Monday 19 May 2008

Workshop on Smart Homes at Pervasive 2008

Today we had our Pervasive at home workshop – as part of Pervasive 2008 in Sydney. We had 7 talks and a number of discussions on various topics related to smart homes. Issues ranged from long term experience with smart home deployments (Lasse Kaila et al.), development cycle (Aaron Quigley et al.), to end-user development (Joëlle Coutaz). For the full workshop proceedings see [1].

One trend that can be observed is that researchers move beyond the living lab. In the discussion it became apparent that living labs can start research efforts in this area and function as focus point for researchers with different interests (e.g. technology and user-centred). However it was largely agreed that this can only be a first step and that deployments in actual home settings are becoming more essential to make an impact.

On central problem in smart home research is to develop future devices and services – where prototyping is based on current technologies and where we extrapolate from currently observed user behavior. We had some discussion how this can be done most effectively and what value observational techniques add to technology research and vice versa.

We discussed potential options for future smart home deployments and I suggested creating a hotel where people can experience future living and agree at the same time to give away their data for research purpose. Knowing what theme-hotels are around this idea is not as strange as it sounds ;-) perhaps we have to talk to some companies and propose this idea…

More of the workshop discussion is captured at: http://pervasivehome.pbwiki.com/

There are two interesting references that came up in discussions that I like to share. First the smart home at Duke University (http://www.smarthome.duke.edu/), which is dorm that is a live-in laboratory at Duke University – and it seems it is more expensive that the regular dorm. The second is an ambient interactive device, Joelle Coutaz discussed in the context of her presentation on a new approach to end-user programming and end-user development. The Nabaztag (http://www.nabaztag.com/) is a networked user interface that includes input and output (e.g. text2speech, moveable ears and LEDs) which can be programmed. I would be curious how well it really works to get people more connected – which relates to some ideas of us on having an easy communication channels.

[1] A.J. Brush, Shwetak Patel, Brian Meyers, Albrecht Schmidt (editors). Proceedings of the 1st Workshop on “Pervasive Computing at Home” held at the 6th international Conference on Pervasive Computing, Sydney, May 19 2008. http://murx.medien.ifi.lmu.de/~albrecht/pdf/pervasive-at-home-ws-proceedings-2008.pdf