Written by: Ben Kay - Web and Applications Developer, 9xb Last week I presented a paper at the 7th International Conference on Open Repositories, at the University of Edinburgh. Open Repositories is a gathering of people involved in the development, management, and application of a digital repository - a means of managing, storing and providing access to digital content. The paper reported on a project at the University of Hull on which I was Principal Developer, which formed the basis of my MSc by Research. The project was funded by the JISC and was lead by Dr. John Whelan, with Dr. Tanko Ishaya, Chris Awre and Professor Martin Goodman as collaborators.
Our aim was to produce an open source, multi-format publishing platform that could be adopted and branded by the Higher Education sector to increase campus-based publishing of both creative and scientific content. The project investigated how a usable Web interface could bring together the three phases of creation, archiving and discovery of this content into one multi-format environment. The result is a platform that provides automatic conversion from submission file format (DOCX) to distribution formats (EPUB, Kindle, and PDF), provides bespoke workflow models, Print-on-Demand, and integrates with the University's digital repository, Hydra. The code and technical documentation for the project can be found on GitHub. The paper described a technical component developed through the project, and follows on from a previous conference paper at the 9th International Conference on Scalable Vector Graphics at Microsoft’s R&D headquarters in Boston, MA in October 2011. The technical component presented at Open Repositories was CAPRI (Campus-based Publishing Repository Integrator), a software that enables the connection between the publishing platform and the University of Hull’s digital repository. The University’s repository uses the Fedora Commons software to store digital materials, including images, project reports, undergraduate dissertations, teaching materials, datasets, collections of past exam papers and doctoral theses. In technical terms, the component meets the project’s requirements through enabling instantiation of a PHP object that has corresponding properties, and methods to manipulate these, that constitutes an abstraction of a Fedora digital object. Through the proxy design pattern, an underlying PHP REST class is used to communicate with the repository’s REST API. This enables the accessing, creating, editing and deleting of digital objects in the repository through the platform. The paper then describes the evaluation of the CAPRI tool, including how authors felt about earlier drafts of work being publicly discoverable through the repository interface. The paper was presented in the Fedora User Group sessions of the conference, alongside other institutions including the University of Notre Dame and the University of Prince Edward Island. One of the main benefit of conferences is discovering and learning from other projects, and sharing results of your own, and through this collectively push forward the development of conference theme. To this end, the presentation was well-received, and was referenced in a series of tweets and resulting blog posts. Our hope is that the recommendations offered in the presentation will inform others looking to harness the flexibility of the Fedora REST API to control digital objects in a repository, and provide information for those who intend to publicly store versioned publishing artefacts in a university repository. The source code for the CAPRI is now available on Github. You can see the slides from the presentation for yourself here: Image credit: kaysgeog