I just came back from 11th International Conference on Human-Computer Interaction with Mobile Devices and Services (shorter version: Mobile HCI) in Bonn, Germany, below is a short summary and my own thoughts about selected presentations and conference itself.
1. Tutorials
The first session was about Mobile Search, Matt Jones explained the importance of mobile search. Matt highlighted mobiles search issues such as data entry methods, using search in different context, differences between search on mobile and desktops (Google and Yahoo study). Take home message from tutorial:
- Data input is not everything - presenting the results is also a very important issue.
- Clustering search results is a good thing.
In the second tutorial session Luca Chittaro introduced us with the domain of information visualization and visual interfaces for mobile devices. Luca presented number of projects he worked on, highlighting the important issues of developing mobile visualizations. He also suggested few important elements of design methodology for mobile visualizations:
- list limitations of mobile device that make information visualization different
- mobile interface should require very little training - take it into account during design
- Design of mobile visualization must be a disciplined process, which contains the following elements:
-- a) mapping (How should we visually encode information?)
--- 2 principles of mapping:
---- A precise mapping between data objects (+ their relations) and visual objects (+ their relations) must be explicitly defined and consistently applied throughout the application.
---- Conceptually important aspects must be perceptively important, and conceptually unimportant aspects must be perceptively unimportant.
-- b) selection (among the data available, what is relevant to the considered task?)
-- c) presentation (how should we lay out the visualization on the available display space?)
-- d) interactivity (What tools sould we provide to explore and rearrange the visualization?)
-- e) human factors (Are we taking into account human perceptual and cognitive abilities?)
-- f) evaluation (How should we test that the visualization is really effective with users on the considered task?)
In the third tutorial Chris Kray introduced the current state of research in Mobile Guides domain. General talk about designing, developing and evaluating mobile guides. Really interesting however not really relevant to my work.
The following session by Marc Langheinrich about Location Privacy was very interesting, and cemented my belief that I've made the right literature review in this domain (although Marc presented some new work that I was not aware of). Presentation rather raised more questions than answers about location privacy, which makes people listening and thinking. Such as can we have both save, secure and privacy friendly AND usable, useful and used phone? or What's privacy? Secrecy!, confidentiality!
Marc also explained the positives and negatives of sharing location.
- Why we share location?
-- by product of positioning technology
-- required to use the service (local recommendations, payment for toll road)
-- social benefits (let friends, family know where am I)
- Why not to share your location?
-- Location implications:
---- where I live/work
---- Who I am
---- Hobbies/interests/memberships
---- People I meet
---- profiling (zip code, implies income, ethnicity, family size)
Fifth tutorial session was about Mobile Interaction with the Real World and was lead by Enrico Rukzio. He explained differences between HCI, Mobile HCI and machine-machine interaction. Enrico also introduced the main interaction types, such as touching, pointing, scanning, user-mediated object interaction and indirect remote controls.
The last tutorial session by Paul Holleis was about Modelling and Developing Mobile Applications. Paul presented basic methods for user modelling:
- cognitive modelling
- GOMS
- Keystroke-Level Model
I liked the example of ATM, showing user's goal vs process design. (Why we need to enter pin, and then take a debit card before we can withdraw money?). Understanding user is an important element of product development.
2. Conference - only selected presentations
Conference started with a keynote by Jun Rekimoto, title: Large Scale Sensing and Integration of Real and Virtual.
Jun presented a very interesting game called "Eye of judgement" - just google for that to see :)
In the next part of the talk he shared with us the lessons he learnt from the development of Place engine - wifi positioning system in Japan. (Something similar to Navizon - people share their access points coordinates which enables wifi-based positioning) Interesting finding: Imperial Park in Tokyo has no wifi :). Really nice feature of the system is that it Place Engine enables not only lat and lon but also altitude (floor number), which is important in a city like Tokyo. Jun presented results of gps vs. wifi postitioning comparison - almost the same, with some exceptions, where gps works worse. Jun suggested a new research topic which is big challenge for them: constructions and maintenance of the big location data.
Another interesting project presented by Jun is Pet Lifelogging (Yonezawa, ACE09), including cat face recognition technology - and analysis of social interaction between cats [something that Clara might be interested in :)]
Next talk was by Yusuke Fukazawa about Automatic Mobile Menu Customization Based on User Operation History. He presented a paper about the design and evaluation of adaptive menu on Android platform:
- platform: Android phone
- participants: 43
- age: 18-25
- duration: 1 week
Take home message: How do you perceive the usability of the ui customization? Adaptive system itself doesn't guarantee better usability, how to assure that this goal is met?
During the same session Julia Schreiber from Volksvagen in her talk Bridging the gap between useful and aesthetic maps in car navigation systems was talking about the usability of maps. The conclusion was that maps adaptation is necessary and it can improve the usability of maps applications.
"Usability" was the most used word of the first conference day :)
In the keynote 2 Peter Moeckel from Deutche Telekom gave us a brief overview of their Research Lab (nice advertisement movie :) ) and user driven innovations. He presented a methodology of designing a new product:
- a) Pipeline - Trends analysis
- b) Exploration - Netnography
- c) Idea generation - focus groups
- d) Selectory execution - evaluation
- e) Commercial launch
Richard Harper - Glancephone. RH presented a system that enables people to glance at others through the front camera on their mobiles. Glancephone was originally invented to support callers in making their "call decisions" - call or do not disturb. Etnographic study showed that people would rather prefer to just use the phone to enable other looking at them.
In the next talk Anupriya Ankolekar showed the Friendlee application. Friendlee is a contact-based social network application, which sorts contacts list based on the calls number and frequency. It changes the contact application into context-awareness friends-list known from social networking sites. Some basic privacy management built in (ex: show my location to College group only). Very interesting project, which raised a lot of questions. A lot of both positive and negative voices heard.
Best paper award winners:
- Sweep-shake: finding digital resources in physical environments - system supporting haptic feedback in mobile search. Shake device vibrates when the user points it toward a location with geo-tagged content. If the content is available then device vibrates, user can refine search by pointing device to bottom-left (video), b-r (audio), t-l (text), t-r (images). Shortcoming of the system is that it does not tell what are the search results, it only can tell whether there are some resources a vailable or not. Really like the idea :)
- PhotoMap: using spontaneously taken images of public maps for pedestrian navigation tasks on mobile devices - the core idea behind PhotoMap is simple: combine real maps (like google maps) with "you are here" maps known from parks, zoos etc. Johannes Schoening presented different implementations of PhotoMap (nokia phone, iphone and android) - nice idea, very good implementation, which has a potential to be a good commercial product.
My overall opinion about the conference is very positive, it was a great organized event, with lot of interesting talks and keynotes. Privacy, which is my main interest, does not seem to be a very popular topic in the community, only a few presenters mentioned that briefly. There was no paper (only one poster) talking directly about privacy, if someone mentioned that it was rather a side effect they had to challenge during the main task.
interesting: I’d say that >50% of all research is made using iPhone platform, the rest is shared between Nokia phones (especially people from Finland :) with some iPhone exceptions ;] ) and Android (rest of the World). I will try to cite on guy that was advertising his poster during madness session: I’ve made something that is boring, crappy etc but I’ve made it on iPhone and now is cool. Come and see me near stand X – I think that this example perfectly illustrates the current state of platform-interest in mobile research, no matter what you do, do it on iPhone and it will make it more interesting.