New German book “on everyday computing” available

In 2005 Friedeman Mattern organized a symposium at ETH Zurich on how computing impacts everyday life (http://www.comp21.inf.ethz.ch/). He edited a book (Die Informatisierung des Alltags. Leben in smarten Umgebungen, @Amazon) which includes versions of most of the talks. The book is in German.

I contributed a chapter to the book (draft version) on the symbioses between humans and computers. In the paper the idea of novel user interfaces that augment human capabilities and improve our ability in what we can do with technology is assessed. It is mainly based on the work done in the DFG funded project Embedded Interaction.

If you can read German I highly recommend the book. It is an interesting collections on viewpoints of pervasive computing. There is also a great chapter (the last in the book) by Friedeman Mattern himself discussing old a new visions of technologies.

acatech workshop: object in context

It was interesting to see that smart objects / smart object services, context, NFC, and RFID become very mainstream. It seems that nearly everyone buys into these ideas now.

Dr. Mohsen Darianian (from Nokia Research, same building as Paul Holleis is at the moment) showed an NFC-advert video which reminded me on the results of an exercise we did on concept videos within an HCI-class at the University of Munich :-)

Overall it seems that acceptance and business models are of great interest and that to create them a lot of technical insight is required. The issues related to user interfaces, interaction, experience become central factors for the success of products and services.

One discussion was on the motivation for people to contribute (e.g. user generated content, write open source code, answer questions in forums, blogs). Understanding this seem crucial to the prediction whether or not a application is going to fly or not.

Besides contributing for a certain currency (e.g. fame, status, money, access to information) it seems that altruism may be an interesting factor for motivating potential users. Even if it is a low percentage within our species the absolute number on a world wide scale could be still enough to drive a certain application/service. There is interesting research on altruism in the animal world (or at the researchers page http://email.eva.mpg.de/~warneken/ ) maybe we should look more into this and re-think some basic assumptions on business models?

Our break out group was in the rooms of the Institute of Electronic Business e.V (http://www.ieb.net/). It is a very pleasant environment and their link to the art school reflects very positive on the atmosphere and projects. The hand drawn semacodes were really impressive.

Workshop dinner, illuminated faucet, smart sink

I first saw a paper about a context-aware sink at CHI 2005 (Bonanni, L., Lee, C.H., and Selker, T. "Smart Sinks: Real World Opportunities for Context-Aware Interaction." Short paper in proceedings of Computer Human Interfaction (CHI) 2005, Portland OR).

Yesterday I saw a illuminated faucet in the wild - one which looked in terms of design really great (in the restaurant they even had flyers advertising the product). But after using it I was really disappointed. It uses the concept of color-illumination of the water based on temperature (red hot, blue cold).

The main issue I see with the user experience is that the visualization is not based on the real temperature using sensor but on the setting of the tap. Hence at the beginning when you switch on hot the visualization is immediately red – even though it is initially cold :-(

Conclusion: nice research idea some time ago, a business person saved a few cents for the senor and wiring, created a product with great aesthetics and a poor user experience; hence I left the leaflet with the ordering address there, don’t want to have it.

Large scale sensor network connected to public displays

The airport Köln-Bonn (CGN) has all the parking spaces monitored with a simple sensor (detects if there is a car or not) and provides displays at the entrance showing the number of open spaces and has active signage in the parking garage leading to the free spaces - additionally it is visualized above each space – probably more a maintenance functions to see if the sensor works.

(looking at the pictures I have probably parked on women-only parking spots...)

3 months have passed very quickly

Nigel, Oliver, Mike, and Andre from Lancaster University have been our colleagues at B-IT for the last 3 months. The time has passed very quickly but it was very inspiring to work together.

We worked together on a project using Bluetooth to enable implicit interaction in public spaces. We have created together an interesting demonstrator in the domain of advertising and in particular exploring the implications for targeting public adverts to customers.

Today was the last day for Mike and Andre in Bonn (so far) and we went for dinner together. Nigel will move on for the second half of his sabbatical to ETH Zurich.

To keep us challenged and busy Nigel’s son asked some of us to convert his transformer from robot to car, and as one can see on the picture this is even for Dagmar not a trivial task ;-)

Ubisense system at IAIS used in the lab

During our lab on context- and location-aware systems we have currently one task to create a novel application using the Ubisense indoor location system. We have set up a little server in C# that reads out the tag-ids and the current x,y,z coordinates and multi-casts it via UDP to several clients.

The students have to create an application that reads the tags and their locations from UDP and make use of it. One first interesting result was a heat-map, that would allow to map out the usage of the room. By the end of term we hope to have a nice set of applications.

For many students it was the first time at the Fraunhofer lab and hence they found interesting things to play with. E.g. Diana explored the PointScreen , a system developed in the division Virtual Environments at Fraunhofer IAIS. Further reading on the point screen:
Li et al. Gesture Frame – a Screen Navigation System for Interactive Multimedia and Strauss et al. : Information Jukebox – A semi-public device for presenting multimedia
information content (2003).

Off-Topic: How defend against Cyberwar

During the last weeks there has been a lot of discussion about the vulnerability of infrastructures after the experience in Estonia. One interesting question to me is if a country will need in the future the power for cyber deterrence.

(Given the current discussions on computer security in Germany I have one option: a bot network that includes all computers in a country that are connected in the internet - could be added as an additional function in the planed German Bundestroianer ;-)

How will wasting energy be judged in the future?

Back in Bonn after my holidays I had to catch up on a few things. When I left the B-IT after midnight I was stunned by the sight of the post tower (http://de.wikipedia.org/wiki/Post_Tower) - as it is all lit up. Another example of a interesting and large scale public display.

However I wonder if behavior where energy is used to quite some extend (or wasted) will be judged in 20 years similar to the way we see nowadays the pollution of rivers and air that was common in the first half of the last century.

Public Displays and Responsibility for Content

Antonio Krüger at the University of Münster is running an infrastructure of public displays that show various kinds of information. Using a web editor a select set of people (mainly staff at the department) can input and manage the information chunks that are presented.

In our discussion it became obvious that running such public displays comes with a lot of responsibilities and that people are very quick at complaining about content (may it be censorship or offending content). This leads then to more or less closed and controlled system – but I wonder if we are not overcautious or the expectations around us are too high.

I took a picture of one door in a restroom in the University. It is converted in a public display by people (anonymously) using a pen – and its content is neither politically correct nor suitable for children. However this is rightly blamed on the people who do the damage and not on the administration or designer that decided that the doors are white and made of a material one can write on.

Visit at the Institute for Geoinformatics in Münster

Yesterday we had the opportunity to see a set of research demos at the lab of Antonio Krüger at the University of Münster. We had some time to discuss research projects and in the afternoon Nigel and I gave a talk at the Geoinformatics seminar.

We saw exciting work in progress - a nearly ready large scale multi-touch display based on frustrated total internal reflection – according to Antonio the world-largest at the moment. The principle of operation of such a display is very appealing and the demo was convincing. For a comprehensive introduction on the topic see Jef Han’s UIST 2005 technote “Low-cost multi-touch sensing through frustrated total internal reflection” http://doi.acm.org/10.1145/1095034.1095054) . I really think we should build one, too (obviously a bit bigger than Antonio’s ;-)

On the Mobile Phone While Working?

It seems that recently I come across many people that speak on the phone while they do their work. In Toronto on the bus to the airport the driver spoke on the phone (telling someone how to find and edit a file in Windows) while driving. Here in Bonn I saw it in shops and on the ferry – it felt really awkward to interrupt people in their phone conversation just to pay my ticket or bread.

At the moment most people speak while holding a handset – but given they use BT-headsets one could image new working practice ;-) e.g. driving a bus and doing a call center job on the side. I would expect there will be some regulation soon…

Public Displays – Making Life More Predictable

On my way home from Toronto it was surprising how many public displays I saw that provided me with ”information about the future”, e.g. telling me when I will be out of time to cross the road, when the next train is due or when my luggage will arrive. These kinds of predictions or contexts are simple to gather and easy to present and best of all: the human is in control and can act on the information. Overall it is reassuring even if the context information is wrong (this is another story about my luggage ;-).

Pervasive 2007 in Toronto

The internatonal conference on pervasive computing in Toronto had an exiting program.

The keynote was by Adam Greenfield on “Everyware: Some Social and Ethical Implications of Ubiquitous Computing” – matching a number of issues we discussed the day before at the doctoral colloquium. The talk was enjoyable even though I think some of the statements made, in particular with regard to opting out and informing the users (e.g. logos) are over-simplified. Furthermore the fact that our society and its values are changing was very little reflected, e.g. privacy is not a constant.

The best paper (by Rene Mayerhofer and Hans Gellersen) “Shake well before use: Authentication based on accelerometer data” was my favourite, too. A further very interesting paper was “Inference Attacks on Location Tracks” by John Krumm. Two papers from ETH Zürich were also quite interesting: “Operating Appliances with Mobile Phones - Strengths and Limits of a Universal Interaction Device” by Christof Roduner et al. showed surprising results for the use of phones as remote control (in short – more usable than one thinks). And “Objects Calling Home: Locating Objects Using Mobile Phones” by Christian Frank et al. showed that phones have a great utility as sensors (in this case to find lost objects.)

We presented a in the video proceedings the smart transport container and a novel supply chain scenario (cutting out all intermediaries and enabling producer to customer transactions).

The tutorial day was excellent - I think the set of tutorials presented can give a good frame for preparing a course or lecture on pervasive computing.

Pervasive 2008 will be in Australia!

Pervasive Computing and Ethics

Together with Boriana Koleva I organized the doctoral colloquium at Pervasive 2007 in Toronto. We had 9 students presenting and discussing their PhD work with us.

One central observation was that we come to a point where we have to make more and more ethical decisions. Many things that are technical feasible and harmless within the lab may have sincere implications in the real world. If technologies for tracking, tracing and mining (e.g. social network analysis, location based services, context-aware systems) are deployed beyond the lab the question of choice becomes a real issue – are users aware of it and can they opt-out?

In the area of context-awareness technology has moved on since I started my PhD on the topic nearly 10 years ago – but amazingly scenarios did change little. Automatically detecting a meeting is still on the students slides. The more I learn and understand about context-awareness the more it becomes apparent that this apparently simple use case is amazingly hard!

Public Displays in Restaurant Bathroom

After the Ubicomp PC meeting we went to a nice restaurant in Toronto for dinner. In the bathroom they had mounted TFT-screens above the urinals showing a TV program or adverts (was hard to tell in the short time I was there). It found it was quite distracting. The displays reminded me of a poster I saw at CHI 2003: You’re In Control: A Urinary User Interface by Maynes-Aminzade and Raffle (http://doi.acm.org/10.1145/765891.766108). Given the distraction experienced, I wonder if a design where visualization and control is spatially separated and hence control is indirect make sense for such applications.

In summer during the soccer world championship in Germany I saw a low-tech version of a bathroom game: a ball on a string in a goal. Here control and visualization is in the same place and it felt more natural to use. (sorry for the low quality pics – they are done with my old phone).

Observation at FRA, Terminal 1 B

The number of power plugs available to the public seems to be very close to zero at Frankfurt airport. If a persons sits somewhere on the floor in an odd corner it is likely that there is energy for the laptop or phone.

The number of Bluetooth IDs visible when scanning is amazing. It seems that many people have it switched on continuously now (quite different from 2 years ago). The friendly name used by people seems fairly boring, mainly the preset model name of the phone, combination of first name an phone model, initials, full name and the occasional “hi there”. These observations are quite encouraging for one of our projects.

Looking at the scans I wonder if it is likely that people who travel together have similar phones (e.g. same manufacturer).

Nothing Matches Real Experience

During our strategy meeting there was time for a canoe adventure and some late night reflections. To get the full experience we chose the rainy afternoon for our trip on the river Lahn. Even though we were pretty wet after the trip (some more some less ;-) it was a great experience.

When comparing real experience vs. virtual experience (e.g. second life) it becomes clear that the central issue is, that the virtual is risk-free with regard to our immediate physical well-being. This sounds great in the first place. But what does it lead to when we live in a risk free environment in the long term? How will it shape our perception in the further?

Enjoying the real experience inspired some ideas for mobile adventure games that take place in the real world with real experience but including virtual aspects. A central design goal would be to create a game, where the technology becomes invisible and the user only realises his or her activity in the real world.

Audio Tapes Soon be History, Printed Paper Next?

News papers in the UK have reported that sales for audio tapes at Currys were down from 83 million in 1989 to 0.1 million in 2006 and hence Currys is going to stop stocking them. We recently discussed how expensive talking toys (in particular dolls) were just twenty years ago. The were based on mini records or endless tapes. By now storage chips are so cheap that a good birthday card can sing you happy birthday.

How long will it take before electronic paper will replace printed paper in the large? At least it is close enough to seriously think about applications and business models that arise.

Ensuring Privacy - trust in the physical

After having a really interesting discussion on privacy with a student at CDTM with regard to implicit interaction I saw the depicted privacy solution on the train back to Bonn. The woman had her notebook camera disabled – not in software – but physically with a scotch tape and a piece of paper. Such solutions are not uncommon and remind one impressively that people want tangible control over their privacy. It seems that people trust in the physical much more than the virtual – and for a good reason.

I reduced the size of the picture as she was preparing an exam paper (school, 7th grade math) and in full resolution details are readable - so much about privacy. One could make that mental note not to edit/view any private document on the train even in small print as it is very quick to talk a phone (even with a phone) and read it afterwards ;-)

Lecture at CDTM in Munich

The CDTM is a joint elite study program from LMU Munich and TU Munich (http://www.cdtm.de/). It offers a set of complementary course for students from different backgrounds including math, business studies, engineering, computer science and media informatics. In their course on multimodal HCI I gave today an introductory lecture on the motivation for and the basics of user interface engineering.

In the seminar room was a real typewrite – in fact a travel type write. It was a real good prop to discuss micro- and macro-efficiency.

One question about the sustainability of a competitive advantage based on user interfaces made me really think. As the user interface is visible it is really hard to protect the competitive advantage and IPR are difficult on this topic. Desktop GUIs are a good example how quickly ideas propagate between competing systems. The only real option to maintain a competitive advantage is to continuously innovate – keeping the status quo means falling behind. The fact that one can not keep new user interface concepts secret (if one includes them in a product) is one of the most exciting aspects of user interfaces research – it is fast moving because everyone shows off their results!

Our Presentations at CHI’07 in San Jose

At this years CHI - the Conference on Human Factors in Computing Systems - we presented 3 contributions: a full paper, a CHI-note, and a work in progress paper. Have a look at them!

Holleis, P., Otto, F., Hussmann, H., and Schmidt, A. 2007. Keystroke-level model for advanced mobile phone interaction. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (San Jose, California, USA, April 28 - May 03, 2007). CHI '07. ACM Press, New York, NY, 1505-1514. DOI= http://doi.acm.org/10.1145/1240624.1240851

Atterer, R. and Schmidt, A. 2007. Tracking the interaction of users with AJAX applications for usability testing. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (San Jose, California, USA, April 28 - May 03, 2007). CHI '07. ACM Press, New York, NY, 1347-1350. DOI= http://doi.acm.org/10.1145/1240624.1240828

Holleis, P., Kern, D., and Schmidt, A. 2007. Integrating user performance time models in the design of tangible UIs. In CHI '07 Extended Abstracts on Human Factors in Computing Systems (San Jose, CA, USA, April 28 - May 03, 2007). CHI '07. ACM Press, New York, NY, 2423-2428. DOI= http://doi.acm.org/10.1145/1240866.1241018

Visit at the Academy of Media Arts Cologne

Today I was invited at the Citizen Media seminar to discuss mobile and ubiquitous computing topics with people working in the project. Fraunhofer IAIS and the Academy of Media Arts Cologne are both partners in the European Citizen Media project (http://www.ist-citizenmedia.org/). It is a difficult question how to create and support a mobile community. The provision of software and infrastructure is obviously required - e.g. Alexander De Luca and Michael Müller (students I supervised in Munich) designed and implemented an open source software as a basis for mobile blogging (mobile reporter) - but the process that forms and evolves specific community is still little understood.

It was great to meet Georg Trogemann, who is professor for audiovisuell art and computer science at the Academy of Media Arts Cologne. Only after I had left I realised that he is the author of a quite interesting book a recently got (Code@Art; ISBN 3-211-20438-5).

We had an interesting discussion on how to most effectively involve users in the design process of novel products. In particular when we expect that technology drives innovation and when future user needs are to be anticipated. I reported from our very positive experience with technology probes (article at IEEE Percom). To me it is central to involve users from the very beginning and throughout all stages of a project and at the same time allow technology to drive innovation beyond current users’ needs.

We had much too little time to see all the interesting projects that are going on there so we have to go back there ;-) the lab and setup in Cologne reminded me of Bill Gaver's group at RCA (when we worked together in Equator some year ago).

What happens if you ask the customer (user)?

Dell started some time ago a web page where the public was asked for comments on improvement (http://www.ideastorm.com) – by now the list is quite impressive – at least in length. Such open community processes are really exciting to watch. I am curious which of those suggestions are implemented and how successful it is to listen to customers direct responses. If it works well others are probably trying it, too…

To make the most out of these suggestions and comments it seems essential to perform a detailed analysis (perhaps taking into account temporal dynamics of the conversation additionally to the content) – looks like another really interesting field for text-mining.

Don’t wash your dishes before loging on to the computer

This morning I washed my dished (manually in a kitchen sink) and afterwards (with dry hands) I tried to log on to my computer using the finger print scanner. My thumb did not work anymore. Looking closer at my thumb it seemed clear that this looked temporarily different as it was in the water for 10 minutes. A few minutes later the thumb is back to normal, but I already logged using a password. What do I learn from this – If I will do a ubicomp project in the kitchen (similar to things we did in Munich) I will not use finger print scanners for authentication.

What would happen if I had to wash dishes before entering the US ;-)

Panel on Users as Producers at Schloss Birlinghoven

In the castle on campus Ute Schütz and Michael Krapp from IAIS and SCAI organize a public panel discussion on citizen journalism and Web 2.0 trends. The discussion looked at the topic from several angles including technology, content, and communication. I had the honour to be on the panel with Wolfgang Back, Frank Patalong, Moritz „mo.“ Sauer and Thomas Tikwinski.

On issue was how much blogs are (mis)used to transport information or to do advertising. The web seems to be to many people a very believable medium. This reminded me of an article I read some time ago on story-telling o n the web (Miller, J. 2005. Storytelling evolves on the web: case study: EXOCOG and the future of storytelling. interactions 12, 1 (Jan. 2005), 30-47. DOI= http://doi.acm.org/10.1145/1041280.1041281). I think something along this lines would an interesting project with students.

The questions what is going to change in the near future on the WWW brought many trend statements out, but for me it is amazing that most of the infrastructure and technology we need to make this happen is out there. One comment which I think is really true is that it is more an more about content and communication again. As Thomas said before the panel – most of the great things that are called Web 2.0 have been in the original proposal for the WWW (e.g. annotations by everyone or users as editors) – but now technology is finally there that people can use it.

Girls' day at Fraunhofer in Birlinghoven

In Germany there are still too few girls and women interested in studying technical subjects – it may have different reasons but I think trying to convince young girls that technology is really exciting is a good way of addressing the problem.

We had this morning 7 girls visiting our lab. Matching our current teaching at B-IT (and using the test implementation Dagmar made ;-) we offered the topic “The computer knows where I am – how does this work”. First we played a mini geocaching game where they had to find a bag of jelly babies behind the castle.

After they had some experience with GPS and an electronic map we explained how it works and even parse together a NMEA-0183 sentence. We also discussed some application ideas, e.g. kids monitoring with GPS. The discussion, in particular the privacy issues that came up, were quite interesting.

I can really see that for some projects running focus groups with kids could be fun for them and a great value for the projects.

Why shopping is fun – thoughts on intelligent user interfaces or why n=1 is not enough

Some weeks ago I saw for the first time one of the intelligent scales in the wild (=outside the lab). At that time I was really impressed how well it worked (sample size: n=1, product: banana, pack: no-bag, recognition performance: 100%). Last time I was too late so there was no time to play or see other people using it – but today I had some 5 minutes to invest.

The basic idea of the scale is simple and quite convincing. The customers put their purchase on the scales. A camera makes a guess what it is and the selection menu is reduced to the candidates that match the guess of the camera. Additionally, there is always a button to get all the options (as in the old version without the camera). It appears that this should make things easier.

I observed people trying to weigh different fruits and vegetables in bags and without bags (obviously I tried it myself, too). It did not work very often but interestingly people did not care much. It looked as most people did not really realise that this is meant to be an intelligent user interface. They probably just wondered why the display is showing always different things, but as they are intelligent themselves they found a way to deal with.

Overall it seems that it does really well on bananas which are not wrapped in a bag (my initial test case) and does not too well on many other things. I think the scales are an interesting example of a invisible interface.

Overall this is again a remainder that user tests that are small may be utterly wrong.

A magic lens for the mass market?

Making things visible that can not be seen with the naked eye? Overlaying personalized information onto objects or images? Such concepts make good fiction but are there interesting use cases? Michael Rohs from T-Labs in Berlin visited B-IT and Fraunhofer IAIS today and he showed us in his talk and demos several such scenarios that appear not to be far in the future or fictional anymore.

Michael has developed during his PhD at ETH Zürich the Visual Codes system (http://people.inf.ethz.ch/rohs/visualcodes/) that provides a basis for augmented reality interaction on mobile phones. Some of his current work, in particular overlaying information on large paper maps, shows impressively the potential of using personal mobile devices, such as phones, as interfaces to combine static and dynamic information. I think for everyone trying out information overlays using a phone can easily imaging that this could be commonplace pretty soon. The question is more what the first pervasive and convincing applications are for mobile augmented reality and when will we find them in the wild. In our discussion a number of interesting application areas came up, in particular games and advertising seem very appealing.

In the morning Michael got a tour a B-IT and some demos. One of our tasks in the practical course developing location and context-aware systems is also related to a magic device from the Harry Potter book – a map with a moving point ;-)

Till Schäfers, a student at B-IT who is currently at T-Labs in Berlin for his master thesis (supervised by Michael and me) gave this morning a presentation on the work he started. We had a longer discussion on issues of teleconferencing – and in particular on mobile teleconferencing – were many interesting ideas and issues came up. (Remark on research&realty: somehow it is sad that even after many year of research in teleconferencing the tools we use to do meeting over the phone in our daily work are still poor – but things are getting better). Another issue we discussed in detail is the questions of how to involve the user in the design process even more while giving at same time the user interface designer the freedom to designs and defines novel ways of interaction and exciting interfaces.

Besides the scientific exchange it is great to have visitors to learn about new gadgets. Michael and Till showed the SHAKE SK6 sensor /actuator attached to a phone – quite an interesting tool for research.

Steffi Beckhaus Showed us Virtual Reality beyond 3D Visual Displays

Steffi Beckhaus, who is professor for computer science at the University of Hamburg, visited our group at B-IT. Meeting her was another classical example how small the scientific community in user interface research is. I met Tanja Döring, one of Steffi’s students, at TEI’07 in Baton Rouge. They had a very interesting paper on novel user interfaces for art historians - “The Card Box at Hand: Exploring the Potentials of a Paper-Based Tangible Interface for Education and Research in Art History”. Looking then up Steffi Beckhaus details I saw that she was at Fraunhofer IMK (which is now part of IAIS) in Birlinghoven some year ago in the virtual environments group.

After lunch we had a few demos (like always on short notice as I forgot to tell before). Florian Alt demoed the current stage of his annotation platform for the web and Paul Holleis showed some examples of the work on modelling physical interaction and cross device prototyping, which we will present at CHI in 2 weeks. We also showed one of the student projects from the last course on developing mobile applications (CardiViz) and the ongoing work of our current lab on context and location awareness. We realized that we have very similar values and methods for teaching. In particular forcing students to bring in their own creativity into projects which they drive and for which we set a corridor seems a very efficient way to teach people who to create novel user experiences.

In her presentation Steffi showed us details about her lab in Hamburg (we were so impressed that we invited ourselves for a visit). In particular the combination of “classical VR” and tangible and novel user interfaces is intriguing. Overall her work is greatly interesting as it looks into the whole body experience (e.g. sound-floor, chairIO) and connects much more than I expected to our research theme of embedded interaction.

In our discussion Steffi brought up a video of the Pain-Station (http://www.fursr.com/) . It is basically a pong game where you get penalized for low performance with actuator that creates pain (I think with heat and a whip). To be successful you can either play well or take more pain than your opponent ;-) This led us to the discussion of how far one should go in designing novel user experiences.

Taking about tactile output Steffi mentioned the project VRIB (for more info see VRIB at Fraunhofer IMK or at Univeristy of Ilmenau) that was done from 2000 to 2004 on novel interaction devices and metaphors. This includes interesting issues that may be also relevant for our work on tactile output on mobile devices.

Wearable Activity Recognition – Talk by Kristof van Laerhoven

How many sensors do we need? That was one point in the discussion after Kristof’s talk. His approach, in contrast to many others, is to use a large number of sensors for activity recognition. This offers more freedom with regard to placement of sensors, variations of sensors, and also provides redundancy but makes the overall system more complex. His argument is that in the long term (when sensors will be an integral part of garments) the multi-sensor approach is superior – let’s wait some 10 years and then discuss it again ;-)

From a scientific perspective and in particular for machine learning (where he sees the greatest challenge) the larger number of distributed sensors is more interesting. I find his porcupine2 (http://porcupine2.sourceforge.net/) stuff quite interesting.

Kristof’s web page: www.mis.informatik.tu-darmstadt.de/People/kristof