All the ideas presented in the context-awareness related research section have illustrated the virtues of extending the mobile computing platform to incorporate a knowledge of the user's environment.

So why is such a useful technique not more widespread? I suggest that the primary reason for this is the difficulty and lack of supporting infrastructure in capturing and processing context data, so hampering the development of context-aware systems. Those applications which do provide support for context-awareness generally have only a limited number of context types. A more diverse range of contexts could further expand the functional possibilities of context-aware systems.

The current lack of support for context-awareness can be likened to developing applications where the whole user-interface has to be written each time. Of course, in reality, when developing applications the designer/programmer can use a windowing architecture in which the common elements of user interfaces (windows, menus, dialogs, etc.) have already been implemented. Its just a job of arranging and perhaps enhancing them to suit the particular application's needs.

The stick-e note research proposes an architecture for context-aware computing to provide a similar infrastructure for developers of context aware applications and interfaces. Such an architecture will make it possible to incorporate context-awareness not only in new specially designed applications, but also in the design of existing user interfaces where the overhead of designing and constructing context aware elements, although helpful to the user, had too great a development overhead in the past.

The role of such an architecture is to provide the framework that can be used by the application to adapt its user-interface or perform some action based on the current context. From the selection of context-aware systems and principles described in the context-awareness section, it can be seen that the potential applications cover a very broad variety of uses, e.g. the notional attachment of electronic information to physical objects or the ordering of a selection list in a GUI based on proximity. The very application-specific nature of some potential uses of context-awareness make it both difficult and undesirable to create a common display system that all of these technologies use. It is more relevant to focus on the underlying metaphor and infrastructure that supports the context-aware application. This is what the stick-e note architecture addresses.

Any comments, criticisms or suggestions on this research are very welcome, please email them to us.

	MCFE Home
	Stick-e Notes		Architecture