CITY MOOD How Does Your City Feel?

CITY MOOD how does your city feel?

Luisa Fabrizi

Interaction Design

One year Master

15 Credits Master Thesis

August 2014

Supervisor: Jörn Messeter CITY MOOD how does your city feel?

2 Thesis Project I

Interaction Design Master 2014

Malmö University

Luisa Fabrizi

[email protected]

3 01. INDEX

01. Index ...... 03

02. Abstract ...... 04

03. Introduction ...... 05

04. Methodology ...... 08

05. Theoretical Framework ...... 09

06. Design Process ...... 18

07. General Conclusions ...... 41

08. Further Implementations ...... 44

09. Appendix ...... 47 02. ABSTRACT

This research aims to make the first step in the direction of creating guidelines to use for building an instrument capable to collect information about citizens’ emotional reaction toward their city. Through the use of an existing mobile application meant to collect data about one’s own emotions I tried to evaluated the availability of people in sharing how they feel. Later on, based on the evaluations of this app, I created two subsequent analog prototypes that can be placed in urban spaces to collect people’s place related emotions. The devices have been used both inside the university and in open urban spaces in Malmö. Using the experiences made with these prototypes, concerning the availability of people in sharing their emotions about spaces, this thesis is used to develop guidelines and design opportunities for further development of interactive street furniture designed to collect place related emotions.

5 03. INTRODUCTION

against the term smart city

In the last years we witnessed the rise of a new and controversial term: smart city; this has become the buzzword politicians, industrials, journalists and urban developers are using with the expectation to convince their public that choices made under the directives of “new” and “smart” are always right and winning; unfortunately very often in our society, the knowledge and critical attitude toward the so called new technologies are scarce and tend to naively and rhetorically divide what is good from what is evil, not understanding that technology is neither moral or amoral, but it is the use that people behind it make that define it as an instrument to empower the society or to threat its freedom. The term smart city is defying itself as a vision of a technological and connected urban environment where distribution, mobility, production and effectiveness are the principal term to define the value and success of the city (Hollands, R. 2008). Of course, this is not a reason to believe that new technologies, when related to the city, are only an instrument of the neoliberal and capitalistic culture; there are other approaches, and they are coming from researchers in the universities, activists and people from all over the world who have seen the power that computation can bring to the collectivity if designed and used the right way. With this paper I am hoping to contribute to this, maybe slightly idealistic, view of what could be done to make our cities better places to live in through the use of digital interactive design; I do this trying to understand how to use citizens’ emotive reactions to places in the city as a way to explore and understand the urban environment and to produce a discussion about it. how does your city feel?

I try to propose emotions as a way to trigger a reflection about the city; this seemed to me as a way to obtain input from as many people as possible: not everyone has the capacity to develop an opinion about a certain matter, but every person has a personal emotional response to what’s happening around her. Once we have put together different people’s emotional responses to a certain environment they can be collectively discussed leading to a new understanding of that environment and the way it is seen and lived; already in the ‘60s Jane Jacobs states that to understand and to build better cities it’s fundamental to start reasoning from the particular and only then going up to see the general image (Jacobs, J. 1961), having the opportunity to see the way citizen’s emotions are scattered and how they describe different areas of the city could be a way to start a new reasoning about our urban environment basing it on the affective portrait of the community. I am also, somehow, trying to propose on a city scale what design developers and researchers do when they test their designs: they try to understand what kind of emotions are triggered in their testers in order to evaluate what they built and to take decision on how to implement it (Isomursu, M. et al, 2007); cities are as well a matter of design and I think it is finally time to take in account what emotions they trigger in their “users”, even if so far, emotions and affect have generally been neglected in the discussions about cities, smart cities and urban environments (de Lange, M, 2013). The kind of research I am going to write about in this paper has to be considered as a first step toward a multifaceted project in the emerging field of Urban Computing that I am intentioned to contribute to in the future; the idea behind this design is to explore in a discrete manner citizen’s feeling about

6 their urban environment and to try to grasp how different portions of the city are seen and experienced; the live data collected should be then shown through a thoughtful map based visualization on screen which would need to be open, interactive and diffused through web, mobile systems and screens around the city. Citizens should then be able to interact with this system, comment the results, cross the emotion-data with other data to try to understand if there are patterns that may describe or explain why the city has certain emotional responses, they may as well propose solutions to situation or problem they detected. A project like this may need the financing and support of the government of the city, which should be very open and not invasive in terms of leaving to the citizen the opportunity to comment and add to the website as much as they would; also it should be a government ready to embrace, to try to understand and to answer to citizens’ requests and needs. In this paper I start to explore how the collection of citizen’s emotive responses to the place in the city could be designed and I try to answer to the following questions:

What is the right mean to use in order to help people to reflect about the relation between their emotions and the place where they are? What kind of interactive instrument could be collectively used in order to gain information about the feelings a certain place triggers in to people. What are the guidelines to follow when designing this kind of interactive system?

I began my research expecting to build an embodied mean, an “emotions collector”, two different ways of collecting seemed to be feasible:

• a wearable based on a personal mobile application (EmotionSense), • an interactive street furniture.

To answer to these questions I proceed first user tested an Android based app for smart-phones called EmotionSense, an emotion tracker, I thought that this would have been a good base to build a wearable; to understand if a wearable was a useful mean to answer to my questions I tested Emotion- Sense on a small number of people, I collected their reactions and idea about it through interviews and a questionnaire. Based on the user tests I found out that this wasn’t a valuable mean for my purpose and I decided to move my research further in the direction of the interactive street furniture. The concept of street furniture was introduced by Eric Paulos (2008) and it indicates new structures, that still need to find their shape, meanings and use, that could be used as interactive system part of a participatory urbanistic network. I continued my research building two analogical prototype to try to answer to the aforementioned questions. The first prototype, made out of cardboard, was a poll asking “How are you today?” I placed it around the university’s building, people had to take a paper chosen from six different stacks, each of which had an emoticon representing a different affective state, then they had to place this slip of paper inside the poll. After three days of testing I found that this was an effective solution. I built a second and more complex analogical prototype. This prototype was made out of MDF and plexiglass, and was composed of six different polls, mirroring the six different emotional states. The question asked this time was “How does this place make you feel?”, as I was looking for more place related responses. To answer to the poll users needed to take a paper from different stacks, all of the papers had written over

7 the question”why?” so that users could decide to write something on this paper before placing it in one of the poll or leave it white. I did a two days test inside the university, where I collected many responses and information and then another three days test outside in the city (Malmö), where I as well obtained many answers and new material to reflect upon; after it I had enough data to answer to the question I was asking at the beginning of my research, I found a valid direction to continue my investigation as I found both a way to trigger people in to reflecting about their emotional response about places and a way to make them sharing those reflection with me. Those findings could be used in the future to build an digital place located physical input used to collect people’s place related emotional statuses.

Image n.1, second prototype, ”How does this place make you feel?”.

8 04. METHODOLOGY

This chapter outlines what are the methodologies I chose to use to answer to the research question mentioned above. The first stage of my research was based on users test; the users were asked to test an Android based smart-phone app called EmotionSense, a behavioural app used to collect emotional data. I collected informations about the user testing through interviews and a questionnaire. As users’ responses about the use of a personal mobile device showed that this was not the right instrument, I shifted my research to less personal collecting devices, and I kept on working on my research in prototyping in order to identify the guidelines to be followed when designing an interactive street furniture. This prototype-based part of the research was conducted in three different stages and was based on the research through interaction design described by Löwgren (2007); this exploration phase was held in order to understand which of the artefact’s features were valuable and had to be kept and enhanced in more advanced stages of the design development. On the first stage I built a first analogical prototype meant to test if people were interested to share their emotions and I placed it for three times in a row in three different places in the university’s buildings, I evaluated the prototype through the number of interaction people had with it in each of the different places where I positioned it and through the observation of people using it. Once I found out that there was a certain number of people willing to share their emotional status with the physical input, I built a second analogical prototype, more complex and that needed more interaction. I placed this as well in different places around the university’s buildings and evaluated it through the observation of people interacting with it, through the interpretation of the collected data and through few interviews. More interviews would have been beneficiary to gain multiple perspectives, but due to the lack of time I was not able to conduct more. On the third stage of the evaluation of the physical input I placed it outside in the city. I chose a complex area of city, in order to have an interesting and flourishing number and kind of answers; this time the evaluation of the prototype was based on the understanding of the place, on the relation between the different spots where it was located and the kind and number of answers, on a close observation and of course on the evaluation of the answers, here as well I should have chose to do some interview to have a better vision of the project. During all the process of my research I enhanced my point of view through theories which I used mostly to direct my thoughts, to gather more ideas on the design and to develop a critical point of view both on my findings and on my design choices.

9 05. THEORETICAL FRAMEWORK

In this chapter I am going to give an overview of the theoretical framework my research is based upon. In the beginning I tried to state what is the general and popular approach to the idea of Mapping Emotions, when the interest in emotion has started and what is the cultural approach toward them. I also describe here the two scales in the actual researches of the emotional mapping, those are: a micro scale, where human bodies are emotionally mapped and a macro, worldwide, scale; my exploration, dedicated to the emotions of the urban scape, places itself as an in between these two; I want my reader to notice that with my research I am not at all trying to give a complete and struc- tured definition of what emotions are, sometimes I also use the term feelings or affect which would not be considerate as appropriate synonymous by an anthropologist or by a psychologist, but that simpli- fied this first attempt I did to approach such a complex research field. I found very relevant the material collected by the artist Christian Nold, that has attempted to create a methodology to collect and display citizens’ emotional reaction to places in the city. My investigation then, shifted toward more scientific approaches; I could not find any research in the Interaction Design field which I could consider comprehensive of all the aspects I wanted to take in account, but I identified the roots of my research in two main fields: Affective Computing and Urban Computing; my research can basically be seen as an attempt to understand and to find a way to connect and relate those two fields.

MAPPING EMOTIONS

As stated by Höök (2013), the first attempt to introduce a reflection around emotion and how they influence our rational thinking and our whole life appears rather late in the history of the “civilized western World” and it was made by Charles Darwin, in his book The Expression of the Emotions in Man and Animals (Darwin, 1872), the first research exploring the world of emotions without consid- ering them as a problem to overcome, but as a natural aspect of being human. In the last years the idea of talking about and understanding emotions has become far more popular than it was before. Even if it is still often considered a sensitive and sometimes inconvenient topic, the approach to emotions is becoming both more scientific and more attentive. Psychologists as Alexan- der Lowen through his Bioenergetic theory (Lowen, 1975), and philosophers as Umberto Galimberti have given a new interpretation of what emotions and feelings are, overcoming the old mind/body dichotomy the whole western culture is build upon since Greek fifth century b.C. philosophy; those new perspectives gave back their value to emotions, relocating them in relation to the body, opening new, more objective and more reliable opportunity for studies, researches and understandings on what emotion are and how they influence and are influenced by our everyday life (Galimberti, 1987). It is on this track that new researches have been made in the last year, such as the one conducted by Nummenmaa et al. (2013) addressed to create a body based emotion-map. The interesting and fascinating fact about this newly made research is that its foundings, explained through an easy to read and beautifully illustrated diagram (See image n.2), have become viral on the internet and could have been found for many weeks on different platforms such as news papers and science focused web pages as well as on more general location such as tumblrs, blog and even on memes based collaborative media. Those are proofs of how much the general interest has lately become focused and fascinated by the

10 idea of understanding the nature of emotions. In their investigation Nummenmaa et al. found that there is a consistency in how humans feel and in how and where they place and sense emotions inside their body. This research is meant to create a new helpful chart which could help us to easily and more exactly discretize and understand specific emotional feeling (Ibid.).

Image n.2, Nummenmaa, L. et al (2013). Bodily maps of emotions.

On a larger scale, but symptomatic of the renewed interest in the value of emotions, is the annual World Happiness Report made by the United Nations Sustainable Development Solutions Network (Helliwell et al., 2013), this year, on the third iteration of the report, Denmark has been elected the happiest country in the world (See Image n.3). It is important to understand that the World Happi- ness Report is not meant to compliment the happiest countries and blame the others, it’s a document created with the aim to find out what makes a country happy, how happiness is perceived variously in different places around the world and, on top of all, with the intention to officially dignifying happiness (and with it all the emotions) as a valuable, new and consistent evaluation approach; we should welcome this new and more human perspective hoping it will help our society to reconsider the high value of happiness over the capitalistic affection for productivity, performance and outcomes. To summarize, a new interest is arising around the subject of emotions, both from a scientific perspective than from a more popular one, as well it is becoming clear how to localize emotions, even on very different scale, is of primary importance to have some kind of understanding of them. So far researches considered and localized emotions on the perspective of the human body and on the perspective of the entire world, and they found out that both those approaches not just make sense, but are relevant to understand the nature of emotions and how they influence and are a significant criterion to evaluate our lives; Considered those two applications it makes sense to think to apply a similar approach to something that has a size in between the human body and the world: cities. In the book Emergence: The Connected Lives of Ants, Brains, Cit- ies, and Software, Steven Johnson theorizes the vision of the city as a macro-organism, alive and responsive to changes and aging and conforming itself as an anthill (Johnson, 2001). This means that if Image n.3, Carlsberg’s advertisment.

11 we have the right perspective (both visual and temporal) on cities we may have the opportunity, on the long run, to understand and analyze them on a completely new level and we may finally be able to intervene with thoughtful wisdom in order to help them to develop in the most desirable manner for their citizens. In this perspective, understanding and visualizing how a city feels may become a mean to see what’s right and what’s wrong about it; may this be the place to start from to make better cities for happier citizens? What does it mean to understand how a city feels?

Emotional Cartography

A preliminary answer to the questions asked in the previous section may be the artistic research Christian Nold has carried on between 2005 and 2007 in several cities around the world; Nold used a biometric sensor, an object that measures the galvanic response of the skin, and he composed it with a GPS and called it Bio Mapping Device (See Image n.3). He then moved in to different cities (Nottingham, Greenwich, San Francisco, Stockport) and asked to local community organizations to collaborate with him.

Image n.4, Nold, C. (2005). Bio Mapping Device.

Nold organized workshops with the people from local community organizations and asked them to wear the Bio Mapping Device and to walk around the city. The Bio Mapping Device measures the emotional intensity response of the person wearing it and the GPS locates it. Once back to the workshop base, people were asked to share their data, comment and process them, in order to make sense, for themselves as well that for the other participants, of what their arousal intensity meant and what had triggered it. So, the subjective mapping were created retrospectively and not just based on the “objective data”, but on the personal elaboration and validation of those data. The subjective maps were then composed in to a collective mapping of the city meant to represent the community reaction to certain areas and specifically located situations.

Image n.5, Nold, C. (2007). San Francisco Emotion Map.

12 Nold’s research is as well an effort to find the most communicative city-map based visualization of the collected data. He uses a different representation for each experiment where he combines the “objective data” (from the Bio Mapping Device) with the “subjective data” (comments, information, reflections,..). There is a very deep reflection on what it means to represent the information, the attempts swing from a scientific and diagrammatic representation approach to more artistic/comic based illustrations. The different choices Nold makes in the representation are justified as well by the different questions he’s trying to answer to; for instance, he reports that the Greenwich Emotion Map, one of the most diagrammatically represented documents, tries to ask the question: “How will our perceptions of our community and environment change when we become aware of our own and each others intimate body states?”; may a project like Emotional Cartography become a democratic mean of democratic involvement in the development of cities future master-plans? (Nold et al, 2009)

AFFECTIVE COMPUTING

Affective computing is a very specific branch in the world of interaction design that explores the influence that users’ emotions have in the experience of the interaction. This research was born within the new approach to the relevance of emotion started in the ‘90s that began to consider emotion as the basis for rational behaviour (Höök, 2013). The first definition and comprehensive study about Affective Computing was made by Rosalind W. Picard, founder of the Affective Computing Research Group at the MIT Media Lab, in 1997 with her book Affective Computing (Picard, 1997); currently, in the research scenario there are very different approach and interpretation of what is Affective Com- puting, what kind of contribution it could give to the field of Human Computer Interaction and how to approach and understand the range of human emotion and the value and the weight they could bring in the conceiving and developing of a design. Here is how Affective computing is explained on the official web page of the Affective Computing Research Group at the MIT Media Lab:

“Affective Computing is computing that relates to, arises from, or deliberate- ly influences emotion or other affective phenomena. Emotion is fundamental to human experience, influencing cognition, perception, and everyday tasks such as learning, communication, and even rational decision-making. However, technol- ogists have largely ignored emotion and created an often frustrating experience for people, in part because affect has been misunderstood and hard to measure. Our research develops new technologies and theories that advance basic understanding of affect and its role in human experience. We aim to restore a proper balance between emotion and cognition in the design of technologies for addressing human needs.” (Picard et al, n. d.)

Differently, in her interview for the web based Encyclopedia of Human Computer Interaction, Kris- tina Höök(2013) opens a debate about the different approaches existing in the field of Affective Com- puting, criticizing Rosalind Picard’s research attitude toward it, here Picard’s view, inspired to cogni- tive psychology, is portrayed as an attempt to discretize human emotions’ nuances through a hasty simplification, in this view affections are identifiable and can be categorized with the aim to develop interactions with machines capable to recognise user’s emotions, influence them and being influenced by them.

13 Contrary, Höök (ibid.) states her approach to the research as Affective Interactional Approach; this approach, as she declares, is more focused on the cultural aspects related to the theme of emotions and toward the development of software and applications meant to help users to understand and better experiencing their own emotions. Central focus of what Höök calls Interactional Approach are the attempts to design to stimulate the user in reflecting upon her emotional states and the pursuit of a non-reductionist representation and categorization of human‘s affective states; in the Interactional Approach the emotional responses sought while designing interactions with a system are not always and necessary positive: the interaction is seen as an experience needing complexity in order to be aes- thetically pleasurable and meaningful; for example the opportunity to communicate with another user and share empathy and sympathy with her is worthwhile and a medium able to support this exchange would be highly considered, while the emotions shared may not always be positive. While the research I am conducting has a lot do with Affective Computing and try to understand some of its different souls and aspects is vital to my work, my intent, at this state of my studies, is not meant to take sides or to judge one as better or more meaningful than another as this would imply a much deeper understanding of the field, of its theoretical foundations as well as a better development of a personal ethical view about those concerns. Anyhow, in the next paragraphs I am going to describe some projects developed in those different research contexts of Affective Computing and I am using them both to outline my research and to start building a personal perspective which is not meant to be seen as a final judgment, but as an ongoing learning and evaluating process.

Mood meter

Mood Meter is a project developed by researchers of the Affective Computing Research Group founded by Rosalind Picard at the MIT Media Lab (Picard et al, 2012). The project was held for ten weeks in the MIT campus during a five month festival organized to celebrate 150 years of the University. Mood Meter was an interactive installation scattered in different locations of the MIT campus; it consisted of four cameras connected with laptops equipped with software able to analyze the recorded images and recognize all the smiles appearing there, then a server received those information and used them to generate a public website displaying graphics of the number of smiles, where they have been collected, with which intensity etc.. Behind each camera a screen or a projection was showing what was being recorded, but whenever a person appeared her faces was hidden behind green blobs when she was smiling and yellow blobs when she was not. This was a clever expedient both to engage more people and make them more aware of the project , but as well a way to communicate a sort of privacy: no faces are being recorded, no personal information are stored. The MIT group focused a lot in making it clear how they were addressing the privacy issue, being this a very sensitive spot (Ibid.).

Image n.6, Picard, R. et al (2012). Mood Meter, visualization.

14 This project is very relevant for my research: in the way it somehow addresses the same kind of questions (understanding place related emotional responses) I am trying to answer to, in the mode it reflects upon problem of visualization of place related data (See image n.6) and looks for an engaging way to communicate patterns, but also because of some of the design choice that I would not have taken and that reflects a very peculiar way to manipulate and use data; I found the data visualization solutions really engaging, the decision to give an “emotive portrait of a community” and the effort to try to understand patterns in the affective responses to different moment of the semester, of the week and of the day and relate them to specific places. What I found not quite correct in respect of the way data are used is the choices to only report “smiles”, which to me seems mostly a way to advertise the campus more than a way to report the emotive situation of the people working and studying there, recognising this attitude made me even more aware of the sensitivity of data and how easy is to manipulate them not only in the way they are displayed but as well in the way they are collected, of course I understand how sometimes researchers have defined technical and founding restrictions that may force them to take certain decision and not other.

Affective diary

Affective Diary is a system developed by Ståhl et al (2008) based on the idea to design a system able to help people to understand and experience their own emotions. In their approach Stål et al were seeking a opportunity to design for the engagement of users in the embodied aspect of the processing of emotions, where the concept of cultural body was kept in account. The Affective Diary is system composed of a mobile phone equipped with a camera, armband and a tablet pc. The system uses the mobile phone to keep traces of the sms sent and received, it uses a blue-tooth connection to disclose the presence of other mobile phones in the proximity and collects all the pictures taken. The armband is provided with a movement sensor and galvanic sensor, which detects the arousal amount of the wearer based on the galvanic conductivity of the skin (it does not, though,detect the quality of the arousal). The armband has a wireless connection to the mobile phone. At the end of the day the user has to connect the phone to the tablet pc, which through a timeline shows the different data recorded at every time. Arousal and movements are displayed through an ambiguous shapes that while giving information are open for the use to interprets them. The user can as well add text or scribbles to the timeline if she feels like doing it (ibid.). As is easily understandable by the name of the project and by the choice of not giving close answers the user, Affective diary is strongly meant to encourage reflection in to users; as many of us could have experienced it is very difficult to keep traces of our own emotion over time (McDuff, D. et al, 2012),; Affective diary can be seen as an easy, clean and interpretable way to collect different status over a long period of time and, being able to reconnect it to happening (pictures), amount and kind of relations (sms pattern, blue-tooth sensor) and moment of the day, it could be an excellent way to trigger behavioural changes (Ståhl et al, 2008) .

Image n.7, Ståhl et al (2008). Affective Diary.

15 URBAN COMPUTING

The term Urban Computing was coined by Eric Paulos and Elizabeth Goodman (2004), and indicates the application of technology in urban environment with the aim of a better understanding, acknowledging and developing of the urban environment itself. The core concept of Urban Computing resides in a grassroots approach seen as an indispensable mean to deepen and enhance the perception that residents and civic government have of the city; of course, the grassroots approach stated before would be insignificant and weak without the modern and spread emergence of computing devices: Urban Computing builds its power and mean on the opportunity given by smart phones and other (mobile or not) computing devices to enhance the interactions of citizens with their urban environment (Paulos E. et al., 2008). Urban Computing’s strategy is the development and design of an orchestrated network of instruments scattered around public spaces as well as in mobile devices and capable to measure the environment they are in and; the kind of devices that we are already finding in our cities and that are probably going to grow in number, power and use are:

• Onboard sensors: embedded in our mobile devices; • Wearable: object embodied in the clothes or in the jewellery that sends informations to the user’s mobile phone through a wireless connection; • Left Behind: low power and low cost sensors scattered around places sending informations; • Scattered: sensors used in a certain context for a framed time; • Infrastructure: sensors placed in the urban environment, with specific tasks and embedded in a sensors’ network (ibid.).

Eric Paulos and his collaborators detected five research key themes that, constituting the framework of Urban Computing, identify it as a multidisciplinary subject where interaction designers, social sci- entists, architects, urban planners and geographers collaborate together. Here are the five key themes:

“People – Who are the people we share our city with? How do they influence our urban landscape? Where do we belong in this social space and how do new technologies enable and disrupt feelings of community and belonging?

Place – How do we derive the meaning of various public places? What cues do we use to interpret place and how will urban technologies re-inform and alter our perception of various places? How does technology create new places?

Infrastructure – How will existing urban systems such as power, water, subways, public transportation, signal lights, toll booths, etc. be used and re-appropriated by emerging technologies?

Architecture – What new techniques and smart surfaces will emerge for interacting with buildings, public surfaces, sidewalks, benches, and other “street furniture”? What role will new structures, shapes, and forms play?

16 Flow – What is a path or route through a city using these new urban tools? How will navigation and movement, either throughout an entire city or within a small urban space, be influenced by the introduction of computing technologies?” (ibid).

Participatory Urbanism

Participatory Urbanism, directly inspired by the mood and spirit of Urban Computing, is more focused on the empowered individual that finally become part of an influential critical mass actively participating to the life of the city: in this perspective residents contribute to their environment through the collection, the sharing, the crossing and the evaluation of data they produced. On this track we could expect Participatory Urbanism to became an instrument that makes citizens (at least the one involved in the generating, sharing and evaluating of the data and content) more attentive and aware of what’s happening around them, more responsible for their own city, feeling more protected and supported by the fact of being part of a community (a critical mass) and better informed about the city; on the other hand, civic governments are more informed on what are the issues it the places they are administrating, they have the opportunity to justify their choices (when coherent with the data and information shared) and can share some of their responsibilities with the citizen’s themselves (Paulos E. et al., 2008). Of course, there are risks in the use of those tools: sensors are often seen as aseptic instruments capable to represent the reality of the world around us, but sensors are setted by people to detect certain input and not others, visualization tool can be focused in showing certain data and hiding others and many other aspect can be handled in order to give a vision of reality that is distorted, strongly ideo- logical or utilitarian; this is why Participatory Urbanism, as well as Urban Computing, could became a strong political tools, they have to be handled with awareness, respect and every time the ideology behind them should be made clear and recognizable, this way those tools won’t be exploited as a pa- ternalistic instrument of soft public coercion.

Discussion in Space

Discussion in Space is an experimental system created to remove barriers between those who live the city and those who govern and build it. This operation was carried in the city of Brisbane, Australia, in 2011 and was used by the City Council during the preliminary phase of the public consultations part of the urban planning agenda. Discussion in Space was designed by the Urban Informatics Research Lab of the Institute for Creative Industries and Innovation of Queensland University of Technology led by Ronald Schroeter (2012). The system consisted of a certain number of screens in different places of the city; the screens were showing a question to the citizens that could be answered sending sms or through a twitter message; being the screens public there were moderator with the power to decline certain messages when considered inappropriate for public displaying. The idea behind this design is to have citizens’ feedback on proposed urban planning projects, and mostly to trigger the interest of young people who instead would not be interested or would not expect to be listened and taken in account in such important decision about their city. The project received a lot of participation; one of the most interesting facts about this project is how the researcher found out that most of the messages they received had been sent through sms and not

17 through twitter accounts, which mostly meant that the user preferred to participate using a “low barrier” mean. Other interesting aspects, understand through online questionnaires, are how important was for the users to see their comments displayed around the city, as well as seeing how many people were using the system.

Image n.8, Schroeter, R. (2012). Discussion in space, screen.

San Francisco CrimeSpotting

San Francisco CrimeSpotting (http://sanfrancisco.crimespotting.org/) is a smart-phone and web app part of a category of applications that Kevin et al (2012) define citizens apps; those are mobile applications, generally designed for a specific urban environment, with the aim to empower citizens giving them a deeper view of the space around them, augmenting the reality in which they are living in and offering them the opportunity to make their voice being heard and/or making them aware of what is going on in their city. In the group of the citizens apps, which can be built with different aims and different systems and level of interaction, collection and distribution of the data, San Francisco CrimeSpotting is a Public Safety app,with the goals of Problem Identification and Creating Awareness (Ibid.); the app (that is not affiliated with the City of San Francisco or the San Francisco Police Department) presents an interactive map showing crimes happening in the city, the crimes are as well classified and can be easily consulted; citizens can comment and leave their opinions about a certain crime. One of the most interesting things about the web-page of San Francisco CrimeSpotting are the state- ments that can be read at the bottom of the homepage: this system is considered by its developers (Stamen Design) not only as a way to create awareness about what’s the criminal state of the city among residents, but is also a “manifesto” to show to the local governments the value and power of data visualizations.

Image n.9, San Francisco CrimeSpotting, available at: http://sanfrancisco.crimespotting.org/

18 06. DESIGN PROCESS

Based on the idea that an embodied instrument would have probably been the best solution to gain collaboration from citizens to share their emotional thought about the city, I started my research expecting one of these two outcomes:

• building a wearable collecting emotional responses based on a mobile application (Emotion- Sense), • an analogical prototype of an interactive street furniture;

the testing of these instruments would have been used to outline guidelines for the best designs of an actual “emotions collector”. I started to approach the design process through the user test of an existing Android based app called EmotionSense as this seemed to be the fastest way to understand if a wearable would have been a valuable mean for the collecting of emotional reactions; EmotionSense is an emotion tracker which had some of the features I wanted to develop further in my project. Seven people tested it for a week, I interviewed two of them, the other five answered to an online questionnaire. From both the interviews and the questionnaire I found out that people didn’t really used EmotionSense, they mostly found it pointless and too intrusive, those founding together made me understand that personal devices, as a wearable, would not be the best choice as a mean to obtain emotional information from people to have a general view of how the city feels. As I decided to put aside the idea of the wearable, I introduced in to my design process the prototyping of place related physical input and their user testing in order to develop guidelines to use when designing an interactive street furniture. At first I designed a fast prototype, made out of cardboard and paper, with a simple design. It was meant to be easily understood and used: a writing on the top of it was asking “How are you today?” and it was supposed to be left in some place so that passer-byes could take one slip of paper with one of the six basic emotions (Handel, 2012) drawn over it, and stick it inside a poll through an hole in the middle of the prototype in order to answer to the question. With this first prototype I wanted to understand if someone would have been interested to share her emotional status through a poll and if and how much the positioning of the poll itself would have mattered on the kind of answers. I tested this first prototype for three days and I used as a base for my tests the C floor of Kranen, one of Malmö University’s building; the C floor is the main and busiest floor in the building, where the entrance, the cafeteria, a study area, and relax area are placed. On each of the three testing days I placed the prototype in a different place of the C floor. Through the different number of answers the poll collected every day I had the opportunity to understand that people were actually willing to share their emotional status and that places where people are passing bye are also the one where it is possible to collect more answers. As the first prototype gave me interesting answers and it mostly revealed itself as a working mean to gather answers from people about their emotional statuses, I designed a second prototype, more complex and basically built on what I found out during my previous user testings. This second prototype was meant to understand if giving a feedback on the received responses could have been a way to obtain more answers this is why part of it is made of transparent plexiglass. Also

19 I wanted to give more than one different way to answer in order to understand what is the one that people prefer. The second prototype, made in MDF and plexiglass, was asking the question “How does this place make you feel?” in order to try to obtain answers related to the place itself and not about a general emotional status. This time users were requested to stick some small slip of papers with the answer “why?” written over it inside one of the six polls behind the plexiglass; each of those polls was representing one of the six basic emotions previously mentioned. The users could chose to write something on the paper or just use it to answer to the general question placing it inside one of the polls. On the prototype there is also a short link to answer to a general questionnaire. I tested this second prototype first inside Malmö University’s buildings and after outside in the city. The first test happened on the bridge connecting Kranen with the building Ubåtshallen, this is a place where people move very fast and never stop; in this occasion the poll obtained 62 responses and over 34% was left blank while 22% of them had a meaningful written feedback on it. On the second day of testing of this prototype I placed it in front of the entrance of the university library. This time I collected 79 answers, the 30% of which were strongly related with the place while another 30% was left white. Only one person answered to the online questionnaire. I considered this user testing very successful as it seamed that people were actually eager to participate and share their emotional status about places and they often understood what was asked them; another reason why I think I had many answers was because I left the choice to the users to decide how much to share. As this first user testing gave me reasons to think I was going in the right direction with my research, I decided to take the second prototype outside in the city to test it in the wild. The place I chose for this second part of the test is Södervärn, a neighbourhood of Malmö in the district of Södra Innerstaden, where an important public transport hub is situated. This time I tested the prototype on three different days and in front of three different bus stops. The first day I placed it in front of a bus stop where city buses going to the south and to the east of the city where stopping; of 20 answers collected the 40% was strongly related with the place while another 40% was left white. On the second day I left the poll where regional buses were stopping, it received 29 responses and less than the 7% of those had a strong relation with the place, while almost the 60% was left white. On the last day I placed the prototype where buses going to the city of Lund where stopping; Lund is a city characterized by its university and most of the people going there are students, this time the poll received 40 answers, the 32% of which strongly place related and about the 10% left white. This iteration of the prototype’s user testing outside in the city gave me the opportunity to confirm most of my previous findings, I also noticed some differences in terms of collaboration and of trust toward the poll and who built it; those differences are probably to be ascribed to the diversity between the university, a semi-close environment where a certain amount of mutual untold trust subsists in between the people that consider each other part of the same community, and a space like the public transport hub of Södervärn where this mutual belonging feeling lacks and consequently people feel less like trusting and helping a stranger who placed a poll in the street asking for feedbacks.

EMOTIONSENSE

EmotionSense is an emotion tracker smart-phone app and is the app I used in the first phase of my research; I made the decision to use this system to began my exploration because it had some features that were very similar to the features I was interested to bring forward in my research and it made

20 sense to use an existing instrument to understand its values and to proof my initial perspectives. Emotion trackers as EmotionSense are also known as behavioural monitoring systems (Lathia et al., 2013) and are used to keep track of the user’s moods and feelings. The idea behind those systems is to understand behavioural patterns, arising awareness around those behaviours and to give the opportunity to the user to change what makes her unhappy, angry or depressed. The idea of tracking behaviours in order to trigger changes is not new (eg: www.lifeguideonline.org), but has been revolved by the large scale diffusion of smart-phones. Those devices, so common in our everyday life are as well a complete novelty and they present an infinite number of opportunity. In the case of behavioural change studies and applications the use of smart-phones could be powerful; a smart-phone is an object meant to be carried around all day everyday, is something we are in constant touch and in constant control of, it has several sensors that combined together are an open door on a myriad of information and insights about its user. As scary at it may seem, when ethically and thoughtfully approached and handled, those instruments are a great manner to monitor and understand patterns on a personal and on a large scale and this is why behavioural app are becoming more and more popular in researchers (Lathia et al, 2013). “Emotion Sense” is an app developed by researcher of The Computer Laboratory at University of Cambridge and released under a ISC license, it has been claimed to be a “pocket therapist” by different reviewers (Huffington Post UK , 2013; Ruki Sayid , 2013); EmotionSense is able, combining passive sensor data collection and machine learning, to provide continuous monitoring of participants’ emotional states while collecting data representative of each person’s social interactions and mobility (Lathia et al, 2013).

How does EmotionSense work?

EmotionSense has a very complex system of sensors and data monitoring that gives extra information and insights on the user’s behavioural pattern. The sensors used are the smart-phone’s accelerom- eter, the Blue-tooth, WiFi, GPS and microphone. This means that the app is able to collect information about the quantity of motion, the position and the level of noise around the user; In order not to bee overwhelming sensors are unlocked weekly. Those sensors are used to describe the device’s physical behavior. The app is able to collect as well the device’s social behaviour which is described by the number of SMSs received and sent, the number of calls, the activity on social media. The screen state and the battery used (device’s status) are collected as well (Lathia et al, 2013, Googledocs). All those information (that remain anonymous and not connected to the app’s users) are collected and both used to show the user what may have triggered the user’s mood and for the sake to enhance the research on smart-phone apps adopted for behavioural change. When first using the app it asks you how many times per day you want to monitor your mood, then, whenever is time for you to do so, you are reminded through a short buzz and an icon on your smart-

Image n.10, Lathia et al (2013). EmotionSense.

21 phone’s screen, basically the same system used for received SMS. The user records her mood through what Lathia et al (n.d.) call “Affective Grid” following the researches of Russell et al. (1989); an affect grid is a two dimensional scale meant for a fast recording of feelings; on the x axis are negative and positive states while the y axis indicates the grade of alertness or sleepiness (Russell et al., 1998). Every time, after recording the mood on the grid a few questions are asked,those questions vary depending on the chosen quadrant on the grid. The recorded emotions are stored and are shown all together on the affective grid. This is supposed to be a mean to reconsider the behaviour that may be behind those emotional response; as every week a new sensor is unlocked every week there is as well a new Affection grid relating the recorded moods to the new sensor or situation.

User Testing

At the very beginning of my research I asked to a small number of friends (10) to try to use Emotion- Sense; It was important for me to understand a certain number of things:

• how people would have reacted using this kind of app; • if the app itself was a valuable way to indicate sensitive data as emotions; • if there were certain people more interested than others in using EmotionSense; • if starting to use this app would have triggered some kind of self consciousness in the people using it; • if any of the people using emotion sense was relating her mood to places . As I already mentioned I mostly tried to involve in this process people close to me, whom wouldn’t have been too reticent in sharing delicate information as their emotional data; in this phase of my research I didn’t need to have all my users placed in Malmö and this is why I had some contribution from people placed both in France and in Poland. Not too surprisingly after the initial enthusiasm some of my “users” stepped back; this happened for several reasons as technical difficulties (the app only runs on certain Android devices), privacy issues (one of my tests didn’t wand to try the app because he would have had to sign up in order to use it) or because it just felt pointless and disturbing to keep track of their emotions and then share it; this is quite understandable, emotions are a sensible topic and not everyone is willing to share them for a research. To have a more comprehensive understanding of the application, its use and of my testers’ opinion I used the app as well for five weeks.

Interviews and questionnaire

I had two different approaches when it came to gaining the results from the user tests I did on Emo- tionSense: I conducted both interviews and surveys; I chose one or the other method depending on the strengthens of my relation with the users, the time they had to dedicate me and the physical distance between me and them. Five people answered to the online questionnaire I sent them, and, interestingly enough three of

22 them were not using the app, because of compatibility of the app with their mobile phone’s system or because of privacy issues (“it reads my sms and call history (and my phone is not supported)”, “ Because I don’t like registration in apps”). Some of the people answering to the questionnaire found the idea of EmotionSense fascinating, but were as well concerned about how useful it could be for them to use a behavioural app, they didn’t seem to be fascinated by it and mostly they were very unhappy with the kind of questions that Emo- tionSense was keeping on asking: “... I find it really hard to answer the questions from time to time. Am I anxious? I don’t know. Compared to what? Extremely difficult is the “To what extend was your day like other days?”, because it can involve sooo many dimensions. In terms of what? In terms of my mood? my experiences? where i’ve been, whom I talked to? It’s all so unspecific.”. Someone said that she would have been much more interested in knowing other’s mood:”...I’d be more interested in other people’s moods, around me.”. It seems as well that people using the app are mostly using it at home and they think that the place where they are influences their emotions (“certain places just make me feel like things aren’t so bad after all and on other places I almost always feel stuck.”, “ There are places where I feel secure (like my home) ,places where I just feel strange (where I don’t know people) or places where I feel relaxed and happy (like the cinemas)..”). Also they didn’t find themselves reflecting much more upon the way they feel (it has to be remembered that the test was taken after one week of user testing, quite a short time); it seems that the opportunity to check the data previously collected is the one taken more in account from the users(“I’m curious how the app can “see me” only after my responses every day, cos I know myself quite well :) but it’s funny and interesting to see almost your psychological profile.”). When it came to the interviews the answers were mostly very similar to the ones given to the online questionnaire, but is quite noticeable how angry one of the users I interview was against the Emotion- Sense app: “It keeps on buzzing, why? It buzzes at seven in the morning asking me how I feel, I’m ****** mad at you because you woke me up, this is how I feel!”. Another user referred how skeptical he was feeling in the beginning about sharing his emotional statuses through an app, then he just referred how he was mostly keeping on forgetting about using the app and as well how difficult it was from time to time to answer to the various questions.

Conclusions on EmotionSense

The user testing of EmotionSense revealed this application as clumsy, disappointing and not appropriate for the kind of project I wanted to develop (collect collective emotional view on places around the city); this made me decide to take a bend in the development of my research and to focus in the creation and user-testing of analogic street furniture prototypes of a to be used to create design guidelines. The reason that made me decide to give up the use of EmotionSense to continue my research are both connected to the responses obtained by the user tests and to a personal better understanding of what would have been the best instrument to answer to my research questions. First of all the user testing of EmotionSense made it clear how many people are very concerned about privacy issues: many users were not keen on sharing all their sensor based data, even if it is made clear when starting to use the app that none of the data(used for further researches on behavioural app) can be reconnected to the user that produced it; also, some of the users were feeling very annoyed by the continuous requests to report their mood and the use of the app were reported as not being a trigger for further reflection about specific places and their relation with emotional responses.

23 While conducting the user-testing of EmotionSense, while using it myself and in parallel going deeper in the research of the relation between cities and emotions I understood that a portable and personal instrument wouldn’t have been the right one for the project I had in mind; first, a wearable would only suit people whom are fascinated by behavioural computing and this wouldn’t be a “democratic approach” for a city based emotional map because it would exclude all the people who don’t have the same kind of interests; also the only way to have a real emotional response about something is to ask for it while that “something” is happening (Isomursu et al, 2007), this means that the only way to have a real emotional reaction about a place is to ask for it in the place itself; differently the user that tested EmotionSense were only using the app while at home, which means they were not actively reflecting over places in the city. Those reasons made me decide to change the direction of my research toward a design that could be considered as a part of the city and this is why I decided to enhance the research of what Paulos (2008) calls “street furniture”, trying to understand what kind of guidelines a designer has to follow when developing a city based instrument able to collect citizens’ emotional reaction to places in the city.

EMOTIONAL POLL

Once founded out that a wearable wouldn’t be the right way to gain city related affective data produced by many and different people I decided to try a different approach I already mentioned before, this would be the use of “city located street furniture”; this would be a mean to understand if in- place digital augmentation may be a valuable method to obtain citizens’ point of view (in this case emotional response) about a certain place in the place itself (Schroeter, 2012) and an opportunity to outline design-guidelines for further development of an interactive street furniture. The idea would be to place physical interfaces in specific places around the city that could be used by pedestrian or cyclist to record their mood at a certain time and place. This way I would avoid both privacy issue, the risk of only obtaining data about a small number of people and the risk of annoying my users keeping on asking them “how do you feel?”. Of course this method may not lead to any interesting material, the input station may not be noticed, the project may not be seen as interesting, someone may think about trolling the station and other kind of issues may arise that would make the project useless or not valuable.

First prototype, How are you today?

I built a prototype of the street furniture made out of paper and cardboard and I placed it in different places around Kranen (One of the building where the university of Malmö is located) to see if this approaches works in the relatively small environment of the university. This prototype I built (the first one) is called “how are you today?” and it is composed of an inclined cardboard sheet placed on a stand. The cardboard sheet has an hole in the centre and behind it there is a box. This first prototype was meant to be used to find out:

• if people are willing to share some information about their emotional state, • what are the best places to ask questions about it.

Those are the reasons why the building phase was very fast and I tried to be very economical with the time spent to build it and with the material used. There is an A3 paper with graphics glued on the

24 cardboard; on the top of the paper there is a question: “how are you today?” and under it the instruc- tion that need to be followed: “please choose your mood and drop it here”, an arrow indicate the hole in the middle of a coloured hexagon. Each side of the hexagon indicate one of the six basics emotions: happiness, excitement, calm, anger, sadness, anxiety (Handel, 2012); the choice of the colours for each emotion is based on different theories on colours as stated by Anna Ståhl (2005). On the left and right side of the cardboard are hanging six stacks of papers, each with a different emoticon symbolizing the six mentioned emotions. Under the hexagon there is a statement explaining what is all the object about followed by a Qr-code; the Qr-code is meant for whom is willing to answer to a questionnaire (“This is a research about the relation between moods and places. If you want collaborate further, contact me answering to the questionnaire using the Qr-code.”). On the bottom of the paper there are two last lines:”thank you!” and “have a candy”. Those are meant to reward who is helping me collaborating in my research. Of course on the side of the cardboard sheet there is a cup full of candies.

Image n.11, How are you today?, pictures.

User testings

I tested “how are you today” for three days, each day I placed it on a different area of the C floor in Kranen; the C floor is the busiest one in the whole structure as here is the main entrance, the cafeteria, there are couches and tables where the students rest and work during the day (and sometimes at night as well) and here are the entrance to the corridor leading to the classes. On the first day, a Monday, I installed “how are you today?” in the main hall in Kranen, between the boat and the glass wall, where the study/rest area is. At 17.30 (all the classes end generally around

25 DAY THREE 20

15 DAY TWO

10 DAY ONE

Image n.12, results from “how are you today?”,

25 16.00) I had 18 answers to the physical pool, but no one compiled to the online questionnaire. On the second day, Tuesday, I moved “how are you today?” to a new spot: Kranen’s main entrance just at the end of the stairs, a very busy transit point. While in the middle of the day, around 13.00 thhe number of responses was scarce at the end of the day, at 17.30, the number of “physical” answers was 71 and all the candies were gone, compared to the previous day that was an outstanding result; again the internet questionnaire had no answer. On the third day of my experiment, a Wednesday, I placed the physical poll in front of one of the entrance to the corridors where the classes are; it is important to notice that on that day, 04/30, there was a public celebration for Walpurgis Night (Valborgsmässoafton) which means that probably many people were not attending lectures or being in the building to study. At 19.30 I had 37 emoticon papers in the box, but no one answered to the questionnaire. In total I had 126 responses in three days.

Conclusions on “How are you today?”users testings

The first thing to notice about the different outcomes I had during the three days of my experiment is how relevant the location choose to place “how are you today?” was. While on the first day people had to decide to go and check out what it was about on the second and third day it was much easier to reach the poll; it seems to be very obvious that installing “how are you today?” in a busy transition point gave it more opportunity to the people to notice it and to have the will to answer to it. I think this outcome is not only due to the visibility of the object, but also to a kind of awkwardness people may experience going to check out a new object that is not completely understandable from a distance. Another noticeable outcome is given by the total number of responses the physical poll received (126) compared to the fact that no one used the Qr code to answer to the questionnaire. This shows how important is to set what Schroeter (2012) calls a “low barrier of entry” in order to have many (or at least some) answers. Of course it has to be considered that the Qr code may not have been the best technology to use, in my next experiment I would like to try to attach to my physical device a short link. Another fascinating aspect regards the value of the reward given to who’s participating to the poll: each of the three days the amount of candies consumed was proportional to the number of answers received; based on this observation I will need to take in account to think some kind of compensation in the design of the final Emotional poll. This experiment gave many insights on how to proceed in the designing of the next prototype, but I still have no information on what kind of different outcome I would receive if I gave to users some

Image n.13, map of Kranen and number of responses to “How are you today?” in the three different spots.

26 kind of feedback on the collected data. Would it be a reason to participate more? Would it affect the results as people often tend to be influenced from what others say/do? Anyhow other experiments have shown that displaying in public the obtained results is an effective mean to gather more content and more visibility, Schoroeter (2012) refers that one of the reason for the contributors to share their opinions during the iteration of Discussion in space (Ibid.)was to make their opinion public.

Second prototype, How does this place make you feel?

With my second prototype I tried to understand if giving some kind of feedback on the answers while receiving them would have had an influence on the outcome. As well I wanted to understand if someone would have been interested in giving more detailed answers to the physical poll. I called my second prototype “How does this place make you feel?” as I wanted to address a question that would have had more place related answers. I chose the world “place” because of the deep meaning that the word brings in itself which is not only related to the physical space, but, as Augè M.(1992) states, a place is affected by identity, relations and history, this way I expect the answers not to be only related to the physical state of the space, but also with other more ethereal aspect of the chosen site. While designing this second prototype I decided to give it a more refined look than the first one in order to engage more people with its slightly more professional feature; this time the material I utilized were mostly mdf sheets of 6mm and plexiglass which I cut and engraved using a laser-cutter.

The physical poll consists of a MDF reclined surface of 45*65*0,6 centimetres (placed horizontally) with a squared hole of 30*30*0,6 centimetres in the centre, in this hole is placed a plexiglass sheet; on the top of the MDF surface I engraved a question: “How does this place make you feel?”, the typing has a height of 2.00 cm: big enough to be easily seen by any passer-bye; this is the main concept I wanted to transmit and this is why is written in a bigger type than any other element on the board. Under this written another one, smaller (height: 1.00 cm), says: “Explain why and then choose your mood”. On the left side, next to the hole in the MDF surface there is a written explanation of what all the board is about:”This is a research about the relation between places and moods. If you want to collaborate further, please respond to the questionnaire using the Qr code or the link below.”; under this written there are a short link and Qr code leading to a questionnaire; the questionnaire in this case is the same used during the previous poll. On the right side of the board, on the top, there are two holes, used to fix a rope that holds a pen. The pen is used to write on some slips of paper. The slips of paper are hanging on the right side of the board, on each paper there is written the question “Why?”. Under the bottom side of the boars, under the squared hole there is a final written: “Thank you, have a candy!”, two holes one on the right and one on the left of the written are used to tight a rope that holds a basket where candies can be placed; as founded with the other prototype some kind of reward is useful to obtain more answers and attention. The plexiglass sheet placed in the centre is divided in six areas by horizontal and vertical engraved lines, on the top of each of those areas there are circular holes of two centimetres diameter, over each of those hole there is a slip of paper indicating the six basilar emotions: happy, sad, excited, angry, anx-

27 ious, calm (Handel, 2012). Under the holes there is as well an emoticon, clarifying the emotion written on the top; for each emotion is used a colour as well, the choice of the colour is based on different theories on colours as stated by Anna Ståhl (2005). Once a user has torn one of the “Why?” slip of paper and has answered to the question she is supposed to place it in one of the six holes to indicate which is her emotional state about the place. Behind the plexiglass sheet there are separated MDF boxes, one for each of the emotions that can be chosen, this way the slips of paper stays separated once they are used for voting, the boxes are designed to be open to make it possible to take the responses without having to break the whole prototype. The choice of the plexiglass sheet was made to leave see-through impression and to make it possible for users or simple passer-bys to see the amount of answers for each emotion, this is a simple analogical way to give a feedback to trigger interest and attention. The whole poll is placed over a stand that keeps it at an height of circa 90 cm in order to be easily readable.

Image n.14, How does this place makes you feel?, picture.

Finally the purpose of this prototype is:

• to draw passer-byes’ attention - this is why I chose a refined look, • to be easily understood and used, • to give a feedback on the previous answers and see if this has some impact, • to give different opportunities to answer - just sticking a paper in one of the

28 poll, writing down an answer or answering to the online survey, • to keep to myself the opportunity to change the prototype during the research in order to adjust it on the different founding.

To understand the data I collected I used a legend to evaluate how much the answers given were useful for an interpretation of peoples’ mood in a special context. It has to be noted that I had many doubts while classifying the data I collected and that I don’t consider the legend I am using as a final solution to categorize this data; both the classification and the legend have to be considered as a first attempt to arrange the information I gathered, an attempt that can be improved and made more appropriate. Here is the legend I am using :

Place related: when the answer given is strictly related to the place considered, could be taken in account as something that could to be changed or has a certain configuration based on a human choice (design);

Somehow place related: when the answer given is related to the place, but can’t be changed or wasn’t influenced by a human choice (e.g.: it’s not possible to change the weather, having exams is part of a student life, etc...) ;

Not relevant: when the answer given has nothing to do with the place considered;

Troll: the answer given is an insult or a goliardic joke.

user testing, inside the university

Before taking this second prototype outside in the city I decided to test it again inside the university’s buildings to see how it worked and if there was some major issues in its usage.

Image n.15, two students interacting with “How does this place makes you feel?” while on The bridge.

29 Day one, The bridge 9,7% 22,6% The first time I experimented the “How does this place make you feel?” prototype was a Thursday from 08.00 in the morning until 18.00 in the afternoon19,4% on a very sunny day. I placed the poll on the covered bridge that connects Kranen with another university building, Ubåtshallen.14,5% The bridge, which is basically a crossable steel truss beam buffered in see-through glass, is about 2,5 m wide and 13 33,9% meters long; it is suspended at about 7 meters from the ground inclined by about half a meter on the Kranen side. The bridge connects the C floor in Kranen (the one where the Cafeteria and resting areas are placed) with a passage point in Ubåtshallen, where stairs and an elevator are.

place related = 14 9,7% 22,6% somehow place related = 9 19,4% white = 21 14,5%

not relevant = 12 33,9%

troll = 6 Image n.16, Pie chart showing the kind of responses collected with “How does this place makes you feel?” on Day one, The bridge.

place related = 14 I’ve chosen this place for different reasons: on top of all many people cross the bridge every day, someone because they havesomehow lectures place in different related = 9 building and someone else to have their lunch break in Ubåtshallen’s Cafeteria. Also this place has a strange nature: it is a corridor, a room with an utilitar- white = 21 ian mean, where people go fast and never stop, but it has as well a nice view: the glass coating makes this corridor a bit less brutal,not relevantoffering = 12 the opportunity to who’s crossing it to have a sense of what’s going on in the world outside and to see some daylight, when it actually happens to be a bright day. At the end of the day the polltroll =had 6 received 62 answers, 14 of those answers were giving an important insight on how the place itself was mood influencing and other 21 were left white. No one answered to the questionnaire reachable through the short link and the Qr code.

0 EXCITED HAPPY CALM ANXIOUS ANGRY SAD Image n.17, results from “how does this place makes24% you feel?”,16% day24% one,13% The bridge.7%

25 Mostly it seems like people were positively influenced by the light and the opportunity to look outside as is evident from answers like: ”20 the light and the view”, “see-trough bridge”, “nice view” founded in the HAPPY pool box that got 15 15total answers. The same amount of answers was received by the

0 EXCITED HAPPY CALM30 ANXIOUS ANGRY SAD ANXIOUS pool box where people were more focusing on the passing through nature of the bridge / corridor: “hate the tunnel”, “strange slope of the floor”, “this place shakes and everyone is on his way to something and the air is bad”, “airport feeling”.

Day two, Orkanen

The second time I tested “How does this place make you feel? “ I did it in a new environment: Orka- nen, Malmö University library; I tested it three weeks before the end of the semester, on a Friday from 8.00 in the morning until 17.00 in the evening, being many students occupied with their exams and thesis project this was a very busy end of the week. Orkanen Library is a modern building, recently built (2005) appositely for Malmö University, the library floor is a huge open space, with different areas where student can study, tables furnished with pc, sofas, a cafeteria and, of course, shelves with books distributed in all the room. As Orkanen has been built by the side of Malmö’s harbour and has huge glass walls from which students can enjoy the view.

Image n.18, students interacting with “How does this place makes you feel?” while in Orkanen.

I choose the University Library because of the large number of visitors it has everyday and because of the diversity of student going there, different ages, nationalities, needs, study courses, disciplines and interests; as well it is a place specifically designed for well being16,5% and to give opportunities to students to concentrate and bound at the same time. I decided to place the physical31,6% poll just outside the entrance of the library, on the fifth floor, as I thought this would have been a good way to indicate that the question asked, “how does this place makes you feel?”, was referring31,6% to the environment of the library, not to the entire building, where also other rooms can be found, and not20,3% even to a specific spot. This second day got 79 answers, 25 of which were specifically place related and 25 were left white. It is noticeable how this time I got no troll answers; as well I received the first answer to the online questionnaire.

16,5% place related = 25 31,6% somehow place related = 16 31,6% white = 25 20,3% not relevant = 13 Image n.19, Pie chart showing the kind of responses collected with “How does this place makes you feel?” on Day two, Orkanen.

place related = 25

somehow place related = 16

white = 25 31 not relevant = 13 During a short close up observation session (from 16.00 to 16.30) I took note of the number of pas- 25 ser-byes, people noticing the board and people actually using it. 72 people used walked in though the entrance, where I placed the board;20 of this 72 people 39 were leaving the library and 32 were walking in (54% leaving, 46% entering); of 15those 72 people 45(52,5%) ignored the board, 23(32%)noticed it and 4(5,5%) gave an answer. Of all the people acknowledging/interacting with the board most of 10 them were leaving the library (20 of the 23 people noticing the board and 4 of 4 people leaving an answer); only one out of the four students5 leaving a reply to the board was on its own, the other where with friends who: took pictures of the0 board, ignored it or took some of the candies. Some people just stopped bye to understand what it wasEXCITED about,HAPPY someoneCALM ANXIOUS elseANGRY tookSAD the candies without watching it, others commented it with the friends they were24% with,16% but24% they13% didn’t7% touch the board.

0 EXCITED HAPPY CALM ANXIOUS ANGRY SAD Image n.20, results from “how does this place makes you feel?”, day two, Orkanen.

Evaluating the answers received it seems like the students like the building and that it mostly re- sponds to the reasons why it was designed: “nice view, get work done, see friends”, “ Nice atmosphere, shared space, people” and “Because usually I get things done in the library, which is a relief” (in the HAPPY box); I found it interesting as well the reasons that make people angry about the library and that could actually being taken in account to make an even better place out of it: “Bad opening times – should be 24/7”, “ Way to much noise and people constantly talking on their mobiles” and “Because you are not allowed to eat”.

Conclusions on “How does this place make you feel?” Users testing inside the University

The results I collected with this second prototype were many and very interesting and so far confirmed my design choices. The prototype received 141 answers in two days, of those 46 were white papers without anything written, while 95 had some kind of writing/drawing; this means that a third of the people who chose to participate chose to do it as fast as they could. The rest (over two third of the whole participants) decided to give some kind of open answer; of those, 6 gave troll answers, but only during the first session, inside the bridge, which may suggest that those answers were given by someone knowing that no one was looking at her/them, it is interesting to notice that 3 of these troll answers were placed in the ANGRY poll; only 25 people gave not relevant answers. All the other 64 answers (45,3%), place related and somehow place related are suggesting that the most

32 of the other participants understood what they were asked and that the prototype actually triggered their desire to share and collaborate giving answers that could actually being taken in account in order to make the place taken in consideration a better place. The first thing that I would take in account evaluating this second prototype is that it was a good choice to give the opportunity to people to respond to the poll with an open answer, but I would keep in mind for further development of the public pool that it is important to leave the chance to give an answer as fast as it is possible in order not to loose collaborators because of a high barrier level (Scho- roeter, 2012). 4,2% It seems that the prototype is noticed/used mostly by people17,7% a the end27,7% off their working day or at the beginning of a break, so,it seems that who has a task to complete tends to be less interested in her own surrounding and tends not to share her time to answer to a public poll; through the close up observation I had the opportunity to notice people not only noticing32,6% the board17,7% and reading the instructions, but acknowledging the previously given answers as well; this probably means that a design that gives a feedback on the kind of answers the poll is receiving may not trigger the will to collaborate sharing one’s own opinion, but generate some sort of reflection upon it (Schroeter, 2012).

4,2% place related = 39

17,7% 27,7% somehow place related = 25

white = 46 32,6% 17,7% not relevant = 25

troll = 6 Image n.21, Pie chart showing the kind of responses collected with “How does this place makes you feel?” during the two days of User Testing inside the University.

place related = 39 Another thing that was quite noticeable both during the close up observation than while walking in front of the prototype duringsomehow the first place and related the = second 25 day of the experiment was the fact that it is easier for people to answer when they are with someone else and that, when in a small group (up to three white = 46 people) only one person generally answers to the poll. not relevant = 25

troll = 6 User testing, out in the city

The two days of testing inside the university’s building gave me the proof that the prototype was a valid instrument for my research and that it was finally time to test it “in to the wild”, which means outside in the city, and finally see if it could become a valid launch pad for the development of a more complex electronic prototype for the gathering of citizens’ emotional responses to places around the city. I chose to carry out this new experiment in Södervärn public transport hub situated in the centre south of the city.

33 Södervärn

Södervärn is a neighbourhood of Malmö in the district of Södra Innerstaden, the southern boundary of the city centre; there is situated a public transport hub. Södervärn is delimited on the north-est side by Spårvägsgata, a fast moving two way road, on the south-est by Nobelvägen, a major street and an important north-south link between Lundavägen the north and Trelleborgsvägen; on the south-west side it is delimited by Dalaplan, an old and important traffic hub, on the north-west by Södra Förstads- gatan that connects from David Hall Bridge to Dalaplan. The area is strongly characterized by the presence of the public transports hub; on its south est side it has a residential nature which mostly consists of a residential development of four floor built in the 1930s. On the south east corner of the neighbourhood there is a school, which was built in the 1924 and that is today headquarters of the SFI institute, a school where newcomers from other nations are taught Swedish; on the side and behind the school there are two green areas. On the north-west side there is big parking garage that was built in 1963. A petrol station is situated in the south-west side (Reisnert A. et al, 1989). On the north-west of the neighbourhood of Södervärn there is the city hospital , in the same area some of the University’s buildings are situated as well. On the north- east there is the neighbourhood of Möllan, which is a multicultural area with mixed functions. The area on the south of Södervärn hhas mostly a residential characterization. public transports

27 different buses stops in Södervärn, 9 running through the city, 16 regional buses directed to 11 different town and cities outside Malmö, one bus directed to Malmö airport and one directed to Co- penhagen.

Image n.22, Map of Malmö’s public transports, only buses going through Södervärn are highlighted.

34 history

Södervärn is a former railway station and city district in Malmö, Sweden. Södervärn railway station was added in 1886 at the Malmö-Ystad railway. Besides Södervärn School, built in 1924 and a few older apartment building (now reconstructed or demolished), the area has been expanded fairly uniform from the 1920s last years to about 1940. Adjacent to the station, on the north of Spårvägsgatan, there is the area called Möllevången. In 1907 Södervärn was reached by the electric tramway line 1 (also line 2 operated for a short period in 1907 Södervärn railway station); In the late 70s, both the station building as well as other remnants of the railroad and the tramline era were tear down and railways were converted in to a bus station (regional and local transport). why Södervärn?

There are many reasons why I chose to test “How does this place make you feel?” in Södervärn. On top of all the fact that this is an important public transport hub of the city, a transition point, that may risk to become what Augè (1992) would consider a non-place: an heartless, franchised space where humans are dis-individualized and act in it only as consumers; differently, I think, Södervärn preserves its own peculiarity and I wanted to have a peak in to it trying to understand the needs and emotion of the humanity that pass through it everyday shaping its personality as a place. Of course what is also fascinating about Södervärn is the incredible number of different people walking through every day: different ages, social statuses, nationality, all together in the same space for few minutes day by day; somehow, being there so many different bus stop there is also the opportunity to have a crass idea of who is travelling on which bus, which means who is stopping at which bus stop and this information could become very helpful to get a better understanding of the data I obtain using my prototype.

J C ar F l G rick us sga ta tan fs v äg n n a a t L t a a g S g n p s a å e rv a B I äg l A s C A ga hlm ta an n sga S F B M tan ö

K d

r G a

F C ö E

d D

s g S a Sigtunagatan p t å a rv n ä gs ga H ta Image n.23, Map of Södervärn traffic hub; bus stops are highlighted. n day one

The first time I took the prototype out in the wild was on a Thursday afternoon, from 15:10 until 17:40. This first time I didn’t want to leave “How does this place makes you feel?” on its own, so I pre-

35 tended to be waiting for the bus and kept a look on it: I was afraid that it would have been vandalized and, even if this would have been an interesting plot twist for my research, I still needed to use it some more time.

Image n.24, The prototype at the bus stop A.

I placed the prototype where the Bus stop A is situated (See image n.22), here there are three buses stopping: • n.1 direction Kristineberg, in the south east of the city of Malmö. • n.3, ring line direction Värnhem, going all around the centre of the city and • n.35, direction Kvarnby, going to the east of the city after passing through the south;

40% 40% I chose this particular bus stop because I was expecting many people to be coming back from the centre of the city to go back home in the suburbs and from my previous experiences it seemed that people are more keen to answer to the board when they are at the end of their task. Also, as I think I can consider line 1 and line 35 “commuter lines” the people answering20% 20% were very likely to be used to stay at that particular bus stop every day for a long time, and they may have probably developed some sort of emotional response and maybe some opinion about the place.

place related = 8 40% 40% somehow place related = 2

white = 2 20% 20% not relevant = 8

Image n.25, Pie chart showing the kind of responses collected with “How does this place makes you feel?” on Day one, bus stop A.

place related = 8 On this first attempt of user testing I made in Södervärn, after 2 hours and a half I collected 20 -an swers and 8 of them were placesomehow related place and related meaningful = 2 for my research.

white = 2

not relevant = 8

36 Image n.26, people interacting with “How does this place makes you feel?”, Södervärn bus stop A.

While standing at the bus stop and looking at people answering I had the opportunity to observe very different reaction of the people seeing the poll. It attracted more attention than interaction, probably the fact of being written only in English could have been an obstacle for many more than one person (this time the 40% of the answers was in Swedish). From where I was standing I could see people’s reaction as soon as they understood what the poll was about, it was going from almost disgusted and annoyed to puzzled, from engaged to fascinated. I had another confirmation that people in a group feel less threatened and are more keen to leave a feedback: a group of five young people, four men and a woman, got interested by the prototype and five of them answered to it.

0 EXCITED HAPPY CALM ANXIOUS ANGRY SAD Image n.27, results from “how does this place makes you feel?”, day one, Bus stop A. Another important thing I noticed15 is that people weren’t interested to the candies, probably because of the presence of two homeless guys12 in the surrounding; parent’s were keeping their children away as they were scared that the candies could have been not clean or safe, what was a “bait” inside the uni- 9 versity had become a reason to stay away from the prototype once outside in the city. 6

0 EXCITED HAPPY CALM ANXIOUS ANGRY SAD

6 Image n.28, The prototype at the bus stop D. 3

0 EXCITED HAPPY CALM ANXIOUS ANGRY SAD

37 6,9% 17,2% 6,9%

10,3% Evaluating the results it seems that people are not happy about this place, and are keen to complain about it, dirt and messines are reasons for anger, while the crowd58,6% and the cars (So many people and People vs cars = accidents) are a cause for anxiety . day two

6,9% place related = 2 17,2% 6,9% somehow place related = 2 10,3% white = 17 58,6% not relevant = 3

troll = 5 Image n.29, Pie chart showing the kind of responses collected with “How does this place makes you feel?” on Day one, bus stop D.

place related = 2 On the second day (a Monday from 13.30untill 20.30) of user testing in the wild, I placed “How does this place makes you feel?” insomehow a different15 place relatedspot of= 2Södervärn, at the bus stop D; I chose this specific spot because it is situated by a greenwhite area= 1712 facing Spårvägsgatan. My aim on this second experiment was to see if the different location, more open9 and looking better than the previous spot (dirty, grey and busy) would have influenced thenot mood relevant of =the 3 people answering. This time I left the prototype on its own and I preferred to place it inside the6 waiting shelter in order to protect it from the wind and from pas- troll = 5 ser-byes; based on the observations3 I made the previous day I slightly modified the prototype: where the writing :”thank you! Have a candy!” appeared I now wrote only a “Thank you!” and I removed the 0 bucket with the candies. EXCITED HAPPY CALM ANXIOUS ANGRY SAD At the bus stop D there four regional buses stopping all directed to towns in the north of Malmö:

0 EXCITED HAPPY CALM ANXIOUS ANGRY SAD Image n.30, results from “how does this place makes you feel?”, day one, Bus stop D. 15

• n.132, directed to Landskrona,12 • n.133, directed to Lomma, • n.134, directed to Löddeköpinge,9 • n.138, directed to Landskrona;6

3 This time the number of answers were proportionally less, only 29 in 7 hours, and the most of them 0 where white answers, also this time theEXCITED pollHAPPY receivedCALM ANXIOUS a quiteANGRY highSAD number of troll answers.

38 5%

25% 32,5% On this second day none of the answers received can really be considered meaningful, but it seemed like the general mood of the answerer was way better than10,3% during 25%the first day in Södervärn. Also the presence of many troll responses may indicate a ludicrous mood. day three

5% place related = 13

25% 32,5% somehow place related = 10

white = 5

10,3% 25% not relevant = 10

troll = 2

Image n.31, Pie chart15 showing the kind of responses collected with “How does this place makes you feel?” on Day one, bus stop F. 12 place related = 13 The third and last day of the Södervärn9 experiment, a Wednesday, I left “How are you today on its own from 10.00 to 22.00 insomehow front of place the related bus stop = 10 F. I chose this particular spot because all the buses going there are going to Lund: 6 white = 5 3

• n.169, directed tonot Lund, relevant0 = 10 • n.171, directed to Lund; EXCITED HAPPY CALM ANXIOUS ANGRY SAD troll = 2 Lund is a city located about 20 km from Malmö. The city is well known because of its big and ancient 15 university campus and many of the commuters going there every day are students, researchers and professors. This last “in the wild” test12 could have been more easily compared to the previous one made because of this very specific audience:9 similar to the one I found inside Malmö University. This time people participating where proportionally not that many (the poll received 40 answers in 12 hours),but more collaborative6 as I received a lot of answers and many of them were useful and properly answering to the question 3“How does this place make you feel?”; on this occasion the online questionnaire received a second answer. 0 EXCITED HAPPY CALM ANXIOUS ANGRY SAD

0 EXCITED HAPPY CALM ANXIOUS ANGRY SAD Image n.32, results from “how does this place makes you feel?”, day one, Bus stop F. A reason to be HAPPY for the Lund commuters seems to be the crowd and the multicultural situation of Södervärn :”A lot of people in motion, people are generally very nice” and “make me happy to see

39 many culture at one place”. What does trigger SADness is the7,9% environment which is described as:”no colour. It seems cold and sad”; also it seems that the dirt, the absence25,8% of seats and the disorganization are a reason of AnxietY: “here is not clean and the seats 23,6%are not nice”, “ too little benches to sit on”, “ bad information about where the bus stops are placed”. 15,7% 26,9% Conclusions on “How does this place make you feel?” Users testing out in the city

place related = 23 7,9%

25,8% somehow place related = 14 23,6% white = 24 15,7% 26,9% not relevant = 21

troll = 7

Image n.33, Pie chart showing the kind of responses collected with “How does this place makes you feel?” during the three days of User Testing in Södervärn. The results I collected duringplace thisrelated user = 23 testing out in the city confirmed most of the finding I obtained during the iteration of the testsomehow inside place the related university’s = 14 buildings. The prototype received 89 answers in three days (21 hours in total), of those 24 were white papers, 65 had something written; whitethis means= 24 that a forth decided to participate, but to do it the fastest way. The rest 3/4s of participants spent a little longer time sharing some kind of comment; I collected 7 troll responses, but none of themnot was relevant placed = 21 in the poll during the first day; on this day I collected a third of the answers written in Swedish;troll = 7 differently from the other all the troll answers were placed in the happy, in the excited or in the calm poll. 21 of the written answers, slightly less than a forth, were not relevant, which means not related to the place. Less than an half of the participants (37) left comments that I considered as place related or somehow place related; this may be connected to a language barrier, probably the use of English was still making sense inside the university, but may have been an obstacle to the understanding of the request outside in the city.

As in the testing inside the University, to give the opportunity to answer choosing how much to share is a winning design choice as over a forth of the participants chose to leave their responses white and it’s an aspect that has to be kept in mind for a development of the prototype from it actual analogical nature to an electronically interactive system. It has to be noticed that on the first day of test, which lasted for 2.30 h, from 15.10 until 17.40, the poll obtained 20 answers, while on the third day, when the poll was left in the test place for 12.00h, almost five times longer, the number of answers (40) was only doubled; this seems to confirm the fact that people collaborate more at the end of their working day, but I would need some more experiment to be sure of it. This time I had again the opportunity to notice how people often check out the poll to understand what it is about and to examine previously collected answers; this seems once more to confirm that a design that gives a feedback triggers some sort of reflection (Schroeter, 2012). I found again that it is easiest for people to interact with the poll when they are in a group, but differ-

40 ently from the previous observations not only one in the group answered. In the end the biggest differences I found between the tests’ iteration inside the university’s buildings and the tests outside in the city is the difference in the amount of trust people seemed to have toward the poll and who built it and less willingness to cooperate. I noticed the trust issue when it came to observe people interacting with the candies I was leaving to thank who was collaborating: while in the university the presence of the candy worked as a way to make the poll more noticeable and appealing, which means that there was some kind of unconscious trust, outside in the city it seemed not be noticed and sometimes almost seen as threatening; mostly when people with kids were approaching the prototype as they were probably fearing that there could have been something wrong with the candies and they didn’t want their kids to touch/eat them; this is why I removed the candies during the second two days, I think anyway that there should be some way to reward who is participating to the poll. When it comes to the willingness to cooperate it is noticeable how during the test inside the Univer- sity’s buildings I collected 141 responses in 19.00 hours, while, outside in the city, in 21.30 hours the number of answers was 89, which means that on the outside testing gathered proportionally 44,4% less responses than the test inside the University; somehow it seems like my fellow students, maybe for a sense of identification, were more eager to help me to bring forward my research;of course this aspect should be investigate further as the reasons for this differences in the number of answers may be connected to many other reasons such as time disposal, positioning of the poll, language barrier, understanding what’s the purpose of the project (what does “a research” mean) and many others.

41 07. CONCLUSIONS and design guidelines

In this chapter I am going to discuss how I answered to the previously asked research questions, what are the foundings of my exploration and what is my knowledge contribution in form of guidelines to be followed when developing a place based interactive instrument to use to collect citizen’t emotional reaction places in the city.

The questions I was asking at the beginning of my research were:

At the beginning of my research I had in mind two different path I had to experiment in order to answer to the research questions; they were the building of a wearable to connect with an existing behavioural app (EmotionSense) or the investigation of the use and develop of street furniture, in the end I decided that the latter was the right mean for the sake of my research. The first reason I based my choice on was that when trying to obtain emotional responses about something it is very important to ask for the response while this something is happening (Isomursu et al, 2007), and this has to be applied as well to emotional responses about places: they have to be solic- ited in the place itself and this is one of the reasons why EmotionSense and the idea to build a wearable to collect place related emotions didn’t seem to be the right choice. Another thing that should be kept in mind is that even there are many people disposed to collaborate in public researches and ready to share their emotional status, they are often not glad to constantly being asked for a feedback, which is one of the stronger dislikes EmotionSense received during the test, it seems that the opportunity to answer and to decide how much to answer is much more valuable and makes people more inclined to participate; in fact the decision I made building the second prototype (“How does this place make you feel?”) to give the opportunity to collaborate choosing between an open answer or a simple vote (which most of the participant choose) revealed itself winning: this way I both set a low barrier level for participating, but I also gave space for a comment to who felt more involved in the cause (Schroeter,2012), giving a choice has probably assured me more participants. A further important aspect that influenced users’ judgment about EmotionSense and consequently my decision to move on with a place located physical input is the privacy issues that a smart-phone app collecting personal information brings with itself; differently, answering through an object which does not have and does not ask for our personal data makes people more eager to participate making the whole inquiring and asking operation more reliable and trustworthy (Picard et al, 2012). In the end I found that the use of a behavioural app wouldn’t have been the right mean for a large and collective use as those instruments capture the interest of a very small number of people while I was looking for an instrument capable to attract the interest and attention of the largest number of people possible.

42 After I found out that a street furniture would have been a better choice for my research, as it seemed a better way to trigger reflections about the relation between places and emotional responses , I started to experiment through physical prototypes, I used candies as a reward offering them as a thanksgiving to who collaborate to my research. This choice seemed to be very fruitful and it attracted a lot of attention inside the University’s buildings, but not when I experimented the second analogical prototype outside in the city. My conclusion about this matter is that it is effective for the research and fair to who is collaborating to share something with them as a compensation, but, outside a close environment, the reward shouldn’t been something potentially risky as food. This difference in the response to the candies as well the lower number of answers once I moved the prototype outside in the city is probably due to a different level of trust and commitment: the Univer- sity could almost be considered a close environment, where everyone is there for the same reasons: there is a mutual recognition, identification and unconscious trust in between the students and toward what is happening inside the buildings; differently, seems that people attending Södervärn do not have the same sense of belonging and mutual expectations, they are still stranger to each others; this could explain why inside the University students were more eager to participate, as they understood the meaning what was written on the prototypes (“This is a research...”) and they knew that who was doing the investigation must have been a student like them, one of them, and this made more trustworthy the candies as well. Differently in Södervärn the sense of what a research is was probably blurry and wasn’t triggering any kind of identification to most of the people, which probably mean less interest to help and to participate; it is easy to understand as well that they there were just following a rule of common sense: ”do not accept candies from strangers!”. Those considerations may also explain why people seemed much more eager to interact with the poll where they were in a group. Another aspect that should be taken in account is the feedbacks sharing, which of course can only be partial through an analogical mean as the second prototype I built, but that seems to attract interests in the passer byes, that, even when they decide not to participate actively, as they can get an idea of the amount of people who answered and how those people felt about the place; this can be considered as a further expedient to involve people to collaborate and reflect. One more important aspect to take in account while designing an instrument to collect place related emotional data is where exactly to collocate it as its location influences both the number of answers received, as seen with the first prototype I experimented (How are you today?) but also what meaning has to be applied to the world “place”, as seen during the user testing inside the university of the second prototype (How does this place makes you feel?); when I placed the poll inside the bridge, on day one, who left a response commented the bridge itself, not the whole building, while, when I left the prototype in front of the library’s entrance the responses were related to the library, not only to the exact place where the poll was positioned. This means that the way an instrument like the prototype I built is collocated expresses and signifies a lot and has to be a thoughtful operation. Probably, to communicate the boundaries of the place I am asking about it would be relevant to use a more appropriate phrasing, but this is a matter that I just started to approach and needs a deeper investigation, as well some further reflection has to be done on the language used. Through the observation people’s behaviour while using the prototypes I also found that the time of the day may have an influence on people disposal to answer to the poll: it seemed in more of an occasion that the end of a working day is much more fruitful that the beginning of it, but more experiments and analysis are needed to prove it; Based on what I found during my research my conclusions are that in order to trigger a reflection in to people on how their emotive status is influenced by places and in order to make people sharing

43 those reflections the design of street furniture seems to be fruitful; in the next paragraphs I am going to outline some guidelines to be used to further investigate this kind of design. design guidelines

Here I am going to list the guidelines that should be used in order to go deeper in the understanding and designing of a street furniture for the collection of citizen’s emotive responses triggered by places in the city, based on the user tests carried out in this thesis:

Reliability: 1- Who is asking the question should be trustable and well known. 2- The reason behind the project have to be clear. 3- The street furniture must have a refined look. 4- It must be clearly stated how users’ privacy is protected.

The place: 5- The question asked must be related to the place where the question is asked. 6- The choice of the placement matters.

Respecting the user: 7- The istrument must not be instrusive. 8- Users must have the choice to participate or not to participate. 9- It is better to let the user decide how much to share (different ways to answer) 10- A lower barrier of entry helps to reach more participant.

Avoiding barriers: 11- Make the instrument easy to understand. 12- Avoid language barriers.

Reward the user: 13- Rewarding who is participating. 14- Give feedback. 15- Transparency.

To remember: 16- Remember: at the end of a working day people are more willing to share. 17- Remember: people are more willing to share when they are in a group.

44 08. FURTHER IMPLEMENTATIONs

In this chapter I am going to suggest which could be the possible context, stakeholders and consideration to be taken in account for a practical implementation of the whole project. As previously mentioned in the research introduction, the main idea of “How does your city feel?” would be to create an emotional map of the city based on the mood collectors I developed as street furniture. The emotional map of the city should be used both to understand the city and citizen’s need, and both to figure out what needs to be changed, what is missing what would make the city a better place to live in. In my idea the project should be promoted by the city for its citizens, to give them a space to discuss how the city should be developed and in order to defend their interests, this approach would give to both urban planners and administrators a tool to understand the needs of city and to base their decision upon. different figures:

Here I am describing the different actors that could take part to the developing and the functioning of the project. It has to be noted that, while administrators and planners are taken in account for the sake of the actualization of the whole project, the citizens are seen as the central figure in the developing and in the growing of this “urban experiment”. Citizens are the main figure in this project, not only because it is intended to make their environment better, but also because the functioning of “How does your city feel?” depends on their will to be involved and to share their feelings, opinions and knowledge.

Promoter- City administration

As mentioned in the “Design Guidelines” an important aspect when asking people to collaborate and share their emotional state or thought is connected with the reliability of who is asking for collaboration(1- Who is asking the question should be trustable and well known.). In this perspective the best promoter for this kind of instrument and project is the city administration: people working for the interest of citizens; this approach would give great meaning to the whole project giving citizens a valuable space where they would be heard and where their view on the city and their opinions would be shared with someone that could take real decisions and make real changes.

Citizens

“How does your city feel?” is thought for citizens’ sake and puts citizens’ need in the centre. Citizens are asked to share the way different place in the city influence their emotive status, they are asked (though the interactive map) to understand how the other citizens are seeing the city as well; they can decide to come up with their own interpretations and opinions, and their requests are publicly shown and have to be taken in account when decision about the city are taken.

45 Urban planners

The last figure of the whole project are urban planners. I think this is a very important last step in the whole process; citizens are seen as central actors in the whole developing and growing of “How does your city feel?” as they are the users of the city, they are the ones that better know what is the city like to be lived, urban planners, differently, are seen as the ones that are able to better interpret what are the requests that citizen have, what are the best ways to respond to those requests; the whole “How does your city feel?” project has to be seen as a new powerful tool to give to those who work and design the changes in the city, a tool that gives them the chance to work not only with the physical and economical facets of urban planning, but as well with the human aspects which is going to greatly amplify their understanding and opportunity to make the best choices for the citizens and the city itself. the program:

Advertising

An important aspect to obtain attention and collaboration has to do with the understanding of the kind of project people are asked to collaborate for (2- The reason behind the project have to be clear.), this is why the first step toward the developing of a map filled with citizen’s emotions would be to advertise the project, explain what is the idea behind it and why it is worth for the citizen’s to collaborate; they should know that there are going to be those new instrument in the streets (the interactive street furniture), they should know how to use them and why, citizen’s are supposed to know that the result of their collaboration is going to be collected and showed through live data on a web based map of the city they could consult, use and comment.

Collecting

This is the (ongoing) phase where the interactive street furniture would be used; probably, as a first iteration of the project, the street furniture should just be placed in few selected-spots, a way to understand how to get the most out of them and to structure the whole project step by step. While the whole system is gaining it own shape and complexity more interactive street furnitures could be placed around the city with the aim, in the end, to cover the whole urban area.

Showing and Understanding

All the data the street furniture are collecting are going to be showed on an interactive map of the city, accessible and open to everyone; this map should be displaced in the city, but it should as well be reachable on a dedicated website. On this interactive map users and visitors should be able to see the data and make sense of them in relation with time of the day, happenings in city, weather, seasons, there should be the opportunity for all the users to cross the existing data with other spatial data, such as kind of services in the area, in- habitants’ income, perceived and actual security, traffic and all the kind of data and information users

46 would consider valuable to have a better understanding of how and why the city is perceived. The aim of the map is to use the power of crowdsourcing and citizen’s love for their city to understand what is missing in the city but as well to understand what works in the city, in order to make it a better place for everyone; the choice of a map and of the use of spatial data is made because on the impact that this kind of representation carries within itself often giving a much clearer and understandable vision of how some events and situations are connected and influence each others. The main features of the map should be the opportunity to easily add material on it, to navigate all the information displayed, to comment what has been found and to make connection between all the data. Making all these step easily accessible to everyone would be an impressive way to empower citizen in the act to understand their city and ask for changes to make it a better place.

Making changes

While the understanding of the city should be an ongoing process that should keep on happening and that can never be considered as completed, the information and the citizen’s opinion obtained have to be used to make effective changes in the city. Those changes would have the strength to be done under the eyes of the citizens, under their request or at least under their guidelines. This process would empower both the administrators and the planners in their decisions as they would have real requests to fulfil, a real understanding of what areas in the city are appreciate and what part of the city need to be changed how and why.

To conclude, with this project, I am hoping to give a space to people where they can learn to understand their city, where they can grow more affectionate for their fellow citizens, and where their voice can be heard by administrations and urban planners giving them the best instruments to make the best decisions to make cities better places for everyone.

47 09. Appendix

How does this place make you feel?

Here are reported the answer received by the “how does this place make you feel?” prototype. The answers are grouped based on the five emotions that were asked to be expressed, and then grouped again based on the information given to justify the choice of a certain emotion. The asterisk (*) means that the has been translated from Swedish.

Day One, The Bridge = 62 total answers

Excited = 10 answers Place related: 1 • the light Not relevant: 6 • candy • uninterpretable drawing • drawing of a happy / excited person → “That’s why” • I get candy! • Exciting* • coffee!! Blank answers: 3

Happy = 15 answers Place related: 5 • the light and the view* • see-trough bridge* • happy courses = happy place • the light* • nice view Somehow place related: 4 • sunny • nice weather* • sun (drawing) • sunny Not relevant: 1 • ‘cause I am happy Troll: 1 Blank answers: 4

48 Calm = 10 answers Place related: 2 • I can see outside* • calm atmosphere* Somehow place related: 1 • I’m sleepy Blank answers: 7

Anxious = 15 answers Place related: 5 • this place shakes and everyone is on his way to something and the air is bad* • hate the tunnel • strange slope of the floor* • airport feeling • tired Somehow place related: 3 • stressed out • stress • stress Not relevant: 2 • the future • related to my stress level right now, but most I feel happy* Troll: 2 Blank answers: 3

Angry = 8 answers Place related: 1 • downhill way* Somehow place related: 1 • heavy exams* Not relevant: 1 • didn’t sleep Troll: 3 Blank answers: 2

Sad = 4 answers Place related: 1 • The computers don’t work Somehow place related: 1 • weather bad Blank answers: 2

49 Day Two, Orkanen = 79 total answers

Excited = 10 answers Place related: 3 • its encourage me to study very well * • A lot of opportunities * • I can find interesting books in ENGLISH Somehow place related: 1 • hopeful to be finished with studying * Not relevant: 3 • Too much coffee! • thereto • unreadable Blank answers: 3

Happy = 21 answers Place related: 9 • nice view, get work done, see friends * • Because I am about to find a book I was looking for * • Beautiful view • Books, Films , Free! • Nice atmosphere, shared space, people * • Make me feel good! I like it! • Because usually I get things done in the library, which is a relief * • Nice Place • lovely place Somehow place related: 1 • cool mood, good looking colleaques Not relevant: 1 • Depending how bad my thesis goes* Blank answers: 10

Calm = 17 answers Place related: 5 • it’s light * • The atmosphere and the sunlight from the big windows makes one calm * • Light and open, nice design • So much space, clean atmosphere * • The beautiful view and the silence Somehow place related: 5 • I can work in peace in concentration * • easy to concentrate • task-focused • Because of the silence and the focus that are apparent in the library * • Because I can work with my studies and be finished Blank answers: 7

50 Anxious = 15 answers Place related: 2 • Anxious because its too noisy • Noisy Somehow space related: 8 • Deadlines Exams • I think about my studies stress, nervous • The need to deliver/achieve • Thesis work • I’ve been working on my BHD all day here and still have no hope that it’ll pass • I always have to do school-work here • represents the lack of time for school-work • Because I am stressed about an exam Not relevant: 3 • Because I am myself * • Should have worked more • Stress Blank answers:2

Angry = 11 answers Place related : 6 • Bad opening times – should be 24/7 • Way to much noise and people constantly talking on their mobiles • Because you are not allowed to eat * • High noise level * • Booking the quiet room to often * • The curtain going down every to minutes when there is sun * Somehow place related: 1 • want free coffee! * Not relevant: 4 • I have an exam * • Because I am stuck with my thesis research topic * • Having a re-exam • capitalism

Sad = 5 answers Somehow place related: 1 • I want to travel Not relevant: 1 • Everything Blank answers: 3

51 out in the city

Day One = 20 total answers

Excited = 1 answer Not relevant: 6 • uninterpretable

Happy = 1 answer

Not relevant: 1 • sweets

Calm = 2 answers Not relevant: 1 • Nothing has happened! That‘s OK!* Blank answers: 1

Anxious = 3 answers Place related: 5 • So many people • People vs cars = accidents Not relevant: 2 • unreadable Angry = 12 answers Place related: 4 • -messy* • -dirty! • -Cigarette smoking • -Dirty downhill way* Somehow place related: 2 • -weird bus times* • -when bus delayed* Not relevant: 5 • -because of all the immigrants* • -Because I think of EU • -Cannot describe my feeling* • -Hej* • -You are standing in the way* Blank answers: 1

52 Sad = 1 answer Place related: 1 • Catastrophy in the environment*

Day Two = 29 total answers

Excited = 6 answers Not relevant: 1 • Social Democrats* Troll: 3 Blank answers: 2 Happy = 7 answers

Place related: 1 • because I like Malmö Not relevant: 1 • Because I am Albanian Girl and I am happy because I can speak English. Troll: 1 Blank answers: 4

Calm = 12 answers Place related: 1 • love malmo Somehow place related: 1 • I wait for the bus and feel good Not relevant: 1 • SD Troll: 1 Blank answers: 7

Anxious = 2 answers Place related: 1 • I wonder weather there is a problem with the bus timing today =) Blank answers: 1

Angry = 1 answer Blank answers: 1

Sad = 1 answer Blank answers: 1

53 Day Three = 40 total answers

Excited =2 answers Troll: 2

Happy = 11 answers

Place related: 4 • A lot of people in motion, people are generally very nice • make me happy to see many culture at one place • Sodevern is the shit • the best place ever Mollan • happy for getting to the bus, but it needs more colour on the bus ********* ?

Somehow place related: 3 • because here I wait for the bus home =) • cause I have music Not relevant: 4 • I have a most fantastic relation with a man of my life. Anna • not? • unreadable • God gave me another day.

Calm = 8 answers Place related: 3 • because I can sit and relax till the bus arrives • Sodevern? It’s ok • I’m used to it

Somehow place related: 4 • tired • reminds me of my birth country • on my way home • I don’t know Not relevant: 1 • I am high

Anxious = 13 answers Place related: 7 • I don’t like traffic • here is not clean and the seats are not nice • drunk people • all the smoking

54 • ‘bout shit (on the ground) * • too little benches to sit on • bad information about where the bus stops are placed *

Somehow place related: 1 • tired Not relevant: 3 • have to organize references * • MFF SM-G 2014 • because working day is over and only half of the tasks performed. Blank answers: 2

Angry = 2 answers Not relevant: 1 • I hate everyone

Blank answers: 1

Sad = 4 answer Place related: 1 • no colour. It seems cold and sad

Not relevant: 1 • for little wildlife * Blank answers: 2

55 REFERENCES:

Augè, M. (1992). Non-Places: Introduction to an Anthropology of Supermodernity, Verso Books, London & New York

Darwin, C. (1872). The Expression of the Emotions in Man and Animals, Available at: http://darwin-online.org.uk/content/frameset?pageseq=1&itemID=F1142&viewtype=text de Lange, M. (2013). The smart city you love to hate, The Mobile City, Amsterdam, Netherlands

Galimberti, U. (1987). Il Corpo, Feltrinelli, Roma

Handel, S.(2012). Classification of Emotions, Available at: http://www.theemotionmachine.com/classification-of-emotions;

Helliwell, J. et al (2013). World Happiness Report 2013, United Nations Sustainable Development Solutions Network

Holland, R. (2008). Will the real smart city please stand up?, City: analysis of urban trends, culture, theory, policy, action, 12:3, 303-310

Höök, K.(2013). Affective Computing. In: Soegaard, Mads and Dam, Rikke Friis (eds.). “The Encyclopedia of Hu- man-Computer Interaction, 2nd Ed.”. Aarhus, Denmark: The Interaction Design Foundation. Available at: http://www.interaction-design.org/encyclopedia/affective_computing.html

Isomursu, M. et al (2007). Experimental evaluation of five methods for collecting emotion in field setting with mobile applications, International Journal of Human-Computer Studies

Jacobs, J. (1961). The Death and Life of Great American Cities, Random House, New York

Johnson, S. (2001). Emergence: The Connected Lives of Ants, Brains, Cities, and Software, Scribner, New York, NY

Kevin, C. et al (2012). Citizen Apps to Solve Complex Urban Problems, Journal of Urban Technology, 19:3, 107-136, DOI

Lathia, N. et al (2013). ES2 Library Documentation. Available at: https://docs.google.com/document/d/1TqThJULb-4e6TGb1gdkAaPCfyuXStjJpbnt7a0OZ9OE/edit

Lathia, N. et al (2013). Smartphones for large-scale behavior change interventions, University of Cambridge, Pervasive Computing, Cambridge, United Kingdom

Lowen, A. (1975). Bioenergetics, Pengion Compass, USA

Löwgren, J. (2007). Interaction design,research practices and design research on the digital materials, Sara Ilstedt Hjelm, Raster Förlag, Stockholm

McDuff, D. et al (2012). AffectAura: An Intelligent System for Emotional Memory, Austin, Texas, USA de Lange, M. (2013). The smart city you love to hate, Utrecht University: Media and Culture Studies, The Mobile City, Amsterdam, Netherlands

Nold, C. et al (2009). Emotional Cartography - Technologies of the Self, Available at: http://emotionalcartography.net/EmotionalCartography.pdf

Nummenmaa, L. et al (2013). Bodily maps of emotions, Department of Biomedical Engineering and Computational Science and bBrain Research Unit, O. V. Lounasmaa Laboratory, School of Science, Aalto University, FI-00076, Espoo, Finland;

Paulos, E. and Goodman, E. (2004). The Familiar Stranger: Anxiety, Comfort, and Play in Public Places, ACM, New

56 York, NY, USA. Available at: http://repository.cmu.edu/cgi/viewcontent.cgi?article=1213&context=hcii

Paulos, E. et al (2008). Citizen Science: Enabling Participatory Urbanism inside Handbook of Research on Urban Informatics: The Practice and Promise of the Real-Time City. Hershey, PA: Information Science Reference, IGI Global. Edited by Marcus Foth, IGI Global

Picard, R. (1997). Affective Computing, MIT Press

Picard, R. et al (n.d.). Affective Computing, Affective Computing Research Group at the MIT Media Lab. Availa- ble at: http://affect.media.mit.edu/

Picard R. (2012). Commentary on: Höök, K. (2013). Affective Computing. In: Soegaard, Mads and Dam, Rikke Friis (eds.). “The Encyclopedia of Human-Computer Interaction, 2nd Ed.”. Aarhus, Denmark: The Interaction Design Foundation. Available at: http://www.interaction-design.org/encyclopedia/affective_computing.html

Picard, R. et al (2011). Mood Meter: counting smiles in the wild, Pittsburg, USA

No author, (2013). Emotion Sense ‘Pocket Therapist’ App Claims To Track Your Mood, Huffington Post UK. Avail- able at: http://www.huffingtonpost.co.uk/2013/05/08/emotion-sense-app_n_3235025.html

Reisnert, A. et al (1989). Malmö Stadsbebyggelse, pp: 90-93, Stadtskontoretstryckery, Malmö

Sayid, R. (2013). Emotion Sense: App to track your mood could turn your mobile into a pocket therapist, Mirror. Available at: http://www.mirror.co.uk/news/technology-science/technology/emotion-sense-app-track-your-1874867

Russell, J. A. et al (1989). Affect Grid: A single-item scale of pleasure and arousal, Journal of Personality and Social Psychology, volume 57, issue 3. Available at: https://www2.bc.edu/~russeljm/publications/JPSP1989.pdf

Schroeter, R. (2012). Engaging new digital locals with interactive urban screens to collaboratively improve the city, Proceedings of the ACM 2012 conference on Computer Supported Cooperative Work Pages 227-236, ACM New York, NY, USA. Available at: http://dl.acm.org/ft_gateway.cfm?id=2145239&ftid=1143621&dwn=1&CFID=465957822&CFTO- KEN=78022464

Ståhl, A. (n.d.). Designing for emotional expressivity, Umeå Institute of Design, Umeå. Available at: http://www.sics.se/~annas/Designing%20for%20%20Eotional%20Expressivity/Designing%20for%20Emotion- al%20Expressivity.pdf

Ståhl, A. et al (2008). Experiencing the Affective Diary, Springer-Verlag London Limited. Available at: http://dl.acm.org/ft_gateway.cfm?id=1536073&ftid=1336588&dwn=1&CFID=465957822&CFTO- KEN=78022464

57 THANK YOU FOR READING!