Resource List Toolkit
A resource list is a generic collection of data such as bibliographic citation information, website links, course reserves, or reading materials. A common characteristic of resource lists in many different scenarios is their lack of reusability. They are often created and stored in proprietary formats. It is also problematic to export lists that are intrinsically stored in some system repositories. Hence, reusing resource lists often requires mediation and laborious manual tasks such as reproducing the same lists for other system use scenarios.
Background
A Resource List management application was a logical development for Edinburgh University Library as previous development work had been done in the areas of distributed searching (d+) and resource list creation (devil). A Resource List Service is also one of the building blocks of the JISC e-Learning framework.
Aims and Objectives
The original aims as outlined in the project plan are repeated below:
“The aim is to develop a software development toolkit (SDK) to mediate the reuse of resource lists created and stored in distributed and heterogeneous systems ranging from e-learning tools, VLEs, portals, dedicated resource list systems and repositories. Our proposed SDK consists of components: a resource list services manager and a demonstrator application. The services manager is an application framework to facilitate a range of web services for managing pre-existed resource lists among heterogeneous systems. The web services are based on the established Create/Read/Update/Delete (CRUD) model and the IMS Resource List Interoperability Specification. The demonstrator is a web-based application that allows potential implementers of the SDK to easily create sample resource lists from federated searching of distributed resources and use them to test the services manager. The demonstrator is also a services consumer prototype which sends out the CRUD services requests to, and receives results from the services manager.”
The general aims of the project did not change during the project lifecycle but there were some changes in specific areas:
- Time constraints meant that the web service is not a complete implementation of the IMS spec. The ‘group’ function was not developed, nor is the complete information model supported
- The IMS spec was revised to include a ‘query’ function considered necessary for any implementation
- It was not possible to find a ready made repository, e.g. Dspace, which would be a good match for the information model described in the IMS spec. A bespoke MySQL database was used instead
- The demonstrator application is more limited in functionality than at first envisaged. We hope to continue working on the development of the demonstrator application to more fully support the range of functionality available in the original Devil application
Methodology
The project was planned as a web service implementation of the IMS RLI specification, in line with JISC web services based model for software development. The specification includes an information model and WSDL which dictated that the development would be a SOAP based web service. Apache Axis was chosen as the tool to facilitate that development.