Workpackage description: 12.1: Open demonstrators

Workpackage number: 12.1

Start date or starting event: Month 1

Lead Partner: HP (4)

Participant short name: ILRT W3C-ERCIM CCLRC HP STILO
Participant number: 1 2 3 4 5
Person-months per participant: 0 0 0 33 0

Total number of deliverables: 8


Description of work

Pilot and demonstrator projects are a critical cornerstone of research and development for the Semantic Web. In particular they play three key roles within this project.

Firstly, they act as realistic application-driven tests of the developed technology and toolkits. Existing Semantic Web toolkits and APIs have tended to evolve bottom up based on predictions of the functionality required; this workpackage tests these APIs against application-driven requirements. It will give critical guidance to the development of those interfaces and suggest new middle-tier components to assist in application development.

Secondly, in deploying any Semantic Web based application there will be a number of ancillary practical issues to be solved: database-layout performance tuning, metadata management techniques, appropriate use of provenance tracking and so forth. The demonstrator development of necessity will address some of these issues and lead to a body of "how to" knowledge that can speed up future application development via the educational outreach program.

Finally, the existence of functioning demonstrators offers developers strong indicators of the potential applications of the Semantic Web. Motivating examples are as important to widespread uptake as the technical know-how and available toolkits. The educational outreach programme is the appropriate mechanism for exploiting such demonstrators.

Type of demonstrator

At this stage we do not fix the precise demonstrators to be developed. As guidance we describe one broad application area we consider promising; however new possibilities emerge by the month and the final choice should be made as part of this workpackage. The critical nature of the Semantic Web is that it should be a web. It is the combination and integration of semantic data across different sources that is key to its power. The broad class of application we start from is one where a rich set of web-accessible information repositories exist (or are in development) and multiple community groups wish to subset, collate, annotate, and organize their own views on this data. Examples of this include:

Each of these applications has common technology requirements: multiple scalable metadata stores, each supporting multiple specialist user communities with their own ontologies, linked together into a common metadata web. These features are core to the evolution of the Semantic Web and key elements of likely demonstrator projects to be selected