The Monday workshop sessions of IFIPTM 2008 were a combination of the second workshop on Context-awareness and trust (CAT) and first workshop on Web 2.0 trust (W2Trust). See the W2Trust website for the full list of papers. In this post, we summarize what we saw.
Jacob Bardram opened the session with an invited talk on context-aware user authentication, bringing up interesting problems, pragmatic workarounds and a solution for a more usable and secure login system for nomadic nurses working in hospitals. The lesson: if security gets in your way, you bypass it, so best design some usable security.
Marcin Sydow presented a proposal to use user similarity information to deduce trust in recommender systems and the like, applying a data mining background in Towards using contextual information to learn trust metric in social networks. The thought of combining user similarity and trust information came up on Wednesday in Georgios Pitsilis’ presentation as well.
Hasan Akram provided some requirement analysis for identity management in ambient environments (with Mario Hoffmann), from the point of view of HYDRA, an EU Integrated Project to produce middleware for ambient environments in e.g. healthcare and agriculture.
Erik Buchmann presented a survey of the user and provider perspective on why privacy-enhancement mechanisms fail (with Thorben Burghardt and Klemes Böhm). We have regulation, self-regulation and privacy-enhancement technology by the boatload, but in practice privacy protection is (by bad design and by nature) too difficult for the user, and in the end people will rather just not think about it. Fun examples: thereturnexchange.com, spock.com, yasni.de, dstudivz.net… and of course Facebook, everyone’s favourite
Natasha Dwyer applied game as a cultural probe to understand the grounds to trust. 10 participants played an investment game, making decisions to risk different-sized investments in the hands of fictional traders who were described with various types of information.
Christian Damsgaard Jensen described a depth-camera based tracking system to provide persistent authentication in smart environments like a university building (with Mads Syska Hansen and Martin Kirschmeyer). Authenticate yourself once, and your tracked blob gets doors unlocked for you as long as you don’t run in the hallways.
Jau-Yuan Chen presented a feature-based scheme for detecting phishing pages (with Kuan-Ta Chen). Instead of the usual text-based analysis, the presented method is based on snapshot images of the pages.
Anders Kofod-Petersen (with Klaeboe, Jervidalo, Aaltvedt, Romnes and Nyhus) described experiences of implementing location-aware friend systems that balance privacy with information symmetry: if you want to know where I am, tell me where you are too. The students who volunteered to participate in the experiments were actually not really interested in upkeeping their privacy – if it’s cool, it doesn’t matter?
Mike Bergmann mapped privacy goals to current web reality and asked whether web 2.0 means privacy 0.2. We seem to be deep within an early alpha phase.
Any heartburn caused by oversimplification can be forwarded to the MobBlog PR centre using the comment form below. We look forward to your commentary.