Research

Workshop about Location Awareness at ECSCW

in ,

Might be good to submit something here: ECSW Workshop 5 - Location-Awareness and Community:

Tracking and positioning technologies, such as GPS, 802.11, Bluetooth and RFID tags, have become mature enough to be widely adopted and employed in systems linking information and communication to geographic places and people. While technical challenges remain, social and usability challenges are proliferating.

We are particularly interested in how presence information from multiple sources can be collected, processed and then presented to users in a useful, usable, and unobtrusive way. There are technical challenges such as how to transform raw data into user meaningful representations. And there are associated social challenges ranging from negotiation of what is public and what is private to what kinds of inferences are actually useful (“This seems like your usual route home” is very different from “are you going to the bathroom again?”). Finally, this raises the issues about what are useful applications for these kinds of data and interfaces.

This workshop is designed to bring together practitioners and academics from a range of disciplines and approaches to address these issues within several thematic areas: Algorithmic solutions, Conceptual Design, Interaction and User Interface Design. The workshop will predominantly focus on interactive work around a common scenario where the participants' differing viewpoints can help shape potential solutions to a common scenario.

Experimental psychologists advocated for (video-)game-based research

in ,

One of the most important paper about the use of video games in experimental psychology: The games psychologists play (and the data they provide) by David Washburn, Behavior Research Methods, Instruments, & Computers, Vol 35 (3), pp185-193.

The paper provides a brief history of psychological research with computer games (with a peculiar emphasis on psychology/physiological research). It also highlights the problems caused by the use of game-based reasearch: - difficult to use commercial video games for data capture (not designed for that) - experimenters have not full control over all the variables - consequently it requires to modify the game which is non trivial and time consuming - games can introduce complexity to the experimental situation that has negative consequences (too much context compared to simple psychological tasks) - there is a perception problem associated with the use of games of computer games in psychological research (because games = entertainment for them), not a problem for people who want to study the effects of game on a psychological process but it's an issue for researchers who want to use game-like task or technology to understand the basic process that underlie behavior. That is why people speak about "game-like tasks" and not just games (just to show that they're serious folks and not gamers )

Benefits of game-based research: - more motivation (also pointed by others) and then better performances - enjoyment and well-being - a common platform across researchers - new opportunities for science

Pervasive gaming design and evaluation: a literature review

in ,

A very nifty document on the iperg website (do not pay attention to the crappy flash interface): "Literature Review, Design and Evaluation Methods for Pervasive Games", February 2005. Edited by Steve Benford and Mauricio Capra, The University of Nottingham Contributors:University of Nottingham: Steve Benford, Mauricio Capra, Martin Flintham, Andy Crabtree, Adam Drozd, Leif Oppermann Fraunhofer Institute: Uta Pankoke-Babatz Interactive Institute: Staffan BjörkSwedish Institute of Computer Science: Annika Waern University of Tampere: Laura Ermi, Anu Jäppinen, Markus Montola.

This review identifies and describes the key design and evaluation techniques to be used in IPerG. It summarises key previous work from the field; clarifies the challenges involved in using these techniques for pervasive games and in IPerG in general and also clarifies the mapping of techniques to showcases. The review is intended to act as a resource to be used across IPerG, especially within the different showcases, and then eventually outside of IPerG, providing a resource for the pervasive game development community at large. The following design methods are reviewed: participatory design, scenario based design, ethnographic field studies of current games, cultural probes, game design patterns, game space and artefacts, player game presence, public performance as a research method, and ethical aspects. The following evaluation methods are reviewed: cognitive walkthrough, questionnaires, ethnography of trials with prototypes laboratory experiments and critical review. These methods will be used in different IPerG showcases, providing a broad experience of design and evaluation methods across the project.

Of course this document came out one year after I was struggling finding such kind information ;) But here it is, this EU project deliverable is a nice account of pervasive gaming evaluation methodologies.

Among all those methods, I would just point that there is a lack concerning interaction analysis, since collaboration is an important issue in pervasive computing. I would also add sequantial analysis and some concrete datamining like the ones I described previously today.

Clusters of Catchbob! users

in

Disclaimer: this post might be extra boring if you're not familiar with statistics and/or catchbob! Today I played a bit with principal components analysis and clustering techniques. The point was to categorize the behavior of CatchBob! users.

On this first picture we see the repartition of the users (each number is a user):

The principal components analysis gave me an interesting information, there seems to be 2 meaningful components to explain those data (component 1 and 2 explains a large proportion of the variance as attested by the following histogram). Time and path are the crux components (as shown on the previous picture).

Then I used two clustering techniques: clara and pam (partitioning around medoids). The results are pretty much the same.

Blogposts visualization: semantic distance and cluster

in

Patrick and his students did a pretty good job to visualize blogposts. They wrote a script that parsed my RDF file, then extracted the most important topics following Wise and colleagues' method. At the end of the road, Patrick used R to compute two visualizations. The first one is a representation of the sematic distance between blogposts (thanks to Multidimensional Scaling). The second one depicts a results of a Clara cluster analysis, in which 10 clusters have been built. At the center of each cluster is a prototypical article. Another visualisation, using “magnet” layout to browse topics and posts is under construction. Even tough these visualizations are nice and interesting, they've having hard times making sense of them! Especially the second one, trying to infer meaning from cluster is difficult (getting back to the prototypical posts at the center of the medoids is a solution to get some insights about it).

Location awareness and its effects on rendezvousing

in

A paper close to my research topic (the impacts of location awareness on collaborative processes/behavior): Effect of Location-Awareness on Rendezvous Behaviour by David Dearman, Kirstie Hawkey, Kori Inkpen, Dalhousie University, Canada. Short paper at CHI 2005:This paper is a very interesting account of an field study. It investigates how location-aware technology impacts social behaviour within the context of rendezvousing (meeting at an agreed upon time and location). Three different technology conditions were investigated: Mobile phones / Location-aware handheld computers / Both mobile phones and location-aware handheld computers. Data was collected via field notes, audio recordings, data logging, questionnaires, and semi-structured interviews.

Results:

Regardless of the technology provided to the participants, all of the pairs were able to complete the rendezvous tasks without much difficulty. However, the results of this study clearly demonstrate that the participants exhibited very different behaviours depending on the technology used. (...) Mobile phones are an easy medium to assist people in communicating information about actions and intentions (i.e. ‘what are you are doing?’ or ‘where are you planning to go?’). This information can be difficult to gather from sensor-based devices such as location-aware handhelds. In contrast, >sensor-based devices are very good at gathering overt contextual information, such as location, in a very unobtrusive manner. However, they provide little assistance in interpreting the associated state of the person. In our study, when participants were given both devices, they easily recognized the strengths of each device and utilized each appropriately (i.e. monitoring their partner’s location with the handheld and using the mobile phone to call when they were confused about what the person was doing).

Psychnology special issue about space/place and technology

in

Psychnology's new issue is about Place and Space in Mediated Communication. It's the first issue among two about this topic.

The choice of the Journal is to reflect on the ways in which Space and Place can be connected to human experience in mediated environments, namely to human 'Presence' there. The selection of papers published in this issue of PNJ offers an overview on these and other ways in which Space and Place matter in the human experience with digital technologies. So many excellent contributions have arrived that two issues of the Journal will be devoted to this theme.

Intersting papers inside, I will talk about some of them later on.

Action replay, mirrors neurons and mentally “going through the motions”

in

A striking and unexpected paper in the FinancialTime about mirror neurons: neurons which are activated when mimicking the actions of others. They have been discovered by Giacommo Rizzolatti (at the University of Parma, Italy). The paper is much more focused on inferring other actions when carrying out a joint action like dancing (uh I use a bit too much the vocabulary of the CSCW/herbet clark's theory of action here!)

Research shows that when subjects watch films of ballet or capoeira (a Brazilian martial art), the same areas in the brain are activated as those used to execute the movements. When watching motion, the brain “moves” along every step of the way, so much so that it stimulates physiological responses - such as increased oxygen consumption - to the point where the weak-hearted might suffer a heart attack merely watching strenuous sports.

This is why mentally “going through the motions” is just about as good as rehearsing to improve a dancer’s or sportsman’s performance. To observe, then, is to dance. (...) This finely honed perception of human movement, the ability to read body language - and readily to perceive and express our own - is known as social intelligence. This capacity to navigate our social world means we can work out “where others are coming from” (are they angry or happy?) or “where they are going” (are they coming to yell at me or to ask for help?) so we know how to react accordingly.

To mirror others is to empathise, using the same mental rehearsal of the body language of others that allows us to put ourselves in someone else’s shoes. Some of us fail miserably, while others can really “feel your pain”. It enables us to recognise a friend at a distance just by their gait and subconsciously to acquire the mannerisms of our spouse. And it is why, like yawning, dancing is contagious.

This is very interesting and relevant to our new project at the lab: we are stuyding mutual modeling: the psychological project involved when people do inference about others' actions/goals/purposes.

Location-based services and collaboration

in

Finally! A very relevant (to my needs/with regards to my interests) about location based services and their uses to support collaboration:From CHI 2005: Life on the edge: supporting collaboration in location-based experiences by Steve Benford, University of Nottingham (UK), Duncan Rowland, University of Nottingham (UK), Martin Flintham, University of Nottingham (UK), Adam Drozd, University of Nottingham (UK), Richard Hull, Hewlett-Packard Laboratories (UK), Josephine Reid, Hewlett-Packard Laboratories (UK), Jo Morrison, NESTA Futurelab (UK), Keri Facer, NESTA Futurelab (UK).

We study a collaborative location-based game in which groups of 'lions' hunt together on a virtual savannah that is overlaid on an open playing field. The game implements a straight-forward approach to location-based triggering in which players must be in the same spatial locale in order to share information and act together. Comparison of video recordings of physical play with system recordings of game events reveals subtle and complex interactions between highly dynamic player behavior and the underlying technology. While players exhibit a fluid approach to group formation, the system embodies a more rigid view, leading to difficulties with sharing context and coordinating actions, most notably when groups of players span virtual locale boundaries or initiate actions while on the move. We propose techniques for extending locales to support more flexible grouping and also discuss the broader implications of our findings for location-based applications in general.

At least I find a very smart account about how a collaborative LBS has been used! I'll put more annotations here later on.

Ubiquitous Computing, Entertainment and Games

in , ,

Interesting workshop at Ubicomp 2005: Ubiquitous Computing, Entertainment and Games:

The workshop’s topics of interest include • New ubiquitous game designs • Mixed traditional and ubiquitous game design • Studies or reports on the compelling aspects of existing ubiquitous games • “Post-mortems” on past ubiquitous game experiences • Taxonomies of ubiquitous games • New, existing and emerging technologies supporting ubiquitous gaming platforms • Mechanisms to evaluate or test ubiquitous games • Design elements that can be learned or adapted from popular console/PC games for creating compelling ubiquitous games • Business models that will enable ubiquitous gaming to be successful in the marketplace

Great people on board! (hi Timo!) Maybe it would be nice to submit something about CatchBob ;)

Location awareness and fluidity of work

in

Kakihara, M., Sørensen, C., and Wiberg, M. (2004). Negotiating the fluidity of mobile work. In Wiberg, M., editor, The Interaction Society: Practice, Theories, & Supportive Technologies. Available from: http://mobility.lse.ac.uk/download/KakiharaSorensenWiberg2002.pdf. Some excerpts:

This paper addresses one particular aspect of organisational life for mobile workers, the constant negotiation of fluid work, based on the assumption that an essential aspect of mobile work is the negotiation of desirable versus disruptive interaction. (...) In order to initiate the debate we ask the question: What are the pertinent issues involved in individuals negotiating mobile work? This is based on the assumption of temporary asymmetry between individual mobile workers in terms of fluid mobile work – what for one person is a perfectly justifiable request can for another be a disruption. (...) [and the most relevant part with regard to my interests:]

Establishment of mutual awareness of location has been promoted as an important element of mobile interaction (Mäenpää, 2001), but also conflicting accounts of the awareness of activities as the primary element has been promoted (Weilenmann, 2001). However, in both cases, a generalised notion of location awareness is being negotiated since Weilenmann argues that the awareness of activity infers awareness of location. It can, therefore, be argued that the use of mobile phones, for example, socially constructs a location based services, both in terms of allowing constant update mutual awareness of locations, as well as in bringing the interaction to the location. Much research has discussed applying specific awareness technologies, AwareWare (Nilsson et al., 2000). The most common of these technologies is a stationary interaction technology in it's own right, namely Instant Messaging, where platforms such as ICQ allow participants explicitly to declare the interactional status. Other systems, such as the one reported by Nardi et al. (2000) supported implicit awareness by monitoring user keystroke rates and therefore enabling others to gain an impression of whether or not the person to be contacted is situated by their desk, or alternatively they perhaps is too busy to be contacted at all. Tang et al. (2001) demonstrate a multi-platform mobile awareness system with implicit location logging. Dix et al. (2000) suggest a generic systems architecture for mobile awareness technologies that integrates the technologies technical "awareness" of internal state with the inclusion of the external context to support mutual awareness between users.

Disclosing or not one's location

in

A relevant paper for my current research: Consolvo, S., Smith, I., Matthews, T., Lamarca, A., Tabert, J., and Powledge, P. (2005). Location disclosure to social relations: why, when and what people want to share. In Proceeding of the SIGCHI conference on Human factors in computing systems, pages 81–90, Portland, Oregon, USA. ACM press: NY. Available from: http://guir.berkeley.edu/pubs/chi2005/p486-consolvo.pdf.

Advances in location-enhanced technology are making it easier for us to be located by others. These new technologies present a difficult privacy tradeoff, as disclosing one's location to another person or service could be risky, yet valuable. To explore whether and what users are willing to disclose about their location to social relations, we conducted a three-phased formative study. Our results show that the most important factors were who was requesting, why the requester wanted the participant's location, and what level of detail would be most useful to the requester. After determining these, participants were typically willing to disclose either the most useful detail or nothing about their location. From our findings, we reflect on the decision process for location disclosure. With these results, we hope to influence the design of future location-enhanced applications and services

"Geocollaboration"

in

(via), this project seems interesting with regard to our research goals: Geocollaboration using Peer-Peer GIS. We're exactly in this geocollaboration topic:

The Penn State researchers define the activity of geocollaboration as a group working together to solve geographic problems facilitated by geospatial information technologies. Others view geocollaboration as any form of collaboration that involves geospatial data.

One level of collaboration in a geospatial context involves sharing of data and/or applications by and among various organizations. This level of geocollaboration is at the less interactive, open community end of the spectrum and illustrates the trend toward increased awareness of the need for better geocollaboration capabilities.

An example of a collaborative GIS tool that has been developed to support multi-agency cooperative work with geospatial data is the Geospatial One-Stop. It uses the approach of providing portals or channels to various web sites that support sharing of geospatial data and applications.

Even though it's more focused on geo-groupware, there's a lot to learn from this field.

\"Geocollaboration\"

in

(via), this project seems interesting with regard to our research goals: Geocollaboration using Peer-Peer GIS. We're exactly in this geocollaboration topic:

The Penn State researchers define the activity of geocollaboration as a group working together to solve geographic problems facilitated by geospatial information technologies. Others view geocollaboration as any form of collaboration that involves geospatial data.

One level of collaboration in a geospatial context involves sharing of data and/or applications by and among various organizations. This level of geocollaboration is at the less interactive, open community end of the spectrum and illustrates the trend toward increased awareness of the need for better geocollaboration capabilities.

An example of a collaborative GIS tool that has been developed to support multi-agency cooperative work with geospatial data is the Geospatial One-Stop. It uses the approach of providing portals or channels to various web sites that support sharing of geospatial data and applications.

Even though it's more focused on geo-groupware, there's a lot to learn from this field.

Mobile phones data mining

in ,

(via), the reality mining project:

The Reality Mining experiment is one of the largest academic mobile phone projects in the US. Our research agenda takes advantage of the increasingly widespread use of mobile phones to provide insight into the dynamics of both individual and group behavior. By leveraging recent advances in machine learning we are building generative models that can be used to predict what a single user will do next, as well as model behavior of large organizations.

We are currently capturing communication, proximity, location, and activity information from 100 subjects at MIT over the course of this academic year. To date, we have collected approximately 350,000 hours (~40 years) of continious data on human behavior. Such rich data on complex social systems have implications for a variety of fields. It is our hope that this research will help us explore research questions including:

  • How do incoming students' social networks evolve over time?
  • How entropic (predictable) are most people's lives?
  • Can the topology of a social network be inferred from only proximity data?

How can we change a group's interactions to promote better functioning? If you have a Symbian Series 60 Phone (such as the Nokia 6600) with a data plan, you can participate.

Close to this SmartFriend project 8a tool that compute statistics on your mobile phone about social things like with whom are you spending the largest amount of time on the phone? bet on your next call or gender stats).

They wrote more about their methodology in this paper (Personal and Ubiquitous Computing). Nathan Eagle, one of the researcher in charge of this reality ming project is interviewed in the Feature. Here are relevant excerpts:

Eagle: I primarily look at mobile phone data that can be broken down into three types: location, communication and proximity patterns. We use cell tower IDs to get approximate locations within a few blocks. Communication logs reveal who is calling and texting whom and how often. And Bluetooth scans every five minutes show who is proximate to you.

Eagle: We can do behavior prediction. Depending on the life you lead, I can predict what you're going to do next based on very limited information. Whether it's your morning Starbucks fix or your Saturday afternoon softball game, everyone lives life in routines. One of our algorithms extracts these routine patterns from everyone's daily lives.

Eagle: There has been a lot of work on building more user-centric interfaces. So the kind of data we gather could automatically change the phone functionality according to a certain demographic. For example, Nokia, one of the sponsors of this research, is selling the same phone to soccer moms, power executives and texting teenagers. With just a few days worth of data, we can characterize the user and their usage. Once we do that, we can customize how the phone looks and operates for specific groups of people.

France Telecom is also working on that topic, using neural network to discriminate social patterns. And their "social serendipity" tool is close to Jamie Lawrence's research project.

Comparison between augmented and virtual reality

in

Tang, A., Biocca, F., and Lim, L. (2004). Comparing Differences in Presence during Social Interaction in Augmented Reality versus Virtual Reality Environments: An Exploratory Study In Proceedings of PRESENCE 2004, 7th Annual International Workshop on Presence, October 13–15, 2004, Valencia, Spain.This paper offers an interesting conclusion with regard to the differences between augmented reality and virtual reality: "the absence of representations of the user’s body in VR environment may lessen sense of spatial presence comparing with AR environment".

I like what this lab does, namely experiments about cognition in AR.

Student needed: diploma or semester project

in

For CatchBob! we need interns or students! Here is the project:Development of a mobile game replay tool

The project consists in developing a tool to replay sessions of interactions with a mobile game called CatchBob!. This game is an experimental platform in the form of a collaborative mobile application for running psychological experiments. In our lab, we are indeed interested in studying how people use mobile devices to carry out collaborative activities.

In this context, a replay tool is a piece of software that displays all the actions undertaken by the players. Such a tool is intended to rebuild and enrich our comprehension of how users performed the activity and to elicit their social and spatial behavior. This visual information might be used both by the researchers (to better understand what happen and compute statistics about it) and the players (to offer a visual support in order to explain the researchers what happened).

Example of the replay tool we have so far: path + time are represented + logfile at the bottom.

The project will both address the visualization of interactions as well as the computation of various indexes.

Keywords: teamwork analysis, design of a visual grammar, dynamic image generation, visualization of qualitative Information.

Technologies: Java, XML, SVG

References:

Contact: Please email Nicolas Nova (nicolas(dot)nova@epfl.ch) to set-up an interview. Please include information on the courses you have taken and projects you have done in the past.