Why context technology? A key component in AmbieSense is the context technology that has the ability to capture and store information about individual user contexts. It is used to improve the information retrieval and filtering of information for mobile users. Context information in AmbieSense describes aspects of an individual user's situation. It can in a wider setting also describe aspects of any actor's situation. Examples of context structures can be: user contexts, document contexts, environment contexts, and cases in case-based reasoning. People’s needs for information and interaction, whilst being mobile, can be summarised as: Individuals have their own needs for information (i.e. personalisation) Personal needs and interests change from situation to situation People with mobiles need wireless access to information Timely delivery of specific information to specific situations is crucial when mobile Vicinity-oriented interaction with info mobile information services are even more important when mobile People travel between real-world situations – so do their needs AmbieSense addresses the information needs of mobile users by trying to improve the individual information retrieval and filtering process. The way this is sought solved, is by developing and using a context technology that can be reused and customised across applications. However, note that each application that tries to use the context technology as a value-adding component, should define its own context template that captures only the most important attributes of the context. Otherwise, less relevant information may be retrieved, and the retrieval process itself might even be slowed down compared to traditional methods. How to make context technology work? The safest way is to involve the user to find out about their information needs, and which context information that really matters for them in your application. Your result may be a context-aware system that outperforms traditional systems in terms of relevance and usefulness of the retrieved information, and sometimes a faster retrieval algorithm. The context technology in AmbieSense comes in two independent mechanisms: Context middleware – can be deployed on any computer or mobile device Context tags – can store context information and communicate with mobile devices Scenarios for our context technology Individual user contexts and digital content related to these, is a key concept in AmbieSense. The context technology covers the following scenarios. Note that the scenarios the context technology: Scenario 1: Planning for future contexts To some extend, mobile users can be able to express future contexts. In user contexts in AmbieSense are viewed as personal activities. They can be part of the past, it can be the current context, or it can be part of the future. Typically, appointments, and planned activities to occur in the future, contain context information. Information about future contexts can to some extent be derived from the user's interaction with the system. Examples of such context information can be: future travels, and long-term user interests. Long-term user interests can be viewed as a persistent query for digital content. It can also be viewed as an information filter that may increase the quality of the personal information experience over time. Scenario 2: Interacting with the current context The current situation that any mobile user is located in is potentially the best and most accurate source of context information. The context information that can be derived from your current situation will likely be the most accurate source to benefit from for any context-aware application. This is because context information can be specified directly by the end-user, indirectly derived from the user-system interaction, or automatically detected by system components and digital sensors. Scenario 3: Retrieving and using the context history The value of context technology will perhaps be best demonstrated when the user wants to retrieve past digital content - like images, films, music, sports, news, and so on. The user will then remember aspects of the past situation, and try to express this "incomplete" fragment of the past context to the system. This may be in the form of text queries to the system. If the image that the user is looking for was taken in Alaska October 2002, and there were three salmons in it. The user will try to articulate this to the system in one way or another. The context technology can help the user in retrieving the right image(s). Hence, the use of context technology in information retrieval or filtering applications has a clear potential. Scenario 4: Content service provision This scenario describes the location and time based delivery/distribution of relevant information to specific situations and relevant users taken into account the different model of content ownership, access rights, cost models, etc. of different content providers and content service providers. Context technology can be used to enable effective publishing processes. The result may be reduced costs for the content service provider. Being mobile doesn’t have a single characteristic. The length of a mobile phase varies just as much as the intentions with the moving. Based on the reasons for the moving, the mobile user wants a different set of information from a source. At travelling points like airports the AmbieSense system architecture will allow for users to know their exit gates, tax free shops, toilets and coffee house just by a quick glance at the map on their hand-held computers. The use of context tags located on different sites will allow for the hotspots in the area to communicate its special offers or flight to the users that want it.