I am co-writing the journal version of the paper Real-Time Geo-awareness – Sensor Data Integration for Environmental Monitoring in the City, a work led by Bernd Resch. My contribution considers the implications of the integration of environment monitoring system as element of urban life. I plan to articulate it as follows:

a) Tools to support practitioners
I reemphasize that this kind of work is first targeted to the information needs of local and regional governments. It can change the work of practitioners that was previously about predicting and accommodating and now it becomes more observing and improving. Indeed, this new ability to render all kinds of ‘machine readable’ environments not only provide new views on the city, but also provide urban and transportation engineers and planners with indicators to evaluate their interventions. For instance Dan Hill and Duncan Wilson foresee the ability to tune buildings and cities through pre and post occupancy evaluations. They speculate that the future of environmental information will be part of the fabric of buildings (see The New Well-Tempered Environment: Tuning Buildings and Cities). However, this integration opens all sorts of issues regarding sampling, density, standardization, quality control, power control, officiality of data, and update frequency (freshness).

b) An imperfect mirror to reality
A complete picture might be hard to achieve with incomplete environmental data patched together by data mining, filtering and visualization algorithms. In many ways we are limited to classic technical issues related to data resolution and heterogeneity. Even mobile sensors do not yet provide high-density sampling coverage over a wide area, limiting research to sense what is technically possible to sense with economical and social constraints. One set of solutions rely on the calibration of mathematical models with only a few sensors nodes and complementing data sources to create a set of spatial indicators. Another, approach aims at revealing instead of hiding the incompleteness of the data. Visualizing the uncertainty of spatial data is a recurrent theme in cartography and information visualization (see Approaches to Uncertainty Visualization). These uncertainty visualization techniques present data in such a manner that users are made aware of the degree of uncertainty in their data so as to make more informed analyses and decision. It is a strategy to promote the user appropriation of the information with an awareness of its limitations (see Notes on Seams, Seamfulness and Seamlessness).

c) Crowdsourcing
Another way to improve the environment data is to alter the current model whereby civic government would act as sole data-gatherer and decision-maker by empowering everyday citizen to monitor the environment with sensor-enabled mobile devices. Recently providers of geographic and urban data have also learned the value of people-centric sensing to improve their services and from the activities of their customers. For instance the body of knowledge on a city’s road conditions and real-time road traffic network information thrive on the crowdsourcing of geodata the owners of TomTom system and mobile phone operators customers generate. Similarly, the users of Google MyMaps have contributed, without their awareness, to the production the massive database necessary for the development of the location-based version of the application. This people-centric approach to gather data raise legitimate privacy concerns. These issues can be handled with a mix of policy definition, local processing, verification and privacy preserving data mining techniques (see Debates on Privacy-Preserving Statistics and Data Mining). These technical solutions necessitate a richer discussion beyond the academic domain on these observing technologies’ social implications.

d) Linking people to their environment
Similar crowdsourcing strategies have been considered for environmental monitoring with individuals acting as sensor nodes and coming together with other people in order to form sensor networks. Several research project explore a wide rande of novel physical sensors attached to mobile devices empowering everyday non-experts with sensing abilities. For instance, Participatory Urbanism investigates the empowerment of citizens to collect and share air quality data measured with sensor-enabled mobile devices. This ‘citizen science’ approach creates value information for researchers of data generated by people going on their daily life, often based on explicit and participatory sensing actions. By turning mobile phones (SensorPlanet), watches (Montre Verte) or bikes (MetroSense) into sensing devices, the researchers hope that public understandings of science and environmental issues will be improved and can have have access to larger and more detailed data sets. This access to environmental data of the city also become a tool to raise the citizen awareness of the state of the environment. Moreover, with the increasing rendering ability of data processing and visualization solutions, citizen can become the actual producers of these awareness tool (example: In the Air).

e) Linking people to their practices
These data gathering and rendering possibilities also implies that we are at the end of the ephemeral, in some ways we will be able to replay the city. In contrast we are also ahead of conflicts to reveal or hide unwanted evidences, when new data can be used to the detriment of some stakeholder and policy makers. Indeed, the capacity to collect and disseminate reconfigure sensor data influence political networks, focussing on environmental data as products or objects that can be used for future political action. Therefore, the openness, quality, confidence and trust in the data will also be subject of debate (e.g. bias to have people record their observations, who gets to report data and who not). The implication of citizens in measuring, sharing, and discussing our environment might increase agencies’ and decision makers’ understanding of a community’s claims, thereby potentially increasing public trust in the information provided by a real-time geo-awareness approach. In consequence this could connect people to the environment in which they live, and provide them with tools for reflection on the impacts of their practices (for example Scanning Objects in the Wild: Assessing an Object Triggered Information System). This objective of improving the environmental sustainability of a city calls for behavior modification can be induced by intervening in moments of local decision-making and by providing people with new rewards and new motivations for desirable behaviors (see Stanford Pesuasive Computing Lab). These kinds of strategies have been common, for instance, in health and fitness applications. However, when we think about persuasion in the real of environment sustainability, what we might want to persuade people of is the ways in which their interests are aligned with those of others (see Paul Dourish’s Points of Persuasion: Strategic Essentialism and Environmental Sustainability). Therefore, this process of alignment and mobilization, by which one can start to find one’s own interests as being congruent with those of others will be critical in the success of these strategies based on real-time geo-awareness.

Relation to my thesis: Applying the kind of transdisciplinary work I have been discussing since the Real-Time Cities Round Table I setup last year. Here trying to provide an HCI perspective to the cutting edge work lead by research in GIS (Research Studio iSPACE) and Environmental Fluid Dynamics (Rex Britter) in the domain of sensor data integration for urban environmental monitoring.

The MIT SENSEable City Lab has unveiled a new project that visualized and analyzed the mobile phone call activity that characterize the crowd Barack Obama’s Inauguration Day. Led by Andrea Vaccari, this work aims at answering the questions: Who was in Washington, D.C. for President Obama’s Inauguration Day? When did they arrive, where did they go, and how long did they stay? The data analyzed consists of hourly counts of mobile phone calls served in Washington, D.C. and includes the origin of the phones involved in the calls. To ensure the complete privacy of the mobile customers.

Besides the dazzling visualizations of Washington DC and the World developed by Mauro Martino, the early findings of the data analysis confirm and quantify the popular impressions. For instance, examining the relative increase in call activity by state reveals some unexpected results. In absolute terms, the most represented states were, unsurprisingly, the most populous: California, Florida, New York, and Texas. In relative terms, the states with the strongest increase were the southern states of Alabama, Georgia, Kentucky and Tennessee, with calls up to twelve times the normal levels. These are states that played a prominent role in the Civil Rights movement and notably are also so-called red states whose voting population went for the Republican candidate, John McCain. Other states with a ten-fold increase in call activity were Illinois, Barack Obama’s home state, and Michigan, Ohio and Indiana, swing states which went blue, voting for President Obama. Most interestingly, comparing these results with U.S. demographic statistics shows that the percentage of African Americans in each U.S. state is a predominant factor determining increase in call activity and therefore participation in the event, which instead was not necessarily influenced by the state’s proximity to Washington, D.C. or its political leaning.

The City joins the mobile call data with a map of Washington D.C. to produce a stirring visualization. The areas around the Mall and Pennsylvania Avenue, where most inaugural activities took place, are highlighted on the map with 3-D building models colored in yellow. In the center of the screen, the map of Washington, D.C. is overlaid with a 3-D color-coded animated surface of square tiles (1 tile represents an area of 150 x 150 meters). Each tile rises and turns red as call activity increases and likewise drops and turns yellow as activity decreases. On the left, a bar chart breaks down the call activity by showing the normalized contributions of calls from the 50 states and 138 foreign countries grouped by continent. The timeline at the bottom illustrates the overall trend of call activity in the city during the week of the Presidential Inauguration.

Similarly, the analyses on call activity in the days before and after January 20 also reveal that the Inauguration was a multi-day event as mobile phone traffic increased markedly throughout the week. The hotspots of activity were clustered in the Northwest neighborhoods of the city, around Downtown, Adams Morgan and U Street.

Relation to my thesis: While these results confirm the popular impressions, they are yet another example on the potential to quantify and compare the presence of people in different areas of the city and to unveil their dynamic movements through time (where people come from compared to a regular day, the Inauguration was a multi-day event). A next step in this kind of research process would be to better link the findings of the data analysis with the visualizations.

Last week I participated to the workshop “La Ciudad Híbrida” organized in Sevilla by José Luis de Vicente. The first 2 days, we used different theoretical lenses and practical references to explore the many different aspects of the hybrid city. Juan Freire engaged us with proposals for a participatory urbanism with the design of open processes and bottom-up solutions (see De la ciudad híbrida al urbanismo P2P: democracia 2.0, gestión local participativa y crowdsourcing). José Luis de Vicente stepped back and led us through the history influenced by urban theorist, architects and artist that have paved the way to the hybrid city of the 21st century, the dynamic city (see Una historia de la ciudad de software: arquitecturas dinamicas y sistemas digitales urbanos). An argumentation completed by Juan Martín Prada who covered the rich field of location media with a philosophic spin (co-existance men-thinks, men-animals, men-deads, the urbanization of the real-time city (networked city) vs. desurbanization of the real-space; the “Here” being replaced by the permanant “Now”.

Before Juan Martin, I played the role of “data cowboy” with a 90min talk (in Spanish) that looked at the contemporary hybrid city through the lenses of my research work augmented by some offline observations. In this intervention entitled “People as sensors; people as actors” (slides with annotations, video), I look at the integration of ubiquitous technologies (and soft infrastructures) and how they afford us new flexibility in conducting our daily activities with simultaneously providing the means to study our activities in time and space.

The other part of the workshop was dedicated to more hands-on activities in which groups had to define, sketch and prototype a citizen-led system or process that take advantage of open/public urban data. With a majority young architects, the focus was first based on “infrastructure”, “mobility”, “space”, but then rapidly also evolved around social issues and even political touching the Critical Cartography approach and critique. There was a lot to learn from the languages employed by participants with different practices (architects, social scientist, biologist, artist).

Finally, Sevilla provided an excellent context to ponder the hybrid city of the present. In addition to its rich history of mixed cultures, architectures and art, engineering work and south-european clichés, the city offers the vestiges of the techno-utopian Expo ‘92. The crumbling infrastructure at the Cartuja Island is source of fascinating sightseeing with an Ariane V on the loose and a monorail in advanced decomposition among other things. The theme for the Expo was “The Age of Discovery”. mmhmm.

Sevilla, a true hybrid city….

…with its real urban furnitures

Thanks José Luis for the invitation!

Related to my past project on Tracing the Visitor’s Eye and World’s Eyes, David Crandall and colleagues at Cornell have created global and city maps and identify popular snapping sites from photos uploaded on Flickr. They ran statistical analyses to identify the most important clusters of “”what the world is paying attention to”. Next they analysed the text tags added to photographs in those clusters, as well as key visual features from each image, to automatically find the world’s most interesting tourist sites. We find that visual and temporal features improve the ability to estimate the location of a photo, compared to using just textual features (slightly similar to IM2GPS: estimating geographic information from a single image). Their paper, Mapping the World’s Photos, recently presented at the WWW 2009 conference explains the process and results in more details.

Representative images for the top landmark in each of the top 20 European cities. All parts of the figure, including the representative images, textual labels, and even the map itself were produced automatically from our corpus of geo-tagged photos. Image courtesy of David Crandall. More on the project’s web page.
Relation to my thesis: Similar to my work, Crandall et al. I have used the spatial distribution of where people take photos to define a relation between the photos that are taken at popular places. I have used similar representations of spatio-temporal traces to illustrate my techniques and finding. However, my motivations has not been to improve photo management and organization applications (see also Mor Namaan’s work). Similar to Currid and Williams’ work on Mapping the Cultural Buzz, I employed georeferenced photos as a proxy to “qualify” the space (e.g. its attractiveness). But rather than mapping and qualifying a timeless space (Currid and Williams talk about “buzz-worthy” without referring to when), I particularly explore the temporal aspect of the data (see The End of Timelessness). For instance, in my work on the NYC Waterfalls, I studies the evolution of the density of georeferenced photos and the flows of photographers to provide evidences of the evolution of the attractiveness of the space (journal paper still in submission).

The Visualization of photographer movement in Manhattan mapped by Crandall is particularly stiking:

To produce these figures, we plotted the geolocated coordinates of sequences of images taken by the same user, sorted by time, for which consecutive photos were no more than 30 minutes apart. Image courtesy of David Crandall

A couple of weeks ago at the Association of American Geographers meeting in Las Vegas, Elizabeth Currid (USC) and Sarah Williams (Columbia University) presented their paper The Geography of Buzz: Art, Culture and the Social Milieu (full aper). As described in Mapping the Cultural Buzz: How Cool Is That?, their work takes snapshots from more than 6,000 events from the data mining of 300,000 photos from Getty Images to categorize them according to event type, controlled for overly celebrity-driven occasions and geo-tagged at the street level. They exploited these photos as proxy for ‘buzz-worthy’ social contexts. As a result, they were able to quantify and understand, visually and spatially, how this creative cultural scene really worked. Based on these evidences, they can argue that those not conventionally involved in city development (paparazzi, marketers, media) have unintentionally played a significant role in the establishment of buzz and desirability hubs within a city.

Density of “fashion” events in New York
Relation to my thesis: After Anthony Townsend’s Augmenting Public Space and Authoring Public Art: The Role of Locative Media, the geography of buzz provides another instantiation of the value of digital footprints to describe the city. However, this use of photos as proxy of human time-space activity raises issues in not to confound behaviors with endorsement. It is one of the limitation of my work on the NYC Waterfalls. Nevertheless, Currid and Williams work certainly strengthen the value of my work on Tracing the Visitor’s Eye, with quotes that could come from straight out of this blog:

“We’re going to see more research that’s using these types of finer-grained data sets, what I call data shadows, the traces that we leave behind as we go through the city,” she said. “They’re going to be important in uncovering what makes cities so dynamic.”

They also provide another argument for the discussion of the contributions of my research work in the context of human-space relationship:

“People talk about the end of place and how everything is really digital. In fact, buzz is created in places, and this data tells us how this happens.”

Finally, an excellent reference in their bibliography that coins data shadows:

Zook, Matthew, Martin Dodge, Yuko Aoyama, Anthony Townsend, “New Digital Geographies: Information, Communication, and Place”, In Geography and Technology, Kluwer Academic Publishers, 2004

Data mining artist Andrea Vaccari gave a thorough interview to ZDNet on his current investigation at MIT SENSEable City Lab also providing context of the lab’s current endeavors. He starts of by argumenting on the necessity to produce quality visualizations responding to the skeptics that call it “info-porn“. I found myself defending the necessity of the urban demos produced at SENSEable to attract attention, stimulate the dialogue and stretch the imagination. Yet, this is only the tip of the iceberg of the work produced in Cambridge that strongly relies on the experience and data generated during the demos to produce scientific content. The opacity of the research review process and publication venues, does not often allow to get a glimpse of the research outcomes. So Andrea goes at it and gets specific on the type of contributions we can expect form the analysis of city-scale data. He takes example of the project in the context of the New York City Waterfalls we collaborated on to describe how the analysis of urban dynamics can become evidences of the evolution of the attractiveness of the most popular areas of a city. Other of his works aim at identifying unexpected events and assist public authorities and first responders.

Andrea recently presented the lab’s work at Etech (slides) where is seems that the buzz was on the fusion of sensor data with social data traffic that “to improve life“. Beside this honorable goal, I am dubious that this is purely reachable with data mining, machine learning to networks analysis and statistics. But maybe the current conclusion of my dissertation plays a bad influence on me.

Relation to my thesis: Andrea describes very well the lab its approach and projects I had the chance to share back in my SENSEable days. The open question that he leaves at the end of his interview are up my alley. Currently writing the conclusion of my dissertation, I take advantage to draw the implications of my work and discuss “human side of urban data” and what it means in context of urbanism; discarding a pure data-driven urbanism and sketching something I would call human-based urbanism, an approach mixing quantitative and qualitative data analysis.

Over the holidays, David Lu produced some new visualizations for our World’s Eyes project featured in the Tourism: Spaces of Fiction exhibit at the Disseny Hub Barcelona. It includes for instance a new look on Barcelona through the eyes and of Britons. David kept track of the messy, organic trajectory delineated by mistakes and occasional success that led to these results. Some of his sketches could have been easily exhibited as well. These sketches follow a messy, organic trajectory delineated by mistakes and occasional success, mainly due to me being in Europe and him on the East Coast (in a classic senseable fashion). Data mining maestro Andrea Vaccari also contributed to the development of this work.

Relation to my thesis: Follow-up of the first iteration. This work is of great use to communicate my work outside of the academia.

Last week I was in Paris for the closing event of the Villes 2.0 program that gathered experts and practitioners on the “digitization” of the city and its novel open and participatory practices. Adam Greenfield sat the stage for the event with his opus “The Long Here, the Big Now, and other tales of the networked city” (video from Visualizar) that pleased me with some new considerations on the implication of this new urban reality on the practices of architects and urban planners. In other words, how can we develop “villes” (cities) 2.0 for us, humans 1.0? My contribution (slides) took place around this notion of “villes 2.0″, a city with new urban actors named Google, wireless networks, but also a city fostering citizens contributions and bottom-up information (see La Ville 2.0 existe déjà, on l’a rencontrée). I described how through our implicit and explici interactions with these new urban actors, we generate data that, when aggregated over space and time, bring novel perspectives (e.g. in real-time, on dynamics of the city. Beyond the possibility to visualize these dynamics, often useful to open or support discussion in the urban area, I suggested with examples of my recent research works that these data can become sources of urban indicators, valuable to develop an urbanism based on evidences (in the sense of signs and indications), particularly promising to perform new types of post-occupency evaluations. I described these notions more extensivily at Visualizar in Madrid, last November (video of From Sentient to Responsive Cities). Other sessions, covered the current and new practices to innovate in the urban realm (see Aborder les questions urbaines par l’innovation) and how this innovation can take place with open access to the edition of the city (e.g. with open APIs), a shared control between private companies and citizens leading with a promotion of the appropriation instead of a stubborn approach of innovation (see La ville 2.0, plate-forme d’innovation ouverte).

Villes 2.0 is a unique program, mixing think-tank and do-tank approaches, with outcomes in the forms of 3 books: “La ville 2.0, complexe … et familière“, “Manifeste pour une mobilité libre et durable” and “La ville 2.0, plate-forme d’innovation ouverte” supported by the exploration of the theories through several urban demos such as a Citywall, a real-time control system, and CityPulse as well as prototypes (see the flickrset).

Within the Ville 2.0, the contemporary “Rue” 1.0 in Paris, governed by anti-terror measures and sustainable development. Here a sensor of some sort next to the vigipirate-promoted transparent trash bags sporting the slogan “viligance” and “cleanliness”.

Relation to my thesis: Testing my research outside of the academic arena. Interestingly, several urban/human geographers approached me after the session, interested on the potentials of my work to support their traditional approaches.

Many thanks to Thierry Marcou, Fabien Eychenne and Daniel Kaplan for the invitation!

I will present the work and concepts developed with David Lu for the World’s Eyes visualizations at the GeoViz 3-days workshop on the “Contribution of Geovisualization to the concept of the Digital City“. The long abstract of “Exploration and analysis of digital footprints through interactive visualization” goes as follows:

In recent years, the large deployment of pervasive technologies in cities has led to a massive increase in the volume of records of where people have been and when they were there. With the ubiquity of mobile and wireless devices, the logs produced become anchored to the physical space with a geographical reference. The emergence of these digital footprints is expected to have significant impact on social sciences and urban studies with this novel access to a massive quantity of quantitative data. Indeed, pioneer works have demonstrated that their accumulation can uncover social behaviros and the spatio‐temporal interactions between people and places. In practice, the visualization and analysis of these urban data could deliver value added information in the different domains touching city management such as urbanism, environmentalism or tourism. For instance the visualization of the density of cellular network activity in a city can help traffic engineers improve public transport provisions in populated areas of a city at peak hours.

However, the dynamic nature of digital footprints and their fluctuating spatial and temporal resolution poses new challenges. Indeed these urban data come from cheap sources and infrastructures not deployed to sense the activities of a city. Therefore it is nececissary to assess the quality of the data through calibration with data collected with more traditional means (e.g. surveys, census data). These shortcoming greatly impair the use of these kind of information for city management and decision making in the near future. However, their visualization are already great tools to communicate with anectodes and evidences uncovered from exploration of this new types of data.

Our specific solution maps digital footprints visitors of a city leave behind them (Figure 1). The analysis of this traces allow to get a perspective on the density and flows of visitors and as a result compare the attractiveness of points of interests (Figure 2). Coordinated visualizations are integrated within individual probe interfaces, which depict the local data in user‐defined regions‐of‐interest. These probes empower users with the ability to observe, coordinate, and compare data over space and time and across multiple local regions. We have been developing this tool not only to interact with these digital footprings, but also to communicate explorations and findings. The visualization and the interaction also serve multiple different actors of the city such as the researchers, to interests groups, practitioners and politicians, to agree and disagree on how to interpret data and contribute contextual knowledge that deepens understanding.

Figure 1. Interactive tool that visualize the density and movements of locals (yellow) and visitors (in pink) from the georeferenced photos taken in Spain and shared shared online.

Figure 2. Interactive tool that visualize the density and movements of photographers visiting Barcelona and sharing their photos online. Yellow glows shows the most attractive points of interest.

Following the MIT SENSEable City Lab’s footsteps (RealTime Rome), Orange Labs and Faber Novel developed Urban Mobs, a tool showcasing their “popular emotion cartography” through the analyzes and visualization of citywide cellular network traffic activity. The featured chronotopes include uncovering “visitors” aggregated digital footprints in cellular network activity generated by foreign phones, linking mobile activities with public transport in the cellular network activity in the Paris subway, and revealing aggregated digital traces in the map of handovers generated by the movements of mobile phones during the national music day.

Visualization of mobile phone calls in Barcelona during the Euro 2008 final

Relation to my thesis: Urban Mobs proves that Real-time Rome work is a pervasive solution that can be replicated to multiple cities and countries to mirror an aspect a city dynamics. Now the goal is to go beyond beautiful visualizations of dynamic urban data and extract their value added information as well as understand their applications and implications.