GeoSensor/Dust Networks

This week, a couple of articles mentioned then now technologically feasible and economically viable solutions to develop and deploy sensor devices with multi-purpose on-board sensing and wireless communication capabilities. There is no doubt that tiny powerful devices will be soon cheap enough to be put everywhere, therefor approaching UC Berkeley Kris Pister’s nanotech vision of “smart dust” containing sensors smaller than the eye could see joined into networks larger than the mind could comprehend.

First Declan Butle (2006) 2020 computing: Everything, everywhere, Nature, 440, 402-405 (part of the Future of Computing), mentions a so-called “sensor web” made of interconnected ‘motes’, ‘nodes’ or ‘pods’ that would generate meaningful digital representations of the world. Examples of experiments using this new interface to the tangible and palpable world include glacier surveillance to soil biodiversity control. The main shift in the way data are collected is that information becomes increasingly spatiotemporal instead of just spatial, as sensor feeds capture the evolution over time of the properties they monitor.

However current challenges include device and data homogeneity and standards

The existence of such large networks points to some major challenges down the line, says Estrin. Sensor webs will frequently be just single layers in a stack of data-collecting systems. These will extract information at different temporal and spatial scales, from satellite remote-sensing data down to in situ measurements.

Managing these stacks will require massive amounts of machine-to-machine communication, so a major challenge is to develop new standards and operating systems that will allow the various networks to understand each other. Sensors and networks of sensors will need to be able to communicate what their data are about, how they captured and calibrated them, who is allowed to see them, and how they should be presented differently to users with different needs. The lack of standards is not an insoluble problem for sensor webs, says Shankar “but it is slowing the field down by several years”.

Second, Emergence of GeoSensor Networks by Anthony Stefanidis mentions that the emergence of affordable geosensor networks introduces a novel data collection scheme, with continuous feeds of data from distributed sensors, covering a broader area of interest. In sensor networks the objective is to get many such devices to collaborate and monitor specific phenomena. Each device then becomes a node of the network.

Finally, Sprinkling RFID sensor tags from the Sky mentions that the Japanese Ministry of Internal Affairs and Communications (MIC) started developing system that allows for detailed information gathering about a disaster area by sprinkling RFID sensor tags from the sky (possibly using helicopters.)

Relation to my thesis: Sensor networks are leaving the labs and entering the real-world. The examples in these articles features data collected in the wild with the generated data centrally crunched and analyzed in a lab. It would be interesting to know how humans cohabits with the ever-growing layers of sensor networks and how these networks could be accessed by everybody (and not only be scientists or network deployers). In other words, how to put the human in the center of attention (that being one of the goal of ubicomp). These articles mention the need of interconnecting devices in order for a real meaning of the data to emerge, but fail to mention the connectivity challenges (will they soon be really solved?). Moreover they overlook the impacts on the environment.

Domino: Exploring Mobile Collaborative Software Adaptation

Marek Bell, Malcolm Hall, Matthew Chalmers, Phil Gray, Barry Brown, Domino: Exploring Mobile Collaborative Software Adaptation, To appear in Proc. Pervasive 2006, Dublin.

Domino is an experiment to support raising complexity (complex patterns of interdependence) in handling software plug-ins, a situation that could become a “plug-in hell” as well as the need of incremental adaptation requested by ubicomp applications (because of real-world dynamics of activities, context and preferences). When in proximity with each others Domino system exchange usage information. A collaborative filtering algorithm then recommends the needed plug-in additions and upgrades. The long term goal is to better understand how patterns of user activity, often considered to be an issue more for HCI than software engineering, may be used to adapt and improve the fundamental structures and mechanisms of technological systems.

Relation to my thesis: Current ubicomp constraints force mutual-adaptation of human and technology. Domino is a example of a incrementally adapting system based on all the users usage. Its social-proximity approach makes it very scalable.

“adaptable and adaptive interaction techniques are likely the only scalable approaches to personalisation.”
Weld, D. et al. Automatically personalizing user interfaces. Proc IJCAI 2003, Morgan Kaufmann, 1613-1619

Tribuna de Prensa

Camp Nou, Barcelona.

The Context Gap: An Essential Challenge to Context-Aware Computing

The Context Gap: An Essential Challenge to Context-Aware Computing is title of the Ph.D. dissertation of Louise Barkhuus (december 2004). She proposes that one of the problems of context-aware computing is the ‘context gap,’ the gap between a sensor-derived technical representation of a context, and the social perception of a context. The context gap is inevitable and inherent in that it cannot be bridged; human context can only be represented technologically to a limited extent.

I set out to explore not only whether the context gap exists but also what contributes to it and what consequences this might have.

She conducted 3 case studies to examine the context gap (with embodied interaction as theoretical framework). The first case study identifies four types of context information that are important to users, and analyzes how the context gap is manifested in this situation. The second study investigates the context gap through three levels of interaction: personalization, passive context-aware computing and active context-aware computing. The third study explores the premises and social structures in the environment and traced the context gap from both the human and the technology side.

4 design rules can be drawn from the case studies:

• Let users describe their context
• Let users define the actions from given sensor information
• Inform users of the implication of their use of the technology
• Reevaluate the applications after some time of use

Louise concludes by emphasizing that

the context gap is found to be an inevitable challenge to context-aware computing, which needs to be addressed in order to continue the goal towards supplying users with a smooth and appropriate interaction by way of context-aware computing.

Relation to my thesis: I aim at analyzing the discrepancy between what a group of users need and what technology is capable of in collaborative context. Louis shows that this type of discrepancies is one of the cause of the context gap described by this thesis. The consequences of the context gap created by discrepancies between how the users lead their lives and the functioning of the potential technology, the users simply did not use the service. Another consequence of the context gap is the lack of perceived usefulness by the systems

Unless it (the context gap) is acknowledged, in the early stages of design, we run the risk of developing inappropriate context-aware computing as seen throughout this dissertation.

Mapping Switzerland

In Mapping Switzerland, an exhibition held in 2004, Hosoya Schaefer showed eleven maps of Switzerland and its global context

Visualizations can help to propose new ways of thinking. They can help to see oneself not only in the historically grown context but also in the flux of globalization.

Full article in Weltwoche from December 23, 2004: [PDF, 6.3Mb]

Via P&V

Relation to my thesis: I am interested in different ways to visualize spatial data

GooMaps

GooMaps displays real-time location of planes. However, tt is not clear where the data are coming from.

Via Mauro

Relation to my thesis: I am interested in real-time tracking of things in uncontrolled environments. GooMaps yet another example of real-time Vessel tracking

Cooperative Collision Warning System

The UC Berkeley-based California Partner for Advanced Transit and Highways (PATH) has built a WiFi/GPS-based system for cars based that tracks and displays the vehicle’s location and those around it. The automobiles are constantly forming ad hoc wireless networks as they pass one another and exchange information about their physical place on the road.

One of the concern is the inherent system failures and limitations

Still, the researchers have a long road ahead before the cooperative collision warning system is ready for commercialization. For example, data transmission protocols and error correction algorithms must be improved so that occasional missed bits of data, a given due to the speed and volume of cars on a freeway, don’t result in hazardous system errors.

Relation to my thesis: An example of a proximity-based system which tries to reduce a driver’s workload but leaves some problems of hazardous system errors. How can these errors can be dealt with on the system and user level?

The Scale of Mass Production and the Scope of Customized Solutions

From Mark van Doorn (Research scientist of Philips Research Laboratories Eindhoven) talk at AmIGro.

“To create personalized Ambient Intelligence on a mass basis however, we must also take into account costs”

Mark talks about the current shift to the experience economy in An Inside Story on the Experience Economy.

Via SmartMobs
Relation to my thesis: Costs of personalize customization in a large scale is a constraint of ambient intelligence.

Conferences, Workshops, Special Issues, etc.

The Mobility Studio Stockholm provides an updated list of conferences, workshop, special issues, etc. with their respective deadlines and links to the call for participations.

Getting real with ubiquitous computing: the impact of discrepancies on collaboration

My position paper on the impact of discrepancies on collaboration in a pervasive context has been published in the spanish HCI journal eMinds. It was written last Fall with the support of Nicolas. Of course, now with 6 months of literature review behind me, I see some flaws. However, it gives an overview I what I intend to do in my PhD thesis.

Girardin, F., Nova, N., Getting real with ubiquitous computing: the impact of discrepancies on collaboration. e-Minds: International Journal on Human-Computer Interaction, Volume 1, Number 1, April 2006.

Abstract:

Ubiquitous computing is still a maturing field of investigation. The vision of the seamless integration of computers to people’s life has yet to happen, if it ever has to become a reality. Nowadays, most mobile, distributed systems and sensor technologies have their faults and limitations. Users of ubiquitous technologies often learn to avoid or rectify the systems failures. However, there is still a lack of quantitative information concerning how they impact the collaboration. Therefore, we propose to use a ‘field of experiment’ approach based on a pervasive game platform. Our aim is to rely on a mix of qualitative and quantitative evaluations to find out how uncertainties modified the collaborative processes.