Deliverables
The use of electronic services is spreading more and more to an increasingly broader group of users, and there is a growing need for support for continuous interaction with multiple services, via different types of devices, and from all sorts of places and locations. Further more, it is desirable that this is done in a way that assures the user control over personal information that services gather and maintain. The user should also be able to control what services do and whether or not, and how, they collaborate with each other.
sView is an electronic service environment with heavy focus on the individual user. It enables a user to collect, store, and run electronic services, locally, or in a distributed fashion. The whole purpose of sView is to serve as a common area for the services to cooperate amongst themselves and to provide a unified access method to the services for the user. To a developer of services, the sView platform is characterized by openness, modularity, security, and a high degree of flexibility. To a user of the platform, it is accessible in many ways and available continuously.
With the sView platform in the FEEL project, the project partners developing the base software platform were able to deploy the FEEL intrusiveness management system in a prototype and run two user studies on the complete system. Using sView it has been easy to construct components for various technologies that represent a common type, for example, user interface components all follow a common specification for how they should function within sView and how they should supply their service to other services in the briefcase. This means that it is easy to introduce a new type of interface capability by constructing a component (service) for the new type of functionality that adheres to the common formats.
Thanks to the modular design of sView it was easy to introduce several different communication channels that were to be intrusiveness controlled.
The demands on sView represent current research topics such as privacy in the context of electronic service usage, service collaboration, and ubiquitous user interface design. The sView system has been designed as a solution to some of these research topics, and to cater for further research on others. The system assumes a client/server model, but instead of having a uniform client without service specific functionality for access to all servers (as in the case with the World Wide Web), access to the servers is channeled through a virtual service briefcase. The briefcase in turn, supports access from many different types of devices and user interfaces. It is also private to an individual user, and it can store service components containing both service logic and data from service providers. This allows the service provider to split the services in two parts. One part provides commonly used functionality and user specific data that executes and is stored within the virtual briefcase. The other part provides network-based functionality and data that is common between all users. Finally, the service briefcase is mobile and it can follow its user from host to host. This allows local and partially network independent access to the service components in the briefcase.
At a high level, the sView system consists of two parts. The core sView specification constitutes an API that defines the basics of service components and personal service environment handling. It specifies service components, a runtime environment for service components, a data structure for persistent and mobile images of service environments (service briefcases), and a server for handling service briefcases. The specification has been implemented as a set of Java packages (which contains mostly interfaces and a few classes). Implementing these APIs and adhering to the design guidelines that accompany the APIs, assures compatibility between sView services and service infrastructure of different origin. The sView reference implementation provides developers with a development and runtime environment for service components as well as a sample implementation of a sView server.
This result is a service oriented intrusiveness management system for telecom and internet based communications. It constitutes a viable technical solution for intrusiveness management in technology saturated communication scenarios. An overall sub-objective to this is to create the solution in a ubiquitous or disappearing computer style. These objectives have been reached, as can be seen by the following text that describes the technical solution as a whole as well as in detail for the component parts. The fulfilment of the objectives is also evident in the prototype implementation that the project has produced. The resulting system, named FEEL Intrusiveness Management system (FEELIM), has progressed through iterative development from a set of individual and basic components. FEELIM presupposes the following hardware scenario. The basic hardware components are laptop computers, mobile phones, public display devices with controlling stationary PCs, and a central server.
The public display devices will consist of smart-boards or projection systems connected to stationary controlling computers. Any personal interaction between the user and the end user services takes place locally on a laptop with wireless networking capabilities.
Each user can use his or her own mobile phone with a special SIM card. FEELIM components are as follows:
- The sView system
- A dedicated user interface
- The Sentinel intrusiveness controller
- A number of end user services
- Several public services
- A media routing component
- Location context information service component
- A logging system and
- An agent-based negotiation mechanism to handle more complex scheduling and routing problem.
The sView system is the consolidating environment for users activities involving the use of electronic services. Each user is given a personal sView system including access to a central sView Enterprise Server. The Sentinel intrusiveness controller is used to establish an appropriate level of intrusiveness for a currently ongoing ad-hoc meeting, and to control other (end user) services in sView regarding their allowed methods of notification and other types of interaction with the user, in terms of intrusiveness. The Sentinel service is loaded into each users sView briefcase. A number of end user services are provided to users as part of FEELIM. The purpose of these services is to provide certain service specific capabilities for the user while engaging in interaction with the sView system.
The services include an email service, an ICQ client, a SMS service and and a telephony service for mobile phones. In addition to the end user services, FEELIM also supplies users with access to two public services which may be used directly by the services in users briefcases for displaying messages and notifications: the sBanner and the iClock. The sBanner displays a scrolling banner of messages across any screen space on public or private displays. The iClock displays a clock on any display device, private or public (it is called sviClock in the private version). On top of this clock display, any text message may be be displayed for a brief period of time. The sBanner and iClock back-end services are installed on each controlling computer. Each users service briefcase is complemented with a location sensing service, which may be used by any other service in sView to ascertain the users current location. Work has also concurrently been done to devise an argumentation-based algorithm to manage the notifications received through sView services.
This algorithm is integrated in the FEELIM system. Each end user service as well as the Sentinel and other supporting services in sView is equipped with logging functionality. The method of logging is based on sending communications in response to important events in the services to a central logging server.
The FEEL project has developed new algorithms for negotiation between software agents, based on the notion of argumentation. These are advanced techniques, which have involved basic research and construction of laboratory prototypes. The algorithms have been tested in simulation and also have been integrated with the FEEL software. The algorithms have been published as academic papers.
Within the FEEL project, these algorithms were designed to address the particular requirements of managing intrusiveness in the pervasive computing context. In this context, their role is to conduct negotiation in order to regulate information flow.
The algorithms have much wider applicability and will be relevant in a variety of distributed computing situations. This includes other information systems applications involving users interacting with realtime information flows, as well as online negotiation within the e-business context.
These algorithms can be used as the basis for applied research in a variety of contexts and also future basic research.
Searching for OpenAIRE data...
There was an error trying to search data from OpenAIRE
No results available