Research Interests...

The 4 main areas of my current research are:-

I believe that Context-awareness is likely to play a crucial role in enabling mobile devices to provide genuinely useful functionality and offer more than simply an extension to Microsoft Outlook. Increasingly, mobile devices are likely to have the ability to sense context, either directly (e.g. GPS for location) or via wireless communications. The three main categories of context that exist are:-

  1. Device context, e.g. available batter power, current network QoS etc.
  2. Environmental context, e.g. weather, time, traffic hold-ups etc.
  3. Personal context, e.g. the users preferences (such as interests) and the user's knowledge/experience.
Uses for such context, include:
  1. Enabling applications (and the behaviour of users) to adapt to the dev ice's context, e.g. managing the tradeoff between processing power, battery life and limited/fluctuating network QoS when streaming video.
  2. Acting as a trigger to cause the presentation of relevant and timely information .
  3. Tailoring the presentation of information based on environmental and personal context.
  4. Reducing the demands on the user for performing/specifying certain tasks.

 The development of mobile context-aware systems raises a number of interesting issues including, usability, scalability and the role of connectivity.

Much of my practical work on mobile context-aware systems has been based on the GUIDE project. This project enabled us to investigate some of the issues surrounding the evaluation of mobile context-aware systems. From our work on GUIDE we learnt a number of valuable lessons for developing mobile context-aware interactive applications.

Platform support for Mobile Context-Aware Systems

Platforms need to be developed which can provide the context required by future context-aware applications. Applications should be able to search 'context servers' for certain 'interesting' types of context. Applications could then choose to receive context from the most appropriate context-server, i.e. the cheapest or the one that can provide context information in the most timely manner.

The following paper describes our initial thoughts on developing such a platform: postscript file or zipped postscript.

The Implications of Mobility on Collaborative Groupware Systems

Collaborative groupware can be affected by mobility for a number of reasons, such as: group members can suffer network disconnection group members can have slow links to other group members. In general mobility implies a number of constraints such as the potential for rapid and drastic changes in the quality of the underlying communications channel. Consequently, there is a strong argument for enabling applications to provide awareness or feedback through the user interface in order that collaborating  users can gain some knowledge of the effect that the constraints imposed by an heterogeneous mobile communications environment might have on the group’s collaboration.

User Interface Design Issues for Mobile (Adaptive) Interactive Application

It is well understood principle for the design of interactive systems that the behavior of a system show appear as predictable to the user. This principle has some interesting implications when considered in the light of application adaptation and context-awareness. For example, how predictable is a video conferencing application when the resolution of the video image changes (for no apparent reason). This is a problem for mobile interactive systems in general because adaptive context-aware systems because system behaviour, such as application functionality and the presentation quality of multimedia information, reacts to changes in context, e.g. the quality of the underlying communications channel. One approach for reducing this problem is provide users with an appropriate, level of 'awareness' for appreciating the current behaviour of the system. To achieve this, metaphors need to be designed that encourage the user to form simple (yet sufficient) mental models for understanding the potential for changes in system behaviour. For example, the 'bars of connectivity' metaphor, made familiar by the mobile phone industry, has been designed to enable users to appreciate the affect of the communications link on a mobile phone's functionality.

Back to my home page...