Keywords

1 Introduction

Nowadays, location-based information systems’ role is progressively gaining importance in our everyday life. With the help of social media applications and mobile technologies (Craig et al., 2002), location based content is being generated easily by the public and substantially used for several purposes such as understanding decision making process about key social, political economic and environmental facts (Porter, 2011). Especially gathering location-based information from local communities is relatively more valuable in order to create sustainable solutions for human needs and improve the quality of life (Elwood, 2008). It is obvious that different approaches such as participatory design (PD), user centered design, ethnographic research methods and participatory geographical information systems (PGIS) have common characteristics in terms of their main aim. Although these practices have been used by different disciplines, it is seen that they share the common ground of understanding the needs of human and coming up with relevant design solutions.

In our ongoing design study, our aim is to make a contribution to the participatory geographical information systems (PGIS) studies by means of novel ideas for information representation and interaction. We have been gathering both content related needs and visualization inspirations from users by conducting PD workshops and individual user input by journal studies in combination. In PGIS studies; more of the participatory approaches can be seen at the information gathering stages, while public involvement has been weak during information visualization stages.

This paper shares our insights from an ongoing research study for exploring new ways of visualization and interaction of spatio-temporal data in daily urban life. Meanwhile we are also looking for sustainable and effective ethnographic data collection methods for spotting intuitive use patterns. With this perspective, our objective is to investigate possible visualization techniques by adopting PD and ethnographic research methods in our study. We envision that these insights may lead us to more sophisticated and effective participatory interactive solutions for interaction with the urban data.

2 Background and Motivation

Throughout the last 50 years, location based data has been elaborately investigated by scholars under the field of Geographical Information System (GIS). Recently, this field has discovered the importance of spatial information analysis (Maguire et al., 1991). Subsequent to spatial information analysis, making explanations and predictions related to spatial information have become much simpler (Clarke, 1997). On the other hand, public knowledge has become an important issue to identify the needs of locals (Elwood, 2008). With the help of volunteered geographic information (VGI) (Goodchild, 2007; Sui, 2008), GIS has been shifted to a more democratic approach. This is because VGI methods provide an environment where the public can be much more engaged with the production of collaborative geographical knowledge. Interactive online instruments such as Google Maps and Open Street Map create open access for the public to build and share their own maps.

While location-based applications are becoming widespread, users’ needs grow and become diverse. A variety of personas use these systems and a huge number of information layers appear on maps. This situation reveals the need for studies on geographic information visualization techniques. Moreover, new contexts of usage such as mobile phones, big screens or dual screens create challenges in both visualization techniques and interaction models. One of the most commonly seen participatory geographic information visualization technique is attaching pins on the maps (Ushahidi Crisis Maps; Global Transition to a New Economy). Color-coding, using attributions of visuals elements, such as circles, squares, lines, using textures, and 3D modeling are other ways of representing geographic information on map (Roth et al., 2010; Fifa Development Globe).

An important issue in location-based information visualization studies is the lack of public participation in the evaluation process of visualization techniques. One of the instruments of public participation in design is cultural probe method. Cultural probes are a range of physical objects which participants interact with. These objects help us to understand user needs and behaviors. Probes can be notebooks, postcards or maps (Gaver et al., 1999). A cultural probe even can be a wearable prototype where direct user feedback is collected, like a wearable recorder to collect city sounds (Gaye et al., 2004). Some geographical visualization techniques were evaluated by conducting user studies (Dockerty et al., 2006; Lange, 2001; Appleton and Lovett, 2003). In addition, some researchers have done collaborative geographical visualization sessions to design much more usable systems (Brewer et al., 2000). Since the geographic visualization topics are varied and complex and technology use changes quickly, we believe that sustainable methods for learning about user mental models and user needs, as well as the need for public engagement to the visualization process is crucial. GIS tools, such as Arc View, have dominantly been used in the field of participatory GIS. These tools would be effective in order to visualize geospatial information but there would also be several usability problems due to the similarities of these kinds of software.

3 Method

With the perspective above, our main approach in this study is to explore how people intuitively perceive their presence and movement on the map, how they visualize it and how participatory methods can be used for learning from the users about these two questions. For this purpose, we collaborated with people from different design domains with a series of participatory design and journal studies in 2014. Here, in this paper, we discuss five consecutive stages (Fig. 1); workshop 1 (user needs + visualization), workshop 2 (conceptualization + visualization), workshop 3 + journal study (visualization), interview (user needs) and a table top game (mental model + information architecture).

Fig. 1.
figure 1

The structure of workshops and journal study

Full size image

3.1 First Stage: Workshop 1 for User Needs and Visualization

All the process started by planning a series of design workshops which were announced to designers as exploratory studies on spatio-temporal data visualization about the city and how people interact and perceive this information within their daily life.

Workshop 1 was announced online through an art and design institution, SALT Istanbul. Five participants who from different design backgrounds like interaction design, architecture, urban planning attended the workshop. After a presentation about data visualization, we briefed the participants about the “Design Network of Istanbul” case study. After a card sorting session to discuss about the content and information architecture of given topics, we separated them into two groups to work on issues related to data visualization (time, navigation, route, action). Working individually for a while, and then brainstorming as a group, teams gathered and presented their ideas to each other, discussed and voted for one of the ideas. Our aim in voting was to see if the competition engages participants more or not, and it ended up being time consuming and left less time for visualizing ideas. At the end, two teams presented their final ideas.

Even though there were some interesting ideas pointed out like the effects of day and night change in the city on human movement and how to visualize them; the results were not satisfying in terms of detail. Our insight from this study is that; the workshop content was too loaded for a one day practice and the participants didn’t engage enough in the topic presented.

3.2 Second Stage: Workshop 2 for Conceptualization and Visualization

With a similar agenda and based on our previous experience, we held a second workshop at Koc University. The workshop was announced online to networks of design domain. Five participants; four students and one graduate attended. The workshop which started with our presentation where we briefly explained what data visualization is and showed relevant examples of it as well as the output of the first workshop. After the presentation, we asked the participants to visualize spatio-temporal information for scenarios that were presented to them. Discussions included personas and different scenarios also using factors like day/night. These scenarios were based on daily movements and actions of certain personas in the city; such as a tourist, disabled person, homeless, couple, family, employee, tradesman and local tourist. For idea generation and sketching, the participants were divided into two groups and each group had five scenarios to visualize. Since there was no voting, two groups gathered only once at the end of the workshop to present their ideas and discuss.

Even though the structure of Workshop 2 was more focused on the visualization part than Workshop 1, time wasn’t enough again for more sophisticated visualization ideas. After Workshop 2, we decided to change our method. We prepared a journal (Fig. 2) and sent it to a group of participants that we announced the previous workshops, before inviting them to the 3rd workshop.

Fig. 2.
figure 2

Journals design for the participatory studies

Full size image

3.3 Third Stage: Journal Study and Workshop 3 for Visualization

At this stage, we created a journal with which users reflect their city experience with visual and verbal representations. Our aim here was to give the users enough time to intuitively reflect how they perceive and represent their presence and movement in the city based on their own daily experiences. There were three important questions that we were interested in this study; (1) what are the things that people choose to record?; (2) how do they visualize them?; (3) How do they perceive their presence in the city?

Journal Structure. The Journal, an A5 sized notebook, which the user can carry around while travelling, is designed with enough pages for seven days (Fig. 2). For each day, there are several parts allocated for different needs. First, there is a grayscale map of Istanbul and a blank page for free drawing. Then, there is a striped page for the user to write their experience. Finally, there is a page with questions, aiming to guide the user to think about some of the issues in the city; such as “Were there any hitches around the city?” or “What type of information did you need to achieve?”

We announced the journal study to a group of the previous workshop participants and other people with design background. During this period, we announced a third workshop. One of the main objectives is to learn about the journal users’ experience with this new medium and discuss them.

Workshop 3. 9 participants, mostly from interactive media design field, attended the third workshop. The workshop lasted for 6 h and consisted of three parts; (1) presentation about visualization methods and discussion about the journal, (2) journal users’ experience sharing, (3) users’ selection of topics and their individual working process on visualization. The workshop started with a detailed and structured presentation about data visualization, followed by briefing on the journal. In the second part, the participants who had the journal shared their experiences. Some of the journal users addressed that they started to pay more attention to their surrounding while completing the journal. In this part, the participants, who didn’t have the journal, had problems with engaging. Later on, we shared our user scenarios from the previous workshops. This was followed by a discussion and idea generation session about user needs. Afterwards, each participant selected a topic to visualize individually. Some of these topics were; historical transformation of the city, sounds of the city, events in one hour, empty areas of the city. For each topic that the participants chose, they determined 3 to 5 types of information to visualize. We imposed this limitation to participants to make them focus on visualization rather than information structure. This individual sketching session lasted for one hour. After an hour, participants presented their work and discussed their ideas all together.

3.4 Interview for User Needs

While we were conducting visualization exploration studies with designers, we also had a parallel process of long discussion sessions with the participants. However, non-designers’ perspective had to be included in the study as well. Therefore we executed in-depth interviews with relatively a more heterogeneous group. The outline of the interviews was based on citizen engagement in city governance, decision-making and planning. The aim of the study was to investigate citizen activities related to social life in the city. In order to have a heterogeneous group, we contacted with political activists, journalists, city planners, students, academicians, shopkeepers, cyclists, health activists, and filmmakers and so on. Thereby we had 17 interviews. During the selection of participants, we applied snowball-sampling method. With the help of these interviews we had considerably detailed information about social movements and activities in Istanbul. We understood how and in which way people use media for taking part in the urban life, what kind of information and actions should be included. We used these insights in the next stage as user needs and mental model.

3.5 Tabletop Game for Mental Model and Information Architecture

As mentioned earlier, the intention is not only the visualization, yet how people interact with this information. Thus, apart from all the above mentioned visualization and user need explorations, we also questioned how people interact, discuss and ideate on maps. With this in mind, we conducted a custom tabletop game for citizen engagement discussions (Fig. 3). Aim of the game was to observe the user experience while citizens of a city partake in the discussion on urban issues, using computer and online-like actions such as discover, share, archive and so on. The context of the discussion was one of our on-going studies on building a citizen engagement support system application.

Fig. 3.
figure 3

A scene from the table top game

Full size image

Briefly, the game was based on events and incidents happening in a city. Thus it was highly related to storytelling. We conducted three game sessions with minimum 4 players and a moderator in each (Fig. 4). Sessions were played around a table. In the center of the table we placed an empty Istanbul map in order to motivate the players to make discussions on map. We collected a series of real events that were related to both public and private city life, such as; a large fire of a historical building, social movements, new regulations on public transport. On each game, we opened 10 event cards one by one and gave five discussion minutes to the players. Each player had a set of action cards and they were asked to make actions with those cards based on their preference to react to the event. The action cards were built based on a previous in-depth interview study that we had conducted with 17 participants. The cards were; see, share, discover, disseminate, archive, create, interpret, argument, story, map, idea, report, group, petition, crowd funding and two jokers. Within 5 min for each event, players discussed and conducted their actions about the event.

Fig. 4.
figure 4

Tabletop Game Setup

Full size image

The game acted as a useful tool to understand interaction models and build information architecture for location-based systems. With the help of the tabletop game we found an interaction pattern and applied it to our citizen engagement support system design. The result of the game sessions is an issue that requires further discussion; for this reason, we will share our findings in another paper that will be on location based citizen engagement tools.

4 Insights and Discussion

In this section, we share our insights from above-mentioned participatory exploration studies; (a) the use of workshops; (b) user needs; (c) visualization inspirations.

Regarding the evolution of workshop structures, observation of the participants’ attitudes and reflections were effective in obtaining beneficial results. The first modification was about providing preliminary information for location-based visualization methods. We prepared a presentation about different methods at the beginning of each workshop as a briefing. Compared to the first two, in Workshop 3, we presented a much more structured presentation based on the categorization of visualization topics, visualization styles and interactivity levels. Also the journal study helped a lot for users to be more ready for the ideation process. As a result, participants seemed to be more willingly engaged in discussions.

The second modification was about the focusing level of each workshop. First two workshops had more generic topics such as; designers’ network in the city or visualizing urban data for disabled people. However, these topics were excessively complex to study at a limited time. Based on this experience, we inevitably narrowed down and identified topics at the workshop with participants. The journal users’ experiences and their true stories have been significantly influential for the scenario building process. In addition, journal study was also effective in getting representative insights about the execution of the study. Encouraging participants to create individual visual stories about daily urban life has also been effective in order to reflect more creative visualization concepts about their individual and fictional urban experiences (Fig. 5).

Fig. 5.
figure 5

Visualization sketches from PD workshops

Full size image

Apart from reflections on the PD workshop methodology, we also gained some insights about alternative user needs for urban data systems. Participants had discussing on concepts such as; spatio-temporal data about day and night change, facades of buildings as information providers, detailed information about pedestrians or urban cycle riders, transportation systems, public sphere information, environmental factors (such as noise, acoustics, waste, animal life), social issues (such as language and immigration), social and cultural events effecting the urban life, historical change of the city landscape and plan.

Although with these early studies we predominantly collected insights regarding user needs, we were also able to gather some ideas related with visualization (Fig. 6). The prominent visualization inspirations have some common features such as; color-coding, metaphoric visualization, texture usage and attributes of lines or dots. To identify different types of information (such as density, activity, pace), locations and actions, participants used color attributes such as; hue, saturation or alpha. Discrete geometric shapes, predominantly circles have been used to visualize human movements while in contrast; repeated amorphous shapes have been used to visualize vehicle movements. Obstacles have been visualized by patterns such as; broken lines, curved lines, zigzags and jagged lines based on their impact. Metaphoric visualization techniques have been used for noise and movement visualization. Density, place attributes and states have been visualized by using textures.

Fig. 6.
figure 6

Chart of visual representations from journals and workshops

Full size image

After we created an outline of the visual representation patterns, we categorized most of the visual elements from journals and workshops; into dots, lines, shapes and patterns (Fig. 6). As Bertin suggests (1983) these are basic and commonly used elements in the area of location based visualizations and maps. We looked into the visual characteristics of each element and what they were meant to stand for by the participant. After we sorted and analyzed the visual information from the journals and workshops, we built a chart. This chart contained the most commonly used elements by the participants. It also contained uncommon and remarkable representation ideas. The numbers at the top of each element indicates how many participants used that visual element. Location based visualization in relation to city context is one of our main research topic as a design research group. City by its nature is a multi-layered and complex phenomenon. In these studies, our aim is to reach self-represented visual and interaction styles that particularly coexist with the city phenomenon.

5 Conclusion

In these early stages of designing participatory location based information systems, we made use of a variety user research and participatory design approaches. With the intention of collecting intuitive insights about users’ needs for such a system, understanding the mental model and exploring how people visually represent and locate themselves in the city, we conducted a series of workshops and other user studies.

One of the main contributions of this early stage exploration study is the evolution of an iterative process of running the participatory studies and how to engage the participants more in the process while we also collected some insights about user needs and visual representation ideas. We believe that the study with journals helped a lot for more participation and saved a lot of time by involving participants’ daily life experience in the process, instead of working only in limited hours of a workshop. Another critical point has been that, although we invited users from different backgrounds and tried to create a heterogeneous group in the 3rd Workshop, this was not the case in the 1st and the 2nd. Hence, there was not enough data to generalize the expectations or needs of a random citizen from an urban information visualization tool. Thus, the whole study has to have a more interdisciplinary approach.

In the end, one of the main intentions here is to explore effective and sustainable ways of learning about users’ needs and mental model while interacting with the urban data. As mentioned above, we envision that these studies will lead us to interactive solutions building upon such ethnographic and participatory approaches. Meanwhile we also contribute to the GIS field by exploring visualization and interaction methods. We are now planning to detail the studies with more use of journals that incorporates with interactive tools and making use of other artifacts for pre-workshop activities recruiting more diverse user backgrounds. At the same time, we are also developing prototypes to evaluate the previously collected ideas.