We have been developing a “Data Webs” for research image data, which focus on lightweight web technologies to link published data and documents. In contrast with approaches that attempt to import information into a prescribed application environment, our approach uses the World Wide Web as a platform, and to direct users to the original publication source for all data. Institutional repositories are a natural home for such image data: we aim to link repository image collections to a wider landscape of data on the web, through the application of Semantic Web and other web standards and tools.
Aims and objectives
We aim to understand how existing Web and Semantic Web tools can be used to improve access to and usability of research images published to the web in institutional repositories, and to link such publications with other published sources such as central bioinformatics databases. In particular, we are identifying tools, standards and other work that can be used to make research image collections more web-accessible via institutional repositories, and to scope of the role that such repositories may play.
Project methodology
Our work focuses on a survey of software tools, web standards, metadata standards and other related research. Our activities include meetings, repository content examination, web search and some hands-on software evaluation. Also, studies of image collections at Cambridge, Southampton, ImperialCollegeand OxfordUniversities, and discussions with their institutional repository team. Information gathered is being recorded as we go along in the project web site. David Shotton, the PI, has led coordination of external meetings, Graham Klyne, the project manager, has coordinated recording of survey work in the web site, and Jun Zhao has led the hands-on software evaluations.
Anticipated outputs and outcomes
The primary deliverable is a report that summarizes our survey findings, recommendations to the JISC concerning future research activities, and an outline software design for creating an data web for research image data. A secondary deliverable is the ImageWeb wiki web site, a live resource that will continue to accumulate information about related tools, standards and research. The web site also contains records of our project meetings and other dissemination activities.
Technology / Standards used
We are advocating the use of common Web standards, notably:
Web standards: HTTP, XHTML, URIs
Semantic Web standards: RDF (model), SPARQL, OWL
The project has used Semantic Media Wiki for its primary web site, supported by Drupal for resource-sharing and some weblog-style commentaries. We are using ePrints to create a prototype for web-publication of research group images. Currently, Excel spreadsheets are a primary mechanism for metadata acquisition.
Lead Institution
Oxford
University
: http://www.ox.ac.uk/
Zoology department: http://www.zoo.ox.ac.uk/
Project partners
UKOLN, University of Bath: http://www.ukoln.ac.uk/