Adaptability and Interfaces: Key to Efficient Pervasive Computing

As mobile and embedded computing devices become more pervasive, the nature of interaction between users and computers is evolving. Computing applications need to become increasingly autonomous and invisible, by placing greater reliance on knowledge of context in which they are running and reducing interactions with users. Moreover, due to the current popular emphasis onànytime, anywhere' computing, applications must cope with highly dynamic environments in which resources, such as network connectivity, bandwidth, security and software services, vary over time and therefore applications' behaviour and functionality have to adapt (adjust) to currently available resources and constraints. One example of varying resources comes from a requirement that users can use multiple and potentially heterogeneous computing devices and switch between those (eg, a workstation to a mobile laptop, or PDA (Personal Digital Assistant), then to a mobile phone) while still participating in distributed computational activities. To enable such change of devices, applications have to adapt to a variety of displays, various CPU capabilities, presence or absence of a hard disk and varying bandwidth of communication channels. Pervasive computing is believed to be a computing paradigm incorporated in a variety of devices (computers, cars, entertainment units, appliances, etc.), which can carry out computational activities in a relatively non-intrusive manner and can impact and support many aspects of work and everyday life. Pervasive systems need to be aware of their environment and associated resources (i.e. aware of their context), able to detect changes in the environment (context changes) and can adjust/adapt their functionality and behaviour to the changes. Pervasive computing systems represent a natural extension and evolution of mobile/ubiquitous/nomadic computing systems, which first emerged in the early 90s. While recognizing diversity and enormity of pervasive computing related issues, I reckon that ultimate success of this paradigm depends on useability of pervasive applications. I would identify the following important issues that need to be addressed by mobile and pervasive computing community. • Adaptability • Flexible customisable interfaces, both present in the device and around it • Context management, including location • Accuracy & credibility vs. disconnection-transparent continuous execution, mobile transaction accuracy thresholds • Direct and indirect methods of location (and other context attributes) verification The rest of this position paper is structured accordingly. Adaptability. System stability could be viewed from the perspectives of functional and behavioural stability. Functional stability refers to systems adherence to design specifications and its ability to achieve the expected outcomes from individual components …