Posted: March 26th, 2007 | 2 Comments »
The main problem for urban planers, traffic engineers, transport authorities to perform urban modeling is the lack of real mobility data. While car traffic is pretty easy to estimate with sensors (cameras) deployed on the main city crossroads, when there is not “real” data, the only options are to perform in-sitiu surveys or rely on rough estimations (e.g. based on the police estimations). But the resulting picture is often inadequate, expensive—or both. Experiments to automatize the process remains very scarce. For instance, Eamonn O’Neill and Vassilis Kostakos (University of Bath, UK, as part of the Cityware project) presented at Ubicomp last year, a study on the deployment of a Bluetooth scanning system at every key points in Bath (with only 7.5% of the pedestrians carrying a Bluetooth enabled mobile phone). On the other hand, the transportation research has been (painfully) processing logs of vehicles location data generated by GPS embedded in cars. Finally, the market research industry has been trying to guess how many eyeballs pass a billboard by providing panelists with GPS units.
In the recent years, the large deployment of GSM phones generated a massive increase in the volume of spatio-temporal data that can be used to infer individuals or groups spatio-temporal behaviors. The MIT SENSEable City Lab (recently featured in the “E”‘s Technology Quarterly) has pioneered in the visualization of the movement dynamics of cellphone users and has rendered a real-time “pulse of a city” (video). The results take the form of luminous maps adorned with moving and colour-coded arrows, dots and patches of light that indicate the speed and population density of people in the city in question. Other example, IntelliOne developed TrafficAid that identifies congestion in real time by using cell phone location provided by GSM operators. The measurement system processes thousands of device locations per second, “snaps” each device to the road network, and monitors movement for a short period of time. Similarly, Timothy Sohn showed (also at ubicomp06) his mobility mode recognition experiment using everyday GSM traces. He used the logs generated by the everyday lives of three data collectors over a period of one month, yielding an overall average accuracy of 85%, and a daily step count number that reasonably approximates the numbers determined by several commercial pedometers.
MIT’s Real-time Rome: Crowd movement in Rome during the celebration of the World Cup victory.
I perceive 3 main domains of applicability of the processing and visualization of these massively collected personal logs, traces (i.e. history of previous locations)
- Provide urban planners, transport authorities and traffic engineers with data to refine their models of citizens spatio-temporal behaviors. For example, fine-grained “origin-destination” (OD) statistics allow to build more accurate traffic simulation.
- Bring new perspective for decision making and policies building. People-movement maps “will be invaluable” in planning housing and transport.
- Raise awareness and affect the discussion making of individuals or of a crowd. The above-mentioned TrafficAid system is one example. A step beyond, Intel Research Berkeley suggests the idea of Participatory Urbanism, AIR, or BioMapping . This feedback mechanism (urban awareness) is one fascinating aspect of such massive spatio-temporal data collection (this is probably what I would discuss if I get the chance to go to the upFing’07 Villes 2.0 workshop). In fact, Nicolas and I have been discussing quite a bit the use of asynchronous location awareness (traces) tools as a way to highlight social spaces inferred from the traces left in the environment .
Inferring spatio-temporal activities raise several issues that are at the core of my research. First, geographic information systems have limited ability to handle the temporal dimension. Then, the many sources of error and imprecision in the sensed data impact the quality and timeliness of the location data. The visualization might take into consideration the uncertainty in the delivered information (as raised by MacEachren et al in Visualizing Geospatial Information Uncertainty: What We Know and What We Need to Know). Finally, histories of location information mainly render a quantitative understanding of the city.
A visual log of physical presence
I have been with Radio-Frequency (i.e. WiFi and GSM) based positioning and mobility inference for the past 3+ years. Now, I explore a different approach to the collection and analysis of spatio-temporal activities. Based on a first experiment I called (inspired by Scott Smith) Tracing the Vistor’s Eye, I explore the potential of using user-generated geotagged content to infer activities. I have been relying on Flickr geotagged photo to define and confirm patterns of how tourists navigate the urban space (e.g. the elephant path along the Rambla, people do not walk to Gaudi’s Sagrada Familia). I believe that the richness of the collection and analysis of such data is that it carries a higher meaning that x;y;timesptamp location data. Taking, uploading, tagging and geotagging a picture can be interpreted as an act of communication rather than a pure implicit log of physical presence (Inspired by Nicolas’ work on automatic and manual location disclosure and The Error of our Ways: The experience of Self- Reported Position in a Location-Based Game. Therefore, my research agenda for this experiment is to:
- Build a framework that infers spatio-temporal activities in urban spaces based on user-generated content. Use tourists exploring cities as context
- Enrich quantitative spatio-temporal data with qualitative information embedded in images
- Integrate the notions of location information quality and timeliness in the data
- Analyze how Flickr users use the accuracy feature to geotag their images (where, what, when, history of use, overall usage over time). This is to provide a study on the quality of explicitly disclosed location information. In other words, what level of location information quality and timeliness must be delivered in order to be useful and relevant?
- Experiment with ways to visualize this type of spatio-temporal data with its uncertainty. Why not use the InfoScope. Here I aim to explore the parameters that influence successful uncertainty visualization.
- Evaluate the People-movement maps/visualizations. Maybe use the people of Barcelona Ecologia as expert users.
- (would be nice) Evaluate the completion of the “feedback control system”.
I have the core of the framework ready. It processed and generated KML files from test data of Barcelona, San Francisco, and Boston. I plan to collect data from the World’s Most Photographed Cities. As a next step, I will add the sense of origin and destination to visual representation of the traces. Moreover, I will infer the origin of the photographers as of being a tourist or a citizen. Finally, the photos will belong to several temporal categories (e.g. morning/afternoon/night, week day/weekend, special events) and location accuracy categories (e.g. street, neighborhood, city).
Early experiments with Flickr images taken in San Francisco between March 11 and March 25, 2007). The complete set of screenshots.
Relation to my thesis: I became acquainted of the multiple issues in collecting data collection during my discussions with the urban ecology agency “Barcelona Ecologia”, my readings in transportation research and the design and deployment of GSM-based personal mobility detection systems. This urban computing experiment is the first of the three experiment I plan to investigate my main research questions. It aims at providing clues to my sub-research questions 1) what level of location information quality and timeliness must be delivered in order to be useful and relevant? and 2) what parameters influence successful uncertainty visualization?
Posted: March 16th, 2007 | 2 Comments »
Ykä Huhtala and Jussi Kaasinen of Nokia Research recently offered for free download their Nokia Sports Tracker for S60 3rd edition phones. Nokia Sports Tracker uses a Bluetooth GPS device or an integrated GPSis a GPS based activity tracker that runs on S60 smartphones. Information such as speed, distance and time are automatically stored to your training diary.
I programmed a similar application in S60 Python a while ago. However I rapidly got annoyed by several things:
- the need to find an place to get good visibility to satellites for the GPS to warm-up and start running
- while running, the phone would sometimes lose the Bluetooth connection to the GPS without being able to automatically reconnect (possibly a S60 Python issue)
- I would often run in forests where I would get bad or no GPS signal impacting the measured distance.
- Wearing a mobile phone + GPS unit while running is a pain (e.g. straps that get loose).
- I ended up taking phone calls or reading/writing SMSs…
Relation to my thesis: There is a real interest in logging personal mobility and activity (i.e. traces) for health (pedometers, BioMapping, Nokia Sports Tracks, Nike+iPod) or social (Jaiku, Marc Hottinger’s dataSpaces) purposes. Such applications can provide valuable self-awareness and replay tools (such as in persuasive or healthcare systems). However, beyond the privacy and ethic issues, the way to integrate them in our daily lives, collect data and way to deliver them in a meaningful way is still at its infancy.
Courtesy of Bio Mapping
Posted: March 5th, 2007 | No Comments »
My paper for the Pervasive 2007 Doctoral Colloquium has been accepted. It is an extended version of the CHI’07 submission.
Bridging the Social-Technical Gap in Location-Aware Computing
Abstract. Building ubiquitous applications that exploit location requires integrating underlying infrastructure for linking sensors with high-level representation of the measured space to support human activities. However, the real world constraints limit the efficiency of location technologies. The inherent spatial uncertainty embedded in mobile and location systems constantly challenges the coexistence of digital and physical spaces. Consequently, the technical mechanisms fail to match the highly flexible, nuanced, and contextual human spatial activities. These discrepancies generate a social-technical gap between what should be socially supported and what can be technically achieved. My research aims at exploring, and hopefully reducing this gap in the context of location-aware computing.
Relation to my thesis: I am pretty glad that people like Shwetak Patel (Supporting Location and Proximity-Based Studies in Natural Settings) and Leif Oppermann (Extending Authoring Tools for Location-Aware Applications with an Infrastructure Visualisation Layer) will also present their work. Hopefully Leif will also talk about his location based mobile phone game Love City. Pervasive 2007 will take place on May 13-16 in Toronto, Ontario, Canada.
Posted: February 27th, 2007 | 3 Comments »
The Amgen Tour of California took place this week. During this year’s edition, 7 riders have been tagged with CSC’s OmniLocation devices to send a constant stream of GPS data to the T-Mobile GSM network (over general packet radio service signals, HTTP streaming and SMS messaging). In consequence, The race’s web spectators had access to almost-real-time location information displayed on Yahoo! maps for the riders with a short delay offset (i.e. about 10 seconds).
Screenshots from the live webcast featuring the rider tracking system.
Relation to my thesis: Using sensors to track performances in sport is nothing new. Yet, this is an example of deployment in a semi-uncontrolled settings (the GSM network provider was a sponsor). Moreover, I am wondering about the web spectator’s experience in watching dots moving (without smooth animation but with icons of riders jumping from one location to another) on a map. Finally, the web application does not seem to take into account the delay offset from the acquisition of the location to the display on the screen. I suspect that a sharp synchronization of the geodata had to be performed prior to the visualization.
Posted: February 16th, 2007 | No Comments »
At the end of his talk on Big Games: Large-Scale, Multi-Player, Real-World Games at Where 2.0 in 2005 (audio), Kevin Slavin came up some “chili computing” (coined by Frédéric Kaplan at Lift) concepts around the experience of location in real-world games. First he mentioned that location is more than GIS data. Location has a wider context and different meanings, it is about being indoors, outdoors, about hearing the busy sound of a street or not. Then he suggest that in fact disinformation and “dislocation” might be more engaging to people as we might become interested into getting lost and making things up (aiming at misrepresentation rather than accuracy). This idea of inventing the real reminds me a discussion with Julian on his current game ideas around the concept of familiar strangers and his inspiring talk When 1st Life Meets 2nd Life.
Relation to my thesis: While my specific work aims at the opposite of using technologies to get lost and tell use lies, I find the approach very relevant. The approach actually reveals that “constraints aren’t what break games. Constraints are what makes them work”. It can be an inspiration for designer of location-aware systems who must deal with quality and accuracy issues could. Instead of hiding the problems or aiming at sharp accuracy, an application might be more engaging when it is less “serious”. This relates to my last week’s discussionwith Frédéric on the nabaztag who tells the weather. Because, the nabaztag is not serious by itself, it is okay if he makes mistakes on forecasting the weather. Actually, its mistakes might lead into a discussion and a collaborative correction of the error.
Posted: February 11th, 2007 | 13 Comments »
On Friday, I gave a talk on Embracing the Real World’s Messiness (slides, video) at the Lift Conference open stage session. Some people in the audience took notes and pictures, including Tom Hume (Future Platforms), Hubert Guillaud (Fing, en français) and Mark Meagher (EPFL).
Relation to my thesis: While I did not present the core of my research, the topic can serve as introduction to my thesis.
I felt a research or engineering talk would not have completely fit to the audience. Therefore, I rather preferred taking the role of the observer of the current integration of sensor technologies in our everyday life in order to question the seamlessness and calmness visions in ubiquitous computing. Even though I feel I only communicated 1/3 of my thoughts, the feedback I received suggest that I delivered my message. In his wrap-up talk (video), Daniel Kaplan shared my observations in highlighting that “we’re using technology to create disorder – you can call it innovation, I call it disorder”. I have been enjoying reading Daniel since he coined the term “Désordinateurs” in reaction from the “Utopie du lisse“.
This talk was based on a few previous blog post, including:
Posted: February 6th, 2007 | No Comments »
Rogers, Y. Moving on from weiser’s vision of calm computing: Engaging ubicomp experiences. In Ubicomp (2006), pp. 404–421.
This paper urges for an alternative agenda in ubicomp research that shifts from Weiser’s calm vision to engaging people (i.e. proactive computing, persuasive computing, engaged living). Yvonne Rogers acknowledges that research in context awareness, ambient intelligence and monitoring/tracking have been somehow fruitful. However they have yet failed to reach Weiser’s world. Indeed, there is an enormous gap between the dream of conformable, informed and effortless living and the accomplishment of UbiComp research. In fact, the fundamental stumbling block has been harnessing the huge variability in what people do, their motives for doing it, when they do it and how they do it. While it has been possible to develop a range of simple ubicomp systems that can offer relevant information at opportune moment, it is proving to be much more difficult to build truly smart systems that can understand or accurately model people’s behaviors, moods and intentions. This makes it difficult, if not impossible, to try to implement context in any practical sense and from which to make sensible predictions about what someone is feeling, wanting or needing at a given moment. Therefore, ubicomp technologies should be designed not to do things for people but to engage them more actively in what they currently do. Rather than calm living it promotes engaged living, where technology is designed to enable people to do what they want, need or never even considered before by acting in and upon the environment. Examples include extending and supporting personal, cognitive and social processes such as habit-changing, problem solving, creating, analyzing, learning or performing a skill.
The author mentions the problems of calm computing in the most prominent ubicomp research themes (i.e. context-aware computing, ambient/ubiquitous intelligence and recording/tracking and monitoring).
Context-awareness
Key questions in context-aware computing concern what to sense, what form and what kind of information to represent to augment ongoing activities. Many of the sensor technologies, however, have been beset with detection and precision limitations, sometimes resulting in unreliable and inaccurate data. While newer technological developments may enable more accurate data to be detected and collected it. However, people often behave in unpredictable and subtle ways in their day-to-day contexts. Therefore, it is likely that context-aware systems will only ever be successful in highly constrained settings.
Ambient and Ubiquitous Intelligence
While there have been significant advances in computer vision, speech recognition and gesture-based detection, the reality of multimodal interfaces – that can predict and deliver with accuracy and sensitivity what is assumed people want or need – is a long way off. In consequence, when a ubiquitous computing system gets it wrong – which is likely to be considerably more frequent – it is likely to be more frustrating and we are likely to be less forgiving.
Recording, Tracking and Monitoring
Much of the discussion about the human aspects in ubicomp has been primarily about the trade-offs between security and privacy, convenience and privacy, and informedness and privacy. This focus has often been at the expense of other human concerns receiving less airing, such as how recording, tracking and re-representing movements and other information can be used to facilitate social and cognitive processes.
Yvonne mentions 2 goals of my research, one being to use ubicomp technologies in the wild, the other to evaluate how to present data and information:
In addition, more studies are needed of UbiComp technologies being used in situ or the wild – to help illuminate how people can construct, appropriate and use them. With respect to interaction design issues, we need to consider how to represent and present data and information that will enable people to more extensively compute, analyze, integrate, inquire and make decisions; how to design appropriate kinds of interfaces and interaction styles for combinations of devices, displays and tools; and how to provide transparent systems that people can understand sufficiently to know how to control and interact with them.
Currently, the more engaging approach is beginning to happen through the areas of playful and learning practices, scientific practices and persuasive practices.
As already mentioned in Comparing AI’s Failures with Ubicomp’s Visions, Yvonne Rogers concludes on “strong” and “weak” UbiComp.
Just as ‘strong’ AI failed to achieve its goals – where it was assumed that “the computer is not merely a tool in the study of the mind; rather, the appropriately programmed computer really is a mind”, it appears that ‘strong’ UbiComp is suffering from the same fate. And just as ‘weak’ AI2 revived AI’s fortunes, so, too, can ‘weak’ UbiComp bring success to the field.
Relation to my thesis: I would argue that current “strong” UbiComp problems not only lays on modelling people and their activities, but also in the integration ubicomp systems in the real-world (e.g. co-existence of systems, real-world constraints). I enjoy the difference between what is “relevant” and what is “smart”, as I find the word smart or intelligent are widely (over)misused. Finally, the agenda proposed in this article, goes in the direction of my research: in sitiu (out of the lab) studies, investigate the playful approach of ubicomp and how to present relevant information rather than seeking the seamlessness utopia.
Posted: February 5th, 2007 | No Comments »
The CommonCensus Map Project redraws the map of the United States based on a survey questions, to reveal the boundaries people themselves feel (i.e. sphere of influence), as opposed to the official state and county boundaries.
The national maps shows the response to the question “On the level of North America as a whole, what major city do you feel has the most cultural and economic influence on your area overall?”
Regional maps show the response to the question “Please choose the name of the local community that you feel is the natural cultural and economic center within your local area.”
Local maps show the response to the question “What do you consider to be your local community?”
Relation to my thesis: People do not always follow the official boundaries to refer to areas. A local neighborood might reveal very fuzzy and fluctuating edges depending on a context. This is what Ian White highlights in User-centered approach on geodata by saying “In practice, a neighbor is defined with average centroid based on population density and then a radial curve is drawn. This barely represents reality in many cases and in the context of use many time useless”.
Due to the low amount of data, the areas of the map are still highly inaccurate and subject to change. It is an example of bottom-up generated information uncertainty.
Posted: January 29th, 2007 | No Comments »
Extracted from an inspiring talk “Lipstick on a pig” given by Clive Grinyer at the European Market Research Event.
London Heathrow Airport Terminal 5 forecast that future travellers would be older. Research into older travellers showed they often go into the toilet, so many new toilets were planned.
However, deeper investigation discovered they were going into the toilets….to hear the announcements. It was the only place they could find where they could clearly hear the flight calls! So now the airport is putting new audio areas where you can clearly hear your flight call….
Relations to my thesis: A nice example of the limitations and (sometimes) subjective analysis in user studies. Then it highlights a very interesting adaptation of some people in a very complex and high-tech infrastructure such as an airport.
Posted: January 26th, 2007 | No Comments »
Still in the theme of around messy and heterogeneous vision of ubicomp, a new graphic from Eboy paints the way technologies are playing out, forming some sort of technolological Tower of Babel.
via LUCI’s group
.
