The project will adopt an Enterprise Architecture (EA) approach. The models will include different viewpoints; i.e. the business processes, the information models (and the interaction with the institution and external resources), and the organisation model. The project team has been a member of the JISC EA practice group and our EA approach will be influenced by the experiences of other universities in implementing EA programmes. The main tool that will be used for modelling is Archi[1].
XCRI-CAP implementation: The implementation of the feed and the web services will follow best practices highlighted in both the XCR-CAP knowledge base and XCRI-CAP wiki. From our preliminary investigation, we anticipate that we will provide two feeds; one for part-time and one for full-time, and similarly the web services will take a single parameter specifying full-time or part-time.
Figure 1 LUCI Architecture
Linked data implementation: Our starting point for the implementation is the Tabloid tool kit produced by the LUCERO JISC project. The tool kit will be adapted to our institution and to the course information advertising process. Figure 1 illustrates some of the adaptations that we will be carrying out.
Course explorer application: The development of the application will follow a user-centred design. Requirements gathering will be conducted as a series of user-centred participative workshops with key stakeholders (part-time students, academic tutors, admission officer and web team members). Workshops will focus on determining possible requirements to exposing the benefit of publishing the course information in linked data. This will ensure that users are fully participatory in the development of the interface. The interface will be created using a lightweight agile approach extracted from the SCRUM model of development. After the requirement elicitation activities, the project will then deliver each iteration of the interface implementation in a series of sprints, with users evaluating the outputs as part of the sprint review meetings.
Processes enhancement: The processes will be modelled in BPMN with the objective to provide an executable version of the processes. XCRI-CAP schema will be used as the information model for the web form that will be passed between the activities in the processes. The team has experience in implementing executable business processes. The tool that we will be using is the open source business process management INTALIO[2]. The implementation of the user activity forms will take the same approach as for the course explorer application; i.e. a user-centred design.
The overall approach for managing project activities will follow a structured set of work packages.
Work package 1: Project management
This will involve management and administration project activities throughout the life of the project.
Work package 2: Project web site/ blog/ wiki
This will use the technical platform used in Cova, Remora, and MCMS5. The work package will involve the constant updating of the web site and the setting up of the wiki. Work done in this package will contribute to the final report.
Work package 3: Enterprise Model Design
This involves refining the institution’s As-is model developed in phase 1 based on further investigation. The To-be EA model will also be built including course data transformation model and linked data schema.
Work package 4: Implementation of XCRI-CAP feed and Web services
The work involves implementation of the course data feed. Existing course information held in the “CID” database will be published into to XCRI-CAP feed and web services.
Work package 5: Implementation of Linked data
Semantic based course data will be constructed in this package. Distributed course related data will be linked in RDF format via knowledge engineering process.
Work package 6: Enhancement of the course information processes
In this package, we will build the user interface for the course data registry. The interface will be integrated with the existing UWL course database and new course data generators through a workflow system.
Work package 7: Implementation of course data discovery
A tool that will allow the search of course linked data will be developed. The discovery will be more human-like rather than keyword matching by interpreting course data triples.
Work package 8: Evaluation & Dissemination
Evaluation and dissemination is integral across the project and is core to our application development approach. We intend to gather a range of experimental evidence to support the project itself and the deliverables it produces. Also informing our evaluation will be a set of research questions such as:
“ What are the business processes and systems needed to expose course information and its related content as a linked data?”
“How can we integrate linked data in the institution culture in a sustainable way?”
Work package 9: Project Report
This will involve producing JISC required reports including an intermediate and final report.