pot effect, the social buzz and progressive increase in the number of people in the immediate vicinity of the installation. Brignull and Rogers further concluded that bottlenecks—not necessarily only physical ones, but also people’s conceptions as to what the installation is, who uses it, how long it takes, and the social system of practices—hinder their motivation to interact with the installation. In this paper we build upon previous work by Gaver et al [6], which argues that users appear to be attracted to ambiguity, and that if used judiciously ambiguity can be a way of provoking new perspectives on everyday life [5]. Furthermore, we ask whether ambiguity can overcome one of the most neglected public display design issues: attracting passersby [16] while approaching the ever present conundrum, user interaction resistance. We present our process in evaluating such issues through the Exquisite Corpse storytelling technique on an airport split-flap display that has been decontextualized and repurposed as an interactive public digital art installation, inspired by the American conceptual artist Jenny Holzer [23]. We reflect on lessons learned and conclude by discussing our results and insights. MADEIRA STORY GENERATOR Madeira Story Generator (MStoryG) is a digital art concept and preliminary prototype for a public installation to be located at an indoor high density and flow location such as a shopping center, high street, or public park. Our objective with MStoryG is to provoke passersby by providing them a space to engage in storytelling through a repurposed decontextualized airport split-flap display [28]. We argue that this apparently familiar, recognizable display, which has been a part of the regional airport since 1974, may take on multiple interpretations and generate productive ambiguity when it is decontextualized and repurposed for another use, such as storytelling. The large (3.5 by 2 meter) (see Figure 1) display should provide an attractive interface for generating, sharing, and consuming visitors’ narrative (story) fragments while the progressive disclosure nature of the display’s mechanism, i.e., writing one line at a time, allied with the split-flap animations and sounds should feel familiar and intriguing. Figure 1. Donated airport split-flap display located at our institute. Related Work For decades American artist Jenny Holzer used ambiguity as a central aspect of her art installations. Holzer plays with the contradiction/tension/opposition between form or medium (usually authoritative) and message (usually subversive, creative, warmly human). In particular, Holzer has a series of installations that focus especially on repurposing public displays—marquees, LED displays and such high visibility displays as the Times Square light board (2001) and the massive video board at the Dallas Cowboys Stadium (2012)—with ambiguous messages as part of her Truisms projects. Repurposing or public installations making use of split-flap displays are somewhat uncommon. Marco de Mutiis’s 2012 FUSE artist-in-residence project entitled “Arrivals”2 was a kinetic installation realized with an old reengineered Solari Board from a train station. The installation consisted of decoupling the split-flap modules, separating them, and displaying singular characters hung in space. Stamen Design’s 2011 installation “Talk to Me”3 utilized a Solari Udine airport display for an exhibition at MoMA in New York. This project involved temporarily replacing the board’s usual flight information by intermittently inserting an easter egg, a hidden feature that would present visitors with a small message at the bottom of the panel. 2. http://www.marcodemutiis.com/flaps 3. http://www.moma.org/explore/inside_out/2011/09/30/h acking-the-solari Other more artistic solutions have included use of split-flap displays for projects like “Signal to Noise” by Lab[au]4 which focuses on the sound produced by 512 mechanical split-flaps. The circular display immerses the spectator in patterns of sonic motion in conjunction with the textual information being transmitted. AMBIGUITY IN DESIGN The Oxford English Dictionary defines ambiguity as “the quality of being open to more than one interpretation; inexactness” [2]. Understandably, ambiguity in critical systems, applications, and websites—really in most humancomputer interactions—is neither valued nor recommended. Usability and efficiency are the gold standard in HumanComputer Interaction (HCI), while ambiguity is seen as a disruptive and opposing force. Instead of regarding ambiguity solely as a problem, however, Gaver et al [6] suggest that ambiguity can also be something intriguing, mysterious, and delightful; something that engages users with innovative systems and allows them to discover, explore and interpret situations for themselves. In order to further understand ambiguity, Gaver et al distinguished three kinds of ambiguity: ambiguity of information (uncertainties about the information displayed); ambiguity of relationship; and ambiguity of context (how things are understood in different contexts). Repurposing and Fostering Ambiguity in MStoryG Ambiguity can be an important factor in crafting, engaging and creating thought-provoking interactive designs [6]. Gaver et al describe the difficulty of finding examples of commercial products that are ambiguous in terms of their preferred context of interpretation. We agree, but would argue that ambiguity can be introduced into task-specific objects by displacing, repurposing and then readapting them to new contexts. We argue that we are fostering the productive ambiguity of the airport display board: an object that is not inherently ambiguous is being opened up by redeployment with a new function (collaborative storytelling) and context (a non-airport public space) to multiple interpretations and possibilities. The MStoryG installation fosters and explores both ambiguity of context (how things are understood in different contexts) and ambiguity of information (uncertainties about the information displayed) in order to promote engagement and provoke the audience. An airport split-flap display is meticulously designed to perform a single well-defined task in a specific locus: to display flight information within the airport. When removed from this context, it should generate a sense of ambiguity that is not 4. http://www.creativeapplications.net/objects/signal-tonoise-by-labau-512-mechanical-split-flaps-and-thenoise-of-data/ derived solely from the display itself as an object, but from its new purpose when decontextualized. Ambiguity should not be mistaken for novelty, since it is highly unusual to see a split-flap display board separated from its traditional context (airport or train station). We assume that our installation fosters ambiguity since travellers are familiar with the board and will read from it a clear reference to airports and train stations around the world. The display supports such a specific task that when it is removed from its original context, ambiguity of context and information should arise. The airport display board donated to MStoryG has never before now been seen apart from its original context; nor has it ever been used to display the kind of non-airport-related, mismatched, deliberately provocative, user-generated content we want to introduce in these experiments. EXPLORATIVE PROCESS Traditional HCI methodologies tell us little about the relationship resulting from the interaction with interactive digital art [3]. Furthermore, we required an exploration phase to understand the viability of the display and our assumptions about repurposing, decontextualizing, and fostering ambiguity. Moreover, we needed to explore the display as a storytelling interface and which interactions would suit the users best. To this end we defined an explorative process that involves two main probing phases employing a virtual replica (software prototype) of the airport display to afford rapid prototyping, deployment, and easy iteration of concepts, unrestricted from the sheer size and weight of the physical airport display. The MStoryG Pilot Installation Probes are tools or instruments ideal for the measurement of certain unknowns. They enable data collection about the use and impact of technology in real-world settings [5], people’s everyday activities and social interactions [9], or when introduced into a public urban environment to explore and research space [22]. In order to assess our goals we deployed a pilot installation of MStoryG as a technology probe in order to explore how people perceived the layout of the split-flap display, typically used to communicate arrival and departures of flights, and whether visitors engaged freely with the installation. A high fidelity software prototype of the airport departure split-flap display was created in Flash AS2 (see figure 2). Every 60 seconds the prototype would query Twitter’s search API for new mentions (tweets) to the MStoryG Twitter account. The display affords a continuum of 26 alphanumeric characters per line, 8 characters in the Airline region and 18 characters in the To sections, providing a potential single (central) block of text of up to 260 characters. Figure 2. The interface of the flash prototype, which is visually similar to the airport display. Initial Probing In the pilot study, our virtual board presented visitors with a sentence stem on the first line (a story starter of the “Once upon a time …” type) and a previously shared story starting on the fifth line (up to the tenth). The software replica was able to accommodate the same 260 characters available for the story fragments. An adjacent laptop was provided with an interface inviting visitors to leave a story fragment up to a maximum of 130 characters and further informing them of the possibility to interact remotely via an “@MStoryG” Twitter mention (see Figure 3). The three-day pilot study was set up in the research institute’s informal meeting room, a point of passage for several labs. During this first experiment we collected 15 contributions and interviewed 10 users. Ambiguity. Passersby immediately recognized the board as “the one from the airport”. The ambiguity of context—an airport board situated in the research facility meeting space—was found to be a positive factor that sparked interest and curiosity among visitors. Passersby did not immediately perceive the purpose of the virtual airport display when it was placed in that context. At first visitors were somewhat baffled about the information displayed. They did not expect to see stories and messages and attempted to make sense of the information, some by directly mapping the labels Airline and To onto the information displayed. One user initially thought the board displayed arrival information of new researchers to the institute. Others thought it was connected to the physical display at the airport and that messages could be sent between them. Interestingly, four visitors interpreted the installation as a childhood game, where each person continues the sentence left by the previous player. This observation at the pilot stage eventually led to the use of Exquisite Corpse in the public installation. Main Insights: Over time, passersby started to realize that the information consisted of user-generated content, which turned out for the most part to be short messages rather than fictional stories or story fragments. Not everybody is a natural born author, and some users found contributing to a fictitious story on request to be a difficult task, with most comments falling into the “I don’t know what to share” category. The installation was put to a slightly different use than we had originally envisioned. The context in which the board is installed and the actions that normally take place in that context influence the way people will use it—sometimes with interesting results. For instance, during out pilot, one user took advantage of the board being located in the meeting room and the possibility of remote interaction through Twitter to leave a message apologizing for his late arrival at the meeting. People recognized the board as the one from the airport, which led to speculation and multiple interpretations of its purpose and function in the meeting room. After contributing, users would remain in locus waiting to see their message appear on the board with the traditional split-flap animations and characteristic sounds. Users contributed once, but would frequently pass by and glance at the installation to check for new contributions by others. In order for stories to constantly flow on the board, the installation needs to engage many users. It is critical to place the installation close to a high-traffic area. Some users suggested improvements to the interface to better support storytelling. They were obviously aware that changes to the virtual board are possible. We are interested to discover whether we will get similar comments utilizing the physical board. Only one (out of 10) user opted to interact through his Twitter account, doing so from within his lab and then moving to the installation space to see it appear on the board. Discussion While users reacted positively to the installation and the use of the “airport display” to transmit and share user-generated content, users did not immediately perceive the objective and purpose of the installation. This may have hindered some interactions with the installation but also succeeded in generating curiosity and “buzz” around it. We were curious about the four user comments that predicted the installation taking on a more collaborative storytelling role. This in fact led to our adoption of Exquisite Corpse, a surrealist game similar in nature to the one suggested by these users, for the public installation. We were also inspired to reflect on the motives behind the board displacement by the user who thought that the board was still somehow connected with the airport and receiving and sending messages to and from a travelling audience. We recognized that deploying the board in this manner would add a further productive layer of meaning by maintaining important ties between the object and its original function and context. A final but important lesson from our first pilot was to make sure that our next probe would be located in a public space with greater foot traffic. passersby and decided to pursue it further with another deployment of the installation in a public space with higher people traffic. The addition of the Exquisite Corpse storytelling technique brought some changes in the way stories appeared on the display. Visual refinements to airport specific text labels such as Time, Airline, and Flight were altered to more abstract symbolical representations (see Figures 4, 5) in order to mitigate users trying to directly map the information to the labels. These new symbols had to be representative enough without constraining users’ initial perceptions of the purpose of the board. Figure 3. The pilot study was located at a point of passage between multiple labs. MSTORYG PUBLIC INSTALLATION Many experiments have focused on creating immersive experiences within urban spaces by augmenting cities with fictional information [13] through mobile location-based games and interactive location-based narratives [20][18]. While relating fictional stories to physical locations is not a novel concept in itself [13], Kearney [10] argues that storytelling innovation depends both on the medium and the story being told, and on the cues embedded in the physical surroundings people draw upon [12]. Figure 4. Layout replicating the physical airport departure split-flap board for the pilot study. Exquisite Corpse Exquisite Corpse is a surrealist collective storytelling technique in which each contributor adds to the story in sequence without being able to see more than a small fraction of what has already been written—blind to the greater part of the story, in other words. The resulting story is inevitably surprising and often beautiful. The American film director Tim Burton’s 2011 Cadavre Exquis [29] experiment is a recent incarnation of the game and a popular example of a community-created story. In Burton’s version, each contributor had at their disposal a maximum of 140 characters to express some continuation of the story. Similarly, in the Novel Iowa City Project [14] eight experienced authors and the general public were invited to participate in a community-based writing project. People contributed to the novel by tweeting their story segment with the hashtag “#icbfn”. In both of these experiments users contributed remotely and interacted through Twitter. User contributions were filtered for any serious disruptions and screened for the optimal story continuity. Study The pilot probe assisted in gathering some initial insights relating to user perception on the deployment of the airport display with user-generated content. We identified ambiguity provided by repurposing and decontextualizing the airport display as something that provoked curiosity in Figure 5. Visual modifications made to the board including new symbols replacing text labels. Evaluation Protocol Greenberg and Buxton argue that evaluation is some cases can be ineffective or even harmful if we blindly follow “by rule” rather than “by thought” [7]. England et al claim that traditional HCI models tell us little about the relationship resulting from the interaction with interactive digital art [3]. An evaluation protocol was defined based on the analytical framework of Mathew et al [15] in order to evaluate whether ambiguity generated enough curiosity to attract users and motivate them to cross the thresholds up to direct interaction, and moreover to evaluate their level of engagement with the overall installation. Three main user engagement trajectories were defined: • Perception phase: user’s first contact with the installation. The user creates an initial understanding of the installation and social ethics. • Interaction phase: user’s first interaction (or intention to interact) with the installation. • Engagement phase: final phase when the user moves from interaction to a deeper understanding of the installation. public flow points inviting people to visit the hall and the MStoryG installation but no project description was provided. Evaluating the Perception Phase: Users may perceive MStoryG as a Dynamic-Interactive installation, where a human actor plays an active role in the modification of the installation or the environment it is set in. Monitoring the perception trajectory and (building on Brignull [1]) in field user observations should allow identification of user activity zones as a measure of physical and social engagement: For the installation, visitors continued the story fragment left by the previous visitor displayed on the first lines of the board. This allowed the board to have “stale” content that could be shown in order to facilitate users’ understanding of its purpose [27]. The new user’s contribution was displayed below the previous story, so that both stories appeared together for 40 seconds. This allowed for the visualization of the continuing story. Afterwards, the latest contribution took up the initial position at the top of the board as the story fragment to be continued. The entire collection of user contributions, the story generated up to that time, was then shown on the display every 15 minutes, therefore keeping the installation active with sound and animation and offering visitors and passersby a view of the story so far. As described by Brignull remote interaction mitigates social embarrassment [1]. We offer passersby the possibility to interact with the installation via Twitter; users could tweet from their own device a story fragment with the mention “@MStoryG”. An adjacent laptop with a Web interface allowed those without Twitter accounts an easy way to contribute. • Peripheral awareness threshold: people are peripherally aware of the installation but at this time do not intend to interact with it. • Focal awareness threshold: people in this activity zone are engaged in social activities relating to the installation. They discuss, explain, point and learn about it but still from a distance without interacting. • Direct interaction: here users are in direct interaction with the installation and formulate a deeper understanding of the installation. We aimed to observe if and how apprehension and social embarrassment constrain user interaction with the installation, and whether the ambiguity of repurposing a familiar public display in a new context as a digital art installation is enough to attract passersby and entice them to cross the threshold of direct interaction. Evaluating the Interaction Phase: For the interaction trajectory, we relied on in field observations for identification of the honey pot effect [1] and analysis of the stories generated (contributed) by users in order to identify any references to the repurposing and ambiguity of the display, the context of the installation, or meta-references to actual events [14]. Evaluating the Engagement Phase: For the final trajectory, an eleven-question (5-point Likert) questionnaire and semi-structured interview were designed to query users on their engagement with the installation, their understanding of the concept of Exquisite Corpse, their overall experience with the installation, whether they were curious about the installation, and if this curiosity influenced their interaction. We also sought to understand users’ initial preconceptions of the installation before interacting and if the actual experience matched their initial mental model. The Installation The installation was located at an indoor public location, a regional science park hall (see Figure 6) adjacent to the University Campus, with a higher flow of people than for the pilot study. The prototype display was retro-projected to a glass storefront that had been covered in semi-transparent vinyl allowing the projected image to be seen on the glass (see Figure 7 and Figure 8). Posters were located at key Stories were limited to 130 characters. We made use of the split-flap board’s Flight (8 characters) and To (18 characters) sections. This allowed for a block of text up to 260 alphanumeric characters or a maximum of two 130character textual contributions to be shown at one time. Figure 6. Schematic of the location of the installation augmented with popular foot traffic paths. Findings The installation was deployed during four afternoons from 1pm-5pm. The majority of passersby were individuals that worked at the science park, students that visit the copy center or snack bar or the general public. Ninety-five passersby were observed entering at least the threshold of peripheral awareness. Sixty of those (63%) did not cross the direct interaction threshold but remained at the focal awareness threshold. Figure 6 represents the most popular foot traffic paths. Passersby would not stray much from their path due to the high visibility of the display and its location but would slow down or stop to glance, read and discuss. Thirty-five (36%) users crossed into the threshold of direct interaction upon invitation of a collocated collaborator. At the end of their interaction with the installation these users were presented with a questionnaire and a short semi-structured interview was conducted. A researcher was present on site during most of the time the installation was active in the regional science park hall. Most of the 60 passersby that walked by without interacting with the installation were approached and asked the reason why they did not engage. Participants mentioned social awkwardness i.e., uncertainty about the quality of their shared story and what others would think, intimidation, and the language barrier (the installation and initial stories shared were in English). Some of those questioned did not know what (story) to share and were happy enough reading others’ contributions, while some participants claimed lack of time. A small but present number of users were unsure of the purpose of the installation and were hesitant to interact, similar to findings by Munson et al [17]. The installation offered the possibility to remotely interact and contribute through Twitter, but this did not engage the more socially embarrassed to interact. In fact, none of the 35 interactions were performed through Twitter, leading us to conclude that social embarrassment was perhaps not the strongest factor hindering interaction with the installation. We overestimated the public’s use of Twitter [17]. Users commented that Facebook was their most preferred and active social network. The 35 visitors that interacted and shared a story waited to see their message appear on the installation with the traditional split-flap animations and sounds. The first person to interact with the installation had the opportunity to begin a story and contributed with “Lisboa 14h00”, information relating to the contextual ambiguity of the airport display. The contributions that followed fell into similar airport, flight, and travelling meta-references. A “hello world” disruption was not enough to disinterest visitors, who quickly continued the thread with “this world is mine what is yours?” Up to this point the content generated was in English, until a visitor preferred to share his story in Portuguese while still continuing in the same context. From this point forward all contributions were made in Portuguese and quickly, meta-references to actual events (consistent with [14]) such as Christmas, a large regional party and the New Year started to appear which then dominated the story content. A total of 261 words were generated from 35 contributions. We found no references to the location where the installation was deployed. This might suggest that the environment surrounding the installation was not rich enough to entice users into including it and imagining stories around it. Figure 7. The adjacent interface for leaving stories and the projected display. Users that had interacted while repeatedly passing by the installation would read any subsequent contributions. We did not observe any recurring interactions. Questionnaire Results on Installation Engagement Here we present the results obtained from the 35 questionnaires (on a 5-point Likert scale) and qualitative data from the semi-structured interviews from users who interacted with the installation. Users strongly disagreed that they felt embarrassed while interacting with the installation (at the point of interaction, not while approaching or at the periphery of focal awareness), with a median of 1 and mode of 1 (σ = 1.371). Some users commented that their level of embarrassment diminished right after interacting with the installation and understanding its purpose. Figure 8. A passerby reading a story in the focal awareness threshold. Users were somewhat undecided as to whether they would use the installation frequently, with a median of 4 and mode of 3 (σ = 1.121). Users did comment that if the installation were located somewhere else, for example in town or on the university campus, they would probably interact more. Consistent with the pilot probe study, not all users are authors. For some it was easy to generate a story; for others less so (median 3, mode 1, σ = 1.371). Confidence in one’s authoring ability is highly personal, and we should look into how to support users with various levels of confidence in authoring. Users responded that they would wait on average less than three (3) minutes to see whether anyone else would continue their story, bearing in mind that the installation was located at a point of passage in a professional setting. Encouragingly, users reported being very interested in reading the final story generated by all contributions (median 5, mode 5, σ = 0.471) and found collaborative storytelling through the Exquisite Corpse technique to be engaging (median of 5, mode of 5 and σ = 0.918). Furthermore, users’ curiosity about the installation especially due to the virtual airport board did lead them to interact with the installation (median of 4, mode of 5 and σ = 1.162). When asked if the installation was what they expected (before interacting with the installation) user responses varied. With a median of 3 and mode of 4 and σ = 1.271, some users did mention leaving messages on the airport display while others commented on it displaying airport relevant information and not allowing for user interaction. Overall, participants rated the installation as “fun” (median of 4, mode of 5 and σ = 0.738). Nov [19] argues that in order to increase and enhance user-generated content contributions that fun was a top motivator, and that there is a strong correlation between motivation and contribution. DISCUSSION The installation was deployed during the third week of December, when there is a lower than average amount of pedestrian traffic at the science park. We did not observe any honey pot effect that would clearly benefit this type of interactive digital art especially for collaborative storytelling. The gaps in time and space between passersby did not allow them to observe others’ interacting with the installation, thus probably hampering the passersby perceived usefulness and interest of the installation [8]. In general we felt that while the display’s location was optimal for viewing from the majority of the popular trajectories, but this point of passage did not favor our target space or wait time required to view the story generated ‘so far’ of 15 minutes. Two participants who displayed an interested in reading the complete story and calculated the remaining time for it to appear on the display and returned later on, with the remaining users commenting that they prefered to read the story somewhere else, for example on Facebook. We argue that similar to the experiment performed by [1], “sites of temporary relaxation” [4] are possibly some of the best urban spaces for identifying the honey pot effect and to entice users in collaborative storytelling. Public installations that require user interaction should be located where there is a balance between leisure space and high traffic trajectories. We found that users were excited by the Exquisite Corpse storytelling technique but had difficulty in generating their own stories to share [11]. Users that contributed were very interested in reading the final story and understanding how their contribution affected the larger story. Contrary to examples from Tim Burton’s Cadavre Exquis and the Novel Iowa City Project, we did not filter out disruptions to the flow of stories. Instead of alienating participants, our initial findings support that potential disruptions led participants to try and generate stories from these— although we recognize that this was done in a somewhat controlled environment with a collocated collaborator and most likely would not be true in more public spaces. Repurposing Objects and Ambiguity Müller, argues that attracting passersby is one of the most critical and overlooked design issues [16]. We presented a repurposed and decontextualized airport split-flap display as a counterpart to new, novel, complicated technologies. We expected our prototype of the familiar display to entice and provoke curiosity in passersby, not only attracting them but also in overcoming social anxiety and embarrassment and motivating interaction. We found passersby to be curious about the repurposing of the display with some generating speculation about its purpose. Furthermore, we found that creative storytelling is not for everyone. Most passersby were more engaged in reading the stories than creating and sharing. We foresee the physical board providing a stronger sense of ambiguity since it is a local landmark that was dislocated from its original location and context, and not a virtual counterpart. The evaluation protocol proved adequate in probing how users reacted to the installation in a public space, while Mathew et al’s analytical framework and activity thresholds were found to be a useful tools for observing and measuring the installation engagement and impact on the surrounding public. CONCLUSION In this paper, we presented a digital art installation concept that employs to a decontextualized and repurposed airport split-flap board in order to support Exquisite Corpse, a collaborative storytelling technique. We argued that the airport board provided ambiguity, thus attracting and enticing passersby to interact with the installation (even if only to read the stories). We defined a process to evaluate a number of features relating to storytelling and the viability of the display. We created a software replica of the airport board to allow us to rapidly prototype concepts, deploy a virtual installation quickly and iterate. Two probes were performed in order to understand users’ perceptions of the split-flap display technology. We analyzed users’ reactions to public storytelling and to their interesting in interacting with a somewhat authoritative public display affording a more democratic use. We feel that the insights presented here are relevant as more and more digital art installations are appearing in public settings. Our research showed that it is possible to evaluate digital art installations with some level of rigor and insight, even when using virtual counterparts as explorative probing tools, without compromising the creative process. FUTURE WORK These studies allowed us to identify beforehand that not all people are attracted to the concept of collaborative storytelling, especially when they are the content creators. Our next steps shall be to focus not only on the users that contribute with stories (which are a minority) but on those that are interested in consuming interesting stories as readers. This could be done by inviting international authors and writers or through Twitter fiction events and an authoring platform. This should open up the installation to individuals that enjoy sharing stories and generate higher quality content for passersby to consume. ACKNOWLEDGMENTS We would like to thank ZON Multimédia for their financial support to the overall project and to the Airport of Madeira for donating their arrival split-flap display. REFERENCES 1. Brignull, H. and Rogers, Y. Enticing people to interact with large public displays in public spaces. Proceedings of INTERACT, (2003). 2. Dictionaries, O. “Ambiguity”. Oxford Dictionaries, 2010. http://oxforddictionaries.com/definition/english/ambigui ty?q=ambiguity. 3. England, D., Fantauzzacoffin, J., Bryan-Kinns, N., Latulipe, C., Candy, L., and Sheridan, J. Digital art. Proceedings of the 2012 ACM annual conference extended abstracts on Human Factors in Computing Systems Extended Abstracts - CHI EA ’12, ACM Press (2012), 1213. 4. Foucault, M. and Miskowiec, J. Of other spaces. diacritics, March 1967 (1986). 5. Gaver, B., Dunne, T., and Pacenti, E. Design: Cultural Probes. interactions, february (1999), 21–29. 6. Gaver, W.W., Beaver, J., and Benford, S. Ambiguity as a resource for design. Proceedings of the conference on Human factors in computing systems - CHI ’03, ACM Press (2003), 233. 7. Greenberg, S. and Buxton, B. Usability evaluation considered harmful (some of the time). Proceeding of the twenty-sixth annual CHI conference on Human factors in computing systems - CHI ’08, ACM Press (2008), 111. 8. Huang, E.M., Russell, D.M., and Sue, A.E. IM here. Proceedings of the 2004 conference on Human factors in computing systems - CHI ’04, ACM Press (2004), 279–286. 9. Hutchinson, H., Hansen, H., Roussel, N., et al. Technology probes. Proceedings of the conference on Human factors in computing systems - CHI ’03, ACM Press (2003), 17. 10. Kearney, R. On Stories. Routledge, London/NY, (2002), 156. 11. Kelliher, A. and Slaney, M. Tell Me a Story. IEEE Multimedia 19, 1 (2012), 4–4. 12. Kjeldskov, J. and Paay, J. Public pervasive computing: making the invisible visible. Computer, (2006), 60–65. 13. Kjeldskov, J. and Paay, J. Augmenting the City with fiction: fictional requirements for mobile guides. Mobile Interaction with the Real World, (2007), 1–6. 14. Likarish, P. and Winet, J. Exquisite Corpse 2.0. Proceedings of the Designing Interactive Systems Conference on - DIS ’12, ACM Press (2012), 564. 15. Mathew, A., Rogers, Y., and Lloyd, P. Post-it note art. Proceedings of the 8th ACM conference on Creativity and cognition - C&C ’11, ACM Press (2011), 61. 16. Müller, J., Alt, F., Michelis, D., and Schmidt, A. Requirements and design space for interactive public displays. Proceedings of the international conference on Multimedia - MM ’10, ACM Press (2010), 1285. 17. Munson, S.A., Rosengren, E., and Resnick, P. Thanks and tweets. Proceedings of the ACM 2011 conference on Computer supported cooperative work - CSCW ’11, ACM Press (2011), 331. 18. Nisi, V., Wood, A., Davenport, G., and Oakley, I. Hopstory: An interactive, location-based narrative distributed in space and time. In Proceedings of TIDSE 2004. Springer, Springer Berlin Heidelberg (2004), 132–141. 19. Nov, O. What Motivates Wikipedians? 50, 11 (2007). 20. Paay, J., Kjeldskov, J., Christensen, A., et al. Locationbased storytelling in the urban environment. Proceedings of the 20th Australasian Conference on Computer-Human Interaction Designing for Habitus and Habitat - OZCHI ’08, ACM Press (2008), 122. 21. Paulos, E. and Beckmann, C. Sashay. Proceedings of the SIGCHI conference on Human Factors in computing systems - CHI ’06, ACM Press (2006), 881. 22. Paulos, E. and Jenkins, T. Urban probes. Proceedings of the SIGCHI conference on Human factors in computing systems - CHI ’05, ACM Press (2005), 341. 23. Read, C.&. Jenny Holzer Bibliography. http://www.cheimread.com/artists/jennyholzer/?view=bio. 2013. 24. Sengers, P. and Gaver, B. Staying open to interpretation. Proceedings of the 6th ACM conference on Designing Interactive systems - DIS ’06, ACM Press (2006), 99. 25. Simmel, G. The sociology SimonandSchuster. com, 1950. of georg simmel. 26. Smith, J. GrandChair: conversational collection of grandparents’ stories. (2000). 27. Taylor, N., Cheverst, K., and Müller, J. Affordances and signifiers of community noticeboards. (2009), 1–4. 28. Split-flap display. In Wikipedia. http://en.wikipedia.org/wiki/Split-flap_display. 29. Tim Burton’s Cadavre Exquis. http://burtonstory.com/connect.php. 2011.