How do the citizens of Riga actually feel about their city? What do they think about urban topics that concern the development, public transport, environment, safety and culture of their city? As part of the Participatory City, initiated by the Connecting Cities Network, Staro Riga Festival invited us to show our work.
The project called Smart Citizen Sentiment Dashboard (SCSD) is an interactive and participatory installation that lets citizens engage with and comment on urgent urban challenges concerning the city of Riga. The SCSD project aims to translate citizen feedback into a visual language, which is displayed on a large LED display. The media facade and its surrounding turn into a stage for social encounter where citizens meet and urban challenges can be discussed.
Staro Riga Festival
During the Staro Riga Festival the installation was set up on a busy pedestrian crossing at the intersection of Marijas Iela and Satekles Iela, which is close to the Riga main train station. The visualization was displayed on a large mobile screen facing towards Satekles Iela, and the sentiment dashboard was set up in front. The installation ran daily in between 6pm and 11pm starting Friday, 14 November 2014 and finished at the National Independence Day on Tuesday, 18 November. During the five days of deployment, in which the installation was running for five hours each evening, the interactive system tracked approximately 1600 interactions.
The motivation was to develop, design and deploy a situated and tangible system that mediates collaborative interactions in public spaces whilst focusing on accessibility and affordance. In other words, the interface should be understandable and easy to use for people. The employed technology makes use of existing ‘Radio frequency identification’ (RFID) as known from smart card technology such as the e-talon in Riga. We build on the widely spread use of these unique ID tags for payless travel purposes, as a large proportion of citizens of Riga carries an e-talon in their pocket. Consequently the use of these cards is a recurring embodied interaction in the smart city. At the same time every interaction is uniquely identifiable and therefore traceable. Our aim was to allow people to use their ID tags beyond technical purposes and express their mood and opinion about specific issues in the technology mediated urban realm. Hence the Smart Citizen Sentiment Dashboard (SCSD) enables participants to express their mood about urgent urban challenges in the city of Riga. The challenges were defined as follows: 1) ATTĪSTĪBA (development), 2) TRANSPORTS (transport), 3) VIDE (environment), 4) DROŠĪBA (safety) and 5) KULTŪRA (culture). By switching a knob on the device participants are able to choose one of the aforementioned categories. By swiping their RFID token (i.e. e-talon) across one of the two emoticons (happy or sad) their mood was transmitted onto the LED display. The SCSD affords three folded interactions: 1) switching: 5 categories can be selected through a rotary switch; 2) swiping: after choosing the category, the electronic ID card needs to be swiped over one of the three mood states (happy, indifferent, sad); 3) pushing: finally a simple push-button (Red Button) allows users to view the overall feedback of all collected moods.
We chose a visualization technique that combines the “seriousness” of the topic with the more accessible style of popular info-graphics. The visualization consists of an abstract sunburst representation, of which each burst corresponds to the sentiment of an individual participant towards the currently selected urban challenge. Each urban challenge is encoded by a different color and an icon representation. Upon switching the rotary knob, the sunburst visualization corresponding to the specific urban challenge, and colored accordingly appears on the facade.
The sentiment ‘value’ for each participant (happy, un-happy) is graphically encoded through the length of the corresponding burst: the longest burst represents a positive sentiment towards the urban challenge at hand, while the shortest corresponds to a negative statement. Our choice for this circular visualization technique was also motivated by its scalability, which allows for an arbitrary number of people to participate and be visually represented. We considered this flexibility a desirable feature in the context of urban environments, often characterized by highly variable and open-ended, and unpredictable flux of people and interactions.
The integration of dynamic visual cues can make visualizations richer, more vivid and therefore easier to understand. Accordingly, our visualization shows a dynamically animated circle over the sunbursts in order to convey the average participants’ sentiment for the given urban challenge. Each new burst from a participant visually appears with a smooth animation and bouncing effect, to highlight the recording of fresh data. A new entry is displayed in a white color to unambiguously distinct it from the rest of the graphical representation. Shortly after it is smoothly taken over by the color of its respective urban challenge.
The Red Button
In order to provide citizens with an overview of previously submitted sentiments, and a more interactive approach to exploring the installation, we integrated a ‘Red Button’ at the bottom of the interface. When pushing this button, a dynamic visualization of the average feedback for all available urban challenges is represented on the facade. As mentioned above, each urban challenge is represented by its corresponding color, and occupies a different part of the circular shape proportionally to the relative participation rate of the according challenge. We aimed to create a simple, playful, yet meaningful approach to enable citizens and participants alike to make a deeper sense of the installation, and the underlying participation results: people can gain insight about which urban challenge is most attractive to vote for, and what is the average sentiment about it of fellow citizens. This, beyond being an overview, the heart visualization symbolizes the overall ‘sentiment’ of the city towards its urban challenges.
The installation was running for five days, in total for 25 hours. During this time the database logged about 1600 interactions. These interactions cannot automatically be counted as single votes by individuals as many participants were exploring the installation in a playful manner through swiping their e-talon more often. In a next step with an elaborate data cleaning process this behaviour can be removed from the data set. However, we can already identify a clear tendency that shows that participants overall voted positive, as 75% of the interactions were happy. If we look more closely into the different urban challenges we can reveal more telling insights of how the citizens of Riga think about their city. ATTĪSTĪBA (development) was the category, which attracted with 356 interactions the least attention of participants. Still 76% (270 counts) of the recorded data was positive. As development maybe interpreted in many ways participants could have felt irritated and therefore may have decided not to vote in this category. TRANSPORTS (transport) and DROŠĪBA (safety) were the most controversial topics, which is by the way in line with other cities where the installation was running before. Transport attracted 379 interactions and safety 434 interactions of which both topics were 65% happy and 35% un-happy. As there were more interactions tracked for the category of safety one may argue that this is due to the higher importance of this topic to participants. VIDE (environment) attracted 447 interactions of which 78% were positive and 22% negative. KULTŪRA (culture) has been the topic most participants wanted to express their sentiments about. The system logged 518 interactions of which 441 (85%) were positive and only 77 (15%) negative. Culture is therefore not only the most frequently used urban topic but also the topic participants are the happiest about. One may argue that this result might be due to the fact that participants were influenced by the many cultural events that were going on at this time during the Staro Riga Festival.
Overall it is to say that the logged interactions by the system were on average 75% positive as well as none of the categories reached a majority of un-happy interactions. This is a result we did not find in any previous deployment and therefore lets us assume that Riga is a city where people seem to be quite content with the addressed urban challenges or tend to see things in a more positive way.
Our observations revealed three different participation patterns:
The serious behaviour
A participant submits exactly one sentiment for each of the explored categories. This pattern would reflect how we expected the interaction mechanism to work – i.e. a person would explore the categories by rotating the knob and would submit one sentiment for a specific preference.
The repetitive behaviour
This was the most frequently observed participation pattern. The participant has submitted the same sentiment (same preference for a certain category) several times within the considered time range. The occurrence of this pattern can be explained with our frequent observation of participants holding their card over the RFID reader (for a certain preference) for several seconds. Thus the system registers several submissions (although our system had restricted votes not to be registered within 5 seconds after each given participation). This behaviour might be due to a usability flaw of our installation – the participating person did not realize the effect of her participation in the visualization, hence tried several times. Another explanation might be the manifestation of a particular sentiment towards one urban challenge: by holding the card over the reader, the user might have wanted to reassure herself that her opinion would be registered by the system.
The playful behaviour
There were many occurrences of this behaviour during the deployment of the installation. The participant has submitted several different preferences for the same category within the considered period of time. This might indicate that s/he did not really want to express an opinion, but rather explored how the installation and the visualization work. While we cannot account for representative polling results, the findings indicate the installation fulfilled its intentions as an urban feedback platform, where people engage meaningfully with locally relevant topics. In the future it would be exciting to deploy several citizen sentiment dashboards permanently across the city as well as working closer with city authorities and local communities. This might open a fruitful dialog in between citizens and stakeholders of Riga.